Re: TEI-L Digest - 3 May 2017 to 4 May 2017 (#2017-101)

--590c713e_4691fe5f_d8c
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

Thank you Matthew and =46ranz,
my mistake=21 I should have read the GL more carefully
Thanks a lot for your explanations.
Best,
Emmanuelle


De:=C2=A0MLH <matthewholford=40hotmail.com>
R=C3=A9pondre:=C2=A0MLH <matthewholford=40hotmail.com>
Date:=C2=A05 mai 2017 at 11:05:31
=C3=80:=C2=A0(Text Encoding Initiative) public discussion list TEI <tei-l=
=40listserv.brown.edu>, emmanuelle.morlock=40MOM.=46R <emmanuelle.morlock=
=40mom.fr>
Sujet:=C2=A0 Re: TEI-L Digest - 3 May 2017 to 4 May 2017 (=232017-101) =20

Dear Emmanuelle,

My understanding is that the two numbers in =40writtenLines (also=C2=A0=40=
ruledLines), is that the first number gives the minimum number of lines p=
er page or column in the codex as a whole, and the second gives the maxim=
um number of lines per page or column in the codex as a whole.

e.g.

columns=3D=222=22 writtenLines=3D=2220 30=22

would not mean that column a had 20 lines and column b 30 lines, but that=
 the codex as a whole had 2 columns per page and between 20 and 30 lines =
per column throughout.

similarly

columns=3D=221=22 writtenLines=3D=2220 30=22

would mean a codex in long lines with between 20 and 30 lines per page th=
roughout.

Best wishes,

Matthew



=46rom: TEI (Text Encoding Initiative) public discussion list <TEI-L=40LI=
STSERV.BROWN.EDU> on behalf of TEI-L automatic digest system <LISTSERV=40=
LISTSERV.BROWN.EDU>
Sent: 05 May 2017 04:00
To: TEI-L=40LISTSERV.BROWN.EDU
Subject: TEI-L Digest - 3 May 2017 to 4 May 2017 (=232017-101)
=C2=A0
There are 10 messages totaling 2938 lines in this issue.

Topics of the day:

=C2=A0 1. =22May contain: Empty element=22
=C2=A0 2. question about attribute =40writtenLines (metadata)
=C2=A0 3. =40xml:base with =40rendition (and maybe other pointers) (8)

----------------------------------------------------------------------

Date:=C2=A0=C2=A0=C2=A0 Thu, 4 May 2017 11:28:58 +0100
=46rom:=C2=A0=C2=A0=C2=A0 Lou Burnard <lou.burnard=40RETIRED.OX.AC.UK>
Subject: Re: =22May contain: Empty element=22

On 03/05/17 16:19, John P. McCaskey wrote:
> When the spec for an element says =E2=80=9CMay contain: Empty element=E2=
=80=9C that
> does not actually mean the spec=E2=80=99d element can contain any other=

> element as long as the contained one is empty, right=3F It actually
> means =E2=80=9CMay contain: No other element,=E2=80=9C right=3F
>
> Should that be changed=3F
>
>

The =22may contain=22 part is boiler plate text provided by the styleshee=
t
in various languages: it might be tricky to change it. The other part is
meant to imply that the element in question cannot contain anything,
i.e. it is empty. Since there is now a proposal in the works (see
https://github.com/TEIC/TEI/issues/1596)=C2=A0 to allow for an empty cont=
ent

Pure-ODD content model elements appear to be strictly weaker than regular=
 languages -- intended=3F =C2=B7 Issue =231596 =C2=B7 TEIC/TEI
github.com
The elements used for content models in pure ODDs include, if I understan=
d chapter 22 correctly: Elements denoting atomic units of the content mod=
el (anyElement, dataRef, elementRef, textNode, val...

model to be represented in an ODD by using an explicit element called
=22<empty/>=22=C2=A0=C2=A0 that text maybe needs to be revisited. You mig=
ht like to
raise a Stylesheets ticket to remind someone to review it.


------------------------------

Date:=C2=A0=C2=A0=C2=A0 Thu, 4 May 2017 12:45:53 +0200
=46rom:=C2=A0=C2=A0=C2=A0 Emmanuelle Morlock <emmanuelle.morlock=40MOM.=46=
R>
Subject: question about attribute =40writtenLines (metadata)

Dear list,

I was just wondering why the attribute=C2=A0=40writtenLines on layout can=
 contain either one or two numbers (to represent the number of lines of o=
ne or two columns), but not more.

My question is mainly out of curiosity, I don=E2=80=99t have any use case=
, but it just seems slightly odd (the I may assume that in the manuscript=
 world it=E2=80=99s very rare to have more than two columns=E2=80=A6) but=
 what if =3F How would you do if you had more than two columns =3F=C2=A0

thanks =21
Best,=C2=A0
--=C2=A0
Emmanuelle Morlock
IE CNRS - Humanit=C3=A9s num=C3=A9riques & TEI (Text Encoding Initiative)=

UMR 5189 HISoMA (Histoire et Sources des Mondes antiques) - Lyon
http://www.hisoma.mom.fr/annuaire/morlock-emmanuelle
MORLOCK Emmanuelle =7C Laboratoire HISOMA
www.hisoma.mom.fr
Activit=C3=A9s principalesSoutien au projets de d'humanit=C3=A9s num=C3=A9=
riques et d'=C3=A9dition de sources anciennes en XML/TEI ;Conception et =C3=
=A9laboration de proc=C3=A9dures de ...

06 85 84 69 16
=40emma=5Fmorlock

------------------------------

Date:=C2=A0=C2=A0=C2=A0 Thu, 4 May 2017 10:16:19 -0400
=46rom:=C2=A0=C2=A0=C2=A0 =22John P. McCaskey=22 <mailbox=40JOHNMCCASKEY.=
COM>
Subject: Re: =40xml:base with =40rendition (and maybe other pointers)

I asked about this over at XML-DEV,
http://lists.xml.org/archives/xml-dev/201705/msg00008.html.

Opinion there is with me and opposite the majority here.

Readers there don=E2=80=99t seem to think there is anything to debate. I =
got one
short answer and one dismissive comment about =E2=80=9Dquestions that sim=
ple.=E2=80=9D
Someone take a look and be sure I didn=E2=80=99t word my question unfairl=
y.

Whichever interpretation TEI adopts, sounds like it should be documented
in the Guidelines or a note somewhere.

If the McCaskey/XML-DEV interpretation is adopted, the GitHub issue I
posted about =40rendition stands. If not, that issue goes away.

John



On 5/3/2017 8:03 PM, Hugh Cayless wrote:
> Those are the cards we've been dealt, yeah.
>
> Sent from my phone.
>
> On May 3, 2017, at 19:53, John P. McCaskey <mailbox=40JOHNMCCASKEY.COM
> <mailto:mailbox=40JOHNMCCASKEY.COM>> wrote:
>
>> So, bottom line: A standalone fragment identifier refers to the
>> loaded document and all the xml:base values above it in the hierarchy
>> are irrelevant. Encoders cannot use xml:base to direct a standalone
>> =23fragment value to a location outside the loaded document.
>>
>> Is that right=3F
>>
>> --
>>
>>
>> On 5/3/2017 6:20 PM, Hugh Cayless wrote:
>>> No. This clarifies the expected behavior when (e.g.) you have
>>> =40xml:base=3D=22=23frag=22. It says nothing whatever about <p
>>> rendition=3D=22=23foo=22>. I don't think these specifications interac=
t in
>>> the way you're positing. Quite the opposite.
>>>
>>> On Wed, May 3, 2017 at 6:13 PM, John P. McCaskey
>>> <mailbox=40johnmccaskey.com <mailto:mailbox=40johnmccaskey.com>> wrot=
e:
>>>
>>>=C2=A0=C2=A0=C2=A0=C2=A0 The special behavior of same-document referen=
ces in R=46C 3986 is
>>>=C2=A0=C2=A0=C2=A0=C2=A0 disavowed by W3C:
>>>
>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 4.4 Interpretation of=
 same-document references
>>>
>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 R=46C 3986 defines ce=
rtain relative URI references, in
>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 particular the empty =
string and those of the form =23fragment,
>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 as same-document refe=
rences. Dereferencing of same-document
>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 references is handled=
 specially. However, their use as the
>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 value of an *xml:base=
 attribute does not* *involve
>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 *dereferencing, and X=
ML Base processors should resolve them
>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 in the usual way. In =
particular, xml:base=3D=22=22 does not reset
>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 the base URI to that =
of the containing document.
>>>
>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 Note:
>>>
>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 Some existing process=
ors do treat these xml:base values as
>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 resetting the base UR=
I to that of the containing document,
>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 so the use of such va=
lues is strongly discouraged.
>>>
>>>=C2=A0=C2=A0=C2=A0=C2=A0 This says:
>>>
>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 R=46C 3986 defines sp=
ecial =E2=80=9Cdereferencing=E2=80=9D of empty strings
>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 and =23fragments. But=
 over here in XML-land, we don=E2=80=99t do
>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 =E2=80=9Cdereferencin=
g.=E2=80=9C That=E2=80=99s not a word we use here. Ignore that
>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 stuff about same-docu=
ment references. Just resolve empty
>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 strings and =23fragme=
nts as specified in the W3C
>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 Recommendation above.=
 And those of you who did carry that
>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 stuff over from 3986 =
to XML-land, shame on you. You messed
>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 things up for the res=
t of us.
>>>
>>>
>>>=C2=A0=C2=A0=C2=A0=C2=A0 A note has:
>>>
>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 5. The meanings of xm=
l:base=3D=22=22 and xml:base=3D=22=23frag=22 have
>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 been clarified;
>>>
>>>=C2=A0=C2=A0=C2=A0=C2=A0 This says:
>>>
>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 In this second versio=
n of this recommendation, we added
>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 paragraph 4.4 specifi=
cally to get you 3986 people to stop
>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 polluting our W3C wit=
h your special cases. Stop doing that.
>>>
>>>=C2=A0=C2=A0=C2=A0=C2=A0 No=3F
>>>
>>>=C2=A0=C2=A0=C2=A0=C2=A0 -- John
>>>
>>>
>>>
>>>
>>>
>>>
>>>=C2=A0=C2=A0=C2=A0=C2=A0 On 5/3/2017 5:09 PM, Hugh Cayless wrote:
>>>>=C2=A0=C2=A0=C2=A0=C2=A0 That's exactly what it does. It's just that =
the behavior of
>>>>=C2=A0=C2=A0=C2=A0=C2=A0 same-document references is prescribed in su=
ch a way that they
>>>>=C2=A0=C2=A0=C2=A0=C2=A0 end up resolving to the current document reg=
ardless of the
>>>>=C2=A0=C2=A0=C2=A0=C2=A0 value of the =40xml:base.
>>>>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0 Put another way, =40xml:base has an influenc=
e on what the client
>>>>=C2=A0=C2=A0=C2=A0=C2=A0 application will retrieve when it dereferenc=
es a URI and
>>>>=C2=A0=C2=A0=C2=A0=C2=A0 retrieves the referenced document, but in th=
e case of
>>>>=C2=A0=C2=A0=C2=A0=C2=A0 same-document references no such retrieval i=
s expected to
>>>>=C2=A0=C2=A0=C2=A0=C2=A0 occur=E2=80=94it's assumed the client alread=
y has the document.
>>>>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0 On Wed, May 3, 2017 at 4:29 PM, John P. McCa=
skey
>>>>=C2=A0=C2=A0=C2=A0=C2=A0 <mailbox=40johnmccaskey.com <mailto:mailbox=40=
johnmccaskey.com>> wrote:
>>>>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 Oh. I thought it obv=
ious they all resolved the same.
>>>>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 I think the same abo=
ut this one. You think otherwise=3F
>>>>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 <div xml:base=3D=22h=
ttp://www.myteiproject.com/=22
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 <http://www.myteipro=
ject.com/> <div xml:base=3D=22images/=22>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 <div> <graphic url=3D=
=22logo.jpg=22> </div> <div
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 xml:base=3D=22http:/=
/www.dictionary.com/words/
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 <http://dictionary.c=
om/words/>=22> <p xml:base=3D=22a.html=22> <ref
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 target=3D=22=23apple=
=22>apple</ref> </p> </div> </div> </div>
>>>>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 I didn=E2=80=99t thi=
nk xml:base had any necessary relation to the
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 current document. I =
thought in XML (not HTML), it just sets
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 a path for URIs belo=
w it.
>>>>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 --
>>>>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 On 5/3/2017 3:42 PM,=
 Hugh Cayless wrote:
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 Well, we don't know=
 about the third one, except that it
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 points to whatever =
element *in the same document* has the
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 =40xml:id =22apple=22=
. When an application attempts to
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 dereference it, it =
should make the URI in =40ref absolute
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 and check it agains=
t the base
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 (http://www.diction=
ary.com/words/a.html
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 <http://www.diction=
ary.com/words/a.html>), discover they
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 are identical, deci=
de it doesn't need to fetch anything,
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 and go looking in t=
he current document for the element
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 with the id =22appl=
e=22.
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 On Wed, May 3, 2017=
 at 3:27 PM, John P. McCaskey
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 <mailbox=40johnmcca=
skey.com
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 <mailto:mailbox=40j=
ohnmccaskey.com>> wrote:
>>>>>
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 Are people proposing that these targets do not all
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 resolve to the same
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 http://www.dictionary.com/words/a.html=23apple
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 <http://www.dictionary.com/words/a.html=23apple>=3F <div
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 xml:base=3D=22http://www.dictionary.com/=22
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 <http://www.dictionary.com/>> <p xml:base=3D=22words/=22>
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 <ref target=3D=22a.html=23apple=22>apple</ref> </p> </div>
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 <div xml:base=3D=22http://www.dictionary.com/words/=22
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 <http://www.dictionary.com/words/>> <p
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 xml:base=3D=22a.html=22> <ref target=3D=22=23apple=22>apple</ref>
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 </p> </div> <div
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 xml:base=3D=22http://www.dictionary.com/words/a.html=22
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 <http://www.dictionary.com/words/a.html>> <p> <ref
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 target=3D=22=23apple=22>apple</ref> </p> </div> <div
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 xml:base=3D=22http://www.dictionary.com/=22
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 <http://www.dictionary.com/>> <p> <ref
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 target=3D=22words/a.html=23apple=22>apple</ref> </p> </div>--
>>>>>

------------------------------

Date:=C2=A0=C2=A0=C2=A0 Thu, 4 May 2017 11:37:19 -0600
=46rom:=C2=A0=C2=A0=C2=A0 =22C. M. Sperberg-McQueen=22 <cmsmcq=40BLACKMES=
ATECH.COM>
Subject: Re: =40xml:base with =40rendition (and maybe other pointers)

> On May 3, 2017, at 5:53 PM, John P. McCaskey <mailbox=40JOHNMCCASKEY.CO=
M> wrote:
>
> So, bottom line: A standalone fragment identifier refers to the loaded =
document and all the xml:base values above it in the hierarchy are irrele=
vant. Encoders cannot use xml:base to direct a standalone =23fragment val=
ue to a location outside the loaded document.
>
> Is that right=3F

Yes and no.=C2=A0 =46or some purposes, the ways in which the answer is 'n=
o'
are pedantic and can be ignored; for others they seem important.
I did not respond to your summary yesterday, because objecting to
the wording you used seemed unnecessarily pedantic.=C2=A0 Also, I
overlooked the second sentence, which I think is the wrong conclusion
to draw.

Given the following fragment of resource http://example.org/eg.xml

=C2=A0 <div xml:base=3D=22http://www.dictionary.com/a.html=22>
=C2=A0=C2=A0=C2=A0 <p>
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 <ref target=3D=22=23apple=22>Apple</ref>
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 <ref target=3D=22a.html=23avocado=22>Avoca=
do</ref>
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 <ref target=3D=22http://www.dictionary.com=
/a.html=23anise=22>Anise</ref>
=C2=A0=C2=A0=C2=A0 </p>
=C2=A0 </div>

we can consider two ways of interpreting the target attributes.

Note that the discussion below ignores some possibly salient facts:

- URIs can denote different resources at different moments.
=C2=A0=C2=A0=C2=A0 (The discussion assumes the URI-resource mapping is no=
t changing.)
=C2=A0=C2=A0=C2=A0
=C2=A0 - A given resource can have multiple representations.=C2=A0 (The
=C2=A0=C2=A0=C2=A0 discussion ignores any resulting complications.)
=C2=A0=C2=A0=C2=A0
=C2=A0 - URIs whose path component ends in .xml and .html do not
=C2=A0=C2=A0=C2=A0=C2=A0 necessarily have particular MIME types, so there=
 is no guarantee
=C2=A0=C2=A0=C2=A0=C2=A0 that a fragment identifier like =23apple will ha=
ve similar
=C2=A0=C2=A0=C2=A0=C2=A0 meanings.=C2=A0 (The discussion assumes that fra=
gment identifiers
=C2=A0=C2=A0=C2=A0=C2=A0 point to elements assigned IDs by the HTML 'id' =
attribute
=C2=A0=C2=A0=C2=A0=C2=A0 and/or the xml:id attribute.)=C2=A0=C2=A0=C2=A0=C2=
=A0

Interpretation 1 ('xml:base values are irrelevant').=C2=A0 Ignore xml:bas=
e
when resolving '=23apple' =5Bbut not when resolving other relative
references=5D.=C2=A0 The target attributes are interpreted as denoting

=C2=A0=C2=A0=C2=A0 (a) http://example.org/eg.xml=23apple
=C2=A0=C2=A0=C2=A0 (b) http://www.dictionary.com/a.html=23avocado
=C2=A0=C2=A0=C2=A0 (c) http://www.dictionary.com/a.html=23anise

None of these have anything to do with any of:

=C2=A0=C2=A0=C2=A0 (d) http://www.dictionary.com/a.html=23apple
=C2=A0=C2=A0=C2=A0 (e) http://example.org/eg.xml=23avocado
=C2=A0=C2=A0=C2=A0 (f) http://example.org/eg.xml=23anise=C2=A0=C2=A0=C2=A0=


Of these, (a) is a same-document reference and R=46C 3986 says it
=22should=22 be dereferenced without a new retrieval action.=C2=A0 If a n=
ew
retrieval action is nevertheless launched, the resource retrieved is
(a).

=5BIt is not clear to me whether the XVAI ('xml:base values are
irrelevant') interpretation takes a position on whether any of these
other than (a) are same-document references.=5D

Interpretation 2 ('same-document references can be surprising', or
SRCBS).=C2=A0 Resolve all relative references against the base URI in the=

usual way.=C2=A0 Dereference same-document references either by looking i=
n
the same document (as recommended by R=46C 3986) or by launching a new
retrieval operation.

The target attributes are resolved to the absolute forms

=C2=A0=C2=A0=C2=A0 (d) http://www.dictionary.com/a.html=23apple
=C2=A0=C2=A0=C2=A0 (b) http://www.dictionary.com/a.html=23avocado
=C2=A0=C2=A0=C2=A0 (c) http://www.dictionary.com/a.html=23anise

All of these are same-document references, so according to R=46C 3986,
they should be dereferenced without a new retrieval action.=C2=A0 If a
retrieval action is nevertheless launched, it will go to URIs (d),
(b), (c) respectively, not (a), (e), (f).=C2=A0 =46rom the fact that (d),=

(b), and (c) can be dereferenced without a new retrieval, it follows
(as far as I can tell) that these three resources can also be denoted
by URIs (a), (e), (f).

The XVAI and SRCBS interpretations agree on the following proposition,
which has important relevance for operations on the data:

P1 The relative reference target=3D=22=23apple=22 can be dereferenced by
locating the element in the current document with xml:id=3D=22apple=22, i=
f
such an element exists.

=46or people whose main interest is the truth or falsity of that
proposition, then, the answer is =22yes, that's right=22 -- the effect is=

the same, and all else is just pilpul.

=46rom P1, it follows (I think) that

P2 The resource identified by target=3D=22=23apple=22 is identified by th=
e
absolute URI http://example.org/eg.xml=23apple ((a) above).

The two interpretations disagree, or seem to disagree, on a number of
other propositions, most obviously:

P3 The relative reference '=23apple' does not identify the resource
identifed by http://www.dictionary.com/a.html=23apple.

XVAI does not actually entail P2, but it is compatible with P3.=C2=A0 (To=

reach P3 it is necessary to assume some rule like =22No two URIs
identify the same thing=22 or =22If we don't know that a URI identifies a=

thing, then it does not identify that thing.=22)=C2=A0 SRCBS entails the
negation of P3.

They also prescribe different URIs for the case that software
determines to perform a fresh retrieval action for the relative
reference =23apple: XVAI prescribes the absolute URI (a), SRCBS
prescribes URI (d).

I don=E2=80=99t believe anyone has seriously suggested XVAI as the
relevant rule of interpretation for examples like the one given;
what I have suggested (and I have understood Hugh Cayless
to be agreeing with) is SRCBS.=C2=A0 Operationally, they can have
similar results in some circumstances (specifically:=C2=A0 they can
both result in no new retrieval action being undertaken in order
to dereference =E2=80=98=23apple=E2=80=99), but they differ in ways which=
 can be
critical.


********************************************
C. M. Sperberg-McQueen
Black Mesa Technologies LLC
cmsmcq=40blackmesatech.com
http://www.blackmesatech.com
********************************************

------------------------------

Date:=C2=A0=C2=A0=C2=A0 Thu, 4 May 2017 14:28:37 -0400
=46rom:=C2=A0=C2=A0=C2=A0 Hugh Cayless <philomousos=40GMAIL.COM>
Subject: Re: =40xml:base with =40rendition (and maybe other pointers)

That's a rather favorable interpretation on your part. One person agrees
with you without elucidating, one says this discussion has jumped the sha=
rk
(which is fair), and the third (Michael Kay) gives a fuller answer which
adds up to =22it depends=22. Michael Kay is quite correct that in the con=
text
where a document retrieval is *expected* to occur, the URI would indeed b=
e
computed with reference to its base and fetched.

The thing is, I'm not aware of any TEI attributes or element/attribute
combinations which are defined as *forcing* a retrieval action. I'd be
happy to be corrected if I'm missing any, of course.

It's fair to ask not just how one might expect them to behave, but what
same-document references *mean* in the context of TEI documents with
=40xml:base. I agree this is something we ought to make clear. I think th=
ere
is some possibility of wiggle room, given that TEI has its own media type=
.
But I also think that we'd be better off adhering to the letter of R=46C
3986. The use of same-document references in TEI documents is ubiquitous,=

and I'm firmly against anything that might break them.

=46or what it's worth, modern web browsers seem to agree with your
interpretation (mutatis mutandis=E2=80=94HTML base is not =40xml:base). A=
s far as I
can tell, probably because of a desire on the part of the Mozilla
developers back in the day to maintain compatibility with IE 4(=21).=5B1=5D=


To further complicate matters, the author of R=46C 3986, Roy =46ielding, =
has
said that using =40xml:base in the way you propose, i.e. to enable shorth=
and
references rather than to set a canonical URI for the current document, i=
s
abusive.=5B2=5D

Given all this, I still agree with Michael Sperberg-McQueen's fuller
explication of the issues at hand. I would interpret same-document
references as pointing to the current document, with the caveat that ther=
e
might be, now or in the future, certain pointer attributes or
element/attribute combinations that mandate a retrieval action. Such a
retrieval would necessarily use whatever base was defined for the URI in
question.

References:
1.
http://w3future.com/weblog/2005/01/13.xml=23stillBugsInTheImplementationO=
fHtmlHyperlinks
2. http://w3future.com/weblog/2005/08/14.xml=23howToUseBaseUris

On Thu, May 4, 2017 at 10:16 AM, John P. McCaskey <mailbox=40johnmccaskey=
.com>
wrote:

> I asked about this over at XML-DEV, http://lists.xml.org/archives/
> xml-dev/201705/msg00008.html.
>
> Opinion there is with me and opposite the majority here.
>
> Readers there don=E2=80=99t seem to think there is anything to debate. =
I got one
> short answer and one dismissive comment about =E2=80=9Dquestions that s=
imple.=E2=80=9D
> Someone take a look and be sure I didn=E2=80=99t word my question unfai=
rly.
> Whichever interpretation TEI adopts, sounds like it should be documente=
d
> in the Guidelines or a note somewhere.
>
> If the McCaskey/XML-DEV interpretation is adopted, the GitHub issue I
> posted about =40rendition stands. If not, that issue goes away.
>
> John
>
>
>
> On 5/3/2017 8:03 PM, Hugh Cayless wrote:
>
> Those are the cards we've been dealt, yeah.
>
> Sent from my phone.
>
> On May 3, 2017, at 19:53, John P. McCaskey <mailbox=40JOHNMCCASKEY.COM>=

> wrote:
>
> So, bottom line: A standalone fragment identifier refers to the loaded
> document and all the xml:base values above it in the hierarchy are
> irrelevant. Encoders cannot use xml:base to direct a standalone =23frag=
ment
> value to a location outside the loaded document.
>
> Is that right=3F
>
> --
>
>
> On 5/3/2017 6:20 PM, Hugh Cayless wrote:
>
> No. This clarifies the expected behavior when (e.g.) you have =40xml:ba=
se=3D=22=23frag=22.
> It says nothing whatever about <p rendition=3D=22=23foo=22>. I don't th=
ink these
> specifications interact in the way you're positing. Quite the opposite.=

>
> On Wed, May 3, 2017 at 6:13 PM, John P. McCaskey <mailbox=40johnmccaske=
y.com
> > wrote:
>
>> The special behavior of same-document references in R=46C 3986 is disa=
vowed
>> by W3C:
>>
>> 4.4 Interpretation of same-document references
>>
>> R=46C 3986 defines certain relative URI references, in particular the =
empty
>> string and those of the form =23fragment, as same-document references.=

>> Dereferencing of same-document references is handled specially. Howeve=
r,
>> their use as the value of an *xml:base attribute does not* *involve *d=
ereferencing,
>> and XML Base processors should resolve them in the usual way. In
>> particular, xml:base=3D=22=22 does not reset the base URI to that of t=
he
>> containing document.
>>
>> Note:
>>
>> Some existing processors do treat these xml:base values as resetting t=
he
>> base URI to that of the containing document, so the use of such values=
 is
>> strongly discouraged.
>>
>> This says:
>>
>> R=46C 3986 defines special =E2=80=9Cdereferencing=E2=80=9D of empty st=
rings and =23fragments.
>> But over here in XML-land, we don=E2=80=99t do =E2=80=9Cdereferencing.=
=E2=80=9C That=E2=80=99s not a word
>> we use here. Ignore that stuff about same-document references. Just re=
solve
>> empty strings and =23fragments as specified in the W3C Recommendation =
above.
>> And those of you who did carry that stuff over from 3986 to XML-land, =
shame
>> on you. You messed things up for the rest of us.
>>
>>
>> A note has:
>>
>> 5. The meanings of xml:base=3D=22=22 and xml:base=3D=22=23frag=22 have=
 been clarified;
>>
>> This says:
>>
>> In this second version of this recommendation, we added paragraph 4.4
>> specifically to get you 3986 people to stop polluting our W3C with you=
r
>> special cases. Stop doing that.
>>
>> No=3F
>>
>> -- John
>>
>>
>>
>>
>>
>>
>> On 5/3/2017 5:09 PM, Hugh Cayless wrote:
>>
>> That's exactly what it does. It's just that the behavior of same-docum=
ent
>> references is prescribed in such a way that they end up resolving to t=
he
>> current document regardless of the value of the =40xml:base.
>>
>> Put another way, =40xml:base has an influence on what the client
>> application will retrieve when it dereferences a URI and retrieves the=

>> referenced document, but in the case of same-document references no su=
ch
>> retrieval is expected to occur=E2=80=94it's assumed the client already=
 has the
>> document.
>>
>> On Wed, May 3, 2017 at 4:29 PM, John P. McCaskey <
>> mailbox=40johnmccaskey.com> wrote:
>>
>>> Oh. I thought it obvious they all resolved the same.
>>>
>>> I think the same about this one. You think otherwise=3F
>>>
>>> <div xml:base=3D=22http://www.myteiproject.com/=22 <http://www.myteip=
roject.com/>
>>>=C2=A0=C2=A0 <div xml:base=3D=22images/=22>
>>>=C2=A0=C2=A0=C2=A0=C2=A0 <div>
>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 <graphic url=3D=22logo.jpg=22>
>>>=C2=A0=C2=A0=C2=A0=C2=A0 </div>
>>>=C2=A0=C2=A0=C2=A0=C2=A0 <div xml:base=3D=22http://www.dictionary.com/=
words/=22>
>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 <p xml:base=3D=22a.html=22>
>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 <ref target=3D=22=23a=
pple=22>apple</ref>
>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 </p>
>>>=C2=A0=C2=A0=C2=A0=C2=A0 </div>
>>>=C2=A0=C2=A0 </div>
>>> </div>
>>>
>>> I didn=E2=80=99t think xml:base had any necessary relation to the cur=
rent
>>> document. I thought in XML (not HTML), it just sets a path for URIs b=
elow
>>> it.
>>>
>>> --
>>> On 5/3/2017 3:42 PM, Hugh Cayless wrote:
>>>
>>> Well, we don't know about the third one, except that it points to
>>> whatever element *in the same document* has the =40xml:id =22apple=22=
. When
>>> an application attempts to dereference it, it should make the URI in =
=40ref
>>> absolute and check it against the base (http://www.dictionary.com/wor=

>>> ds/a.html), discover they are identical, decide it doesn't need to
>>> fetch anything, and go looking in the current document for the elemen=
t with
>>> the id =22apple=22.
>>> On Wed, May 3, 2017 at 3:27 PM, John P. McCaskey <
>>> mailbox=40johnmccaskey.com> wrote:
>>>>
>>>> Are people proposing that these targets do not all resolve
>>>> to the same http://www.dictionary.com/words/a.html=23apple=3F
>>>> <div xml:base=3D=22http://www.dictionary.com/=22 <http://www.diction=
ary.com/>>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0 <p xml:base=3D=22words/=22>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 <ref target=3D=22a.h=
tml=23apple=22>apple</ref>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0 </p>
>>>> </div>
>>>>
>>>> <div xml:base=3D=22http://www.dictionary.com/words/=22 <http://www.d=
ictionary.com/words/>>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0 <p xml:base=3D=22a.html=22>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 <ref target=3D=22=23=
apple=22>apple</ref>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0 </p>
>>>> </div>
>>>>
>>>> <div xml:base=3D=22http://www.dictionary.com/words/a.html=22 <http:/=
/www.dictionary.com/words/a.html>>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0 <p>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 <ref target=3D=22=23=
apple=22>apple</ref>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0 </p>
>>>> </div>
>>>>
>>>> <div xml:base=3D=22http://www.dictionary.com/=22 <http://www.diction=
ary.com/>>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0 <p>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 <ref target=3D=22wor=
ds/a.html=23apple=22>apple</ref>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0 </p>
>>>> </div>
>>>> --
>>>>
>>>>

------------------------------

Date:=C2=A0=C2=A0=C2=A0 Thu, 4 May 2017 15:31:42 -0400
=46rom:=C2=A0=C2=A0=C2=A0 =22John P. McCaskey=22 <mailbox=40JOHNMCCASKEY.=
COM>
Subject: Re: =40xml:base with =40rendition (and maybe other pointers)

What is the short bottom-line guidance for someone trying to encode a
document=3F Is it this=3F

=C2=A0=C2=A0=C2=A0 To point inside a TEI document, as for =40rendition, u=
se pointers of
=C2=A0=C2=A0=C2=A0 the form =23destination. Any later changes to xml:base=
 values will not
=C2=A0=C2=A0=C2=A0 mess up your internal =23 pointers.

=C2=A0=C2=A0=C2=A0 Don=E2=80=99t try to point outside your TEI document b=
y using xml:base plus
=C2=A0=C2=A0=C2=A0 a =23destination in the pointer attribute. No xml:base=
 attribute will
=C2=A0=C2=A0=C2=A0 be prepended to a pointer that begins with a =23.

Even if I and others find that idiosyncratic and surprising, it=E2=80=99s=

unambiguous, has practical benefits, requires no knowledge of R=46Cs or
W3C specs, is easy to articulate, and is easy to encode to.

John



On 5/4/2017 2:28 PM, Hugh Cayless wrote:
> That's a rather favorable interpretation on your part. One person
> agrees with you without elucidating, one says this discussion has
> jumped the shark (which is fair), and the third (Michael Kay) gives a
> fuller answer which adds up to =22it depends=22. Michael Kay is quite
> correct that in the context where a document retrieval is *expected*
> to occur, the URI would indeed be computed with reference to its base
> and fetched.
>
> The thing is, I'm not aware of any TEI attributes or element/attribute
> combinations which are defined as *forcing* a retrieval action. I'd be
> happy to be corrected if I'm missing any, of course.
>
> It's fair to ask not just how one might expect them to behave, but
> what same-document references *mean* in the context of TEI documents
> with =40xml:base. I agree this is something we ought to make clear. I
> think there is some possibility of wiggle room, given that TEI has its
> own media type. But I also think that we'd be better off adhering to
> the letter of R=46C 3986. The use of same-document references in TEI
> documents is ubiquitous, and I'm firmly against anything that might
> break them.
>
> =46or what it's worth, modern web browsers seem to agree with your
> interpretation (mutatis mutandis=E2=80=94HTML base is not =40xml:base).=
 As far
> as I can tell, probably because of a desire on the part of the Mozilla
> developers back in the day to maintain compatibility with IE 4(=21).=5B=
1=5D
>
> To further complicate matters, the author of R=46C 3986, Roy =46ielding=
,
> has said that using =40xml:base in the way you propose, i.e. to enable
> shorthand references rather than to set a canonical URI for the
> current document, is abusive.=5B2=5D
>
> Given all this, I still agree with Michael Sperberg-McQueen's fuller
> explication of the issues at hand. I would interpret same-document
> references as pointing to the current document, with the caveat that
> there might be, now or in the future, certain pointer attributes or
> element/attribute combinations that mandate a retrieval action. Such a
> retrieval would necessarily use whatever base was defined for the URI
> in question.
>
> References:
> 1.
> http://w3future.com/weblog/2005/01/13.xml=23stillBugsInTheImplementatio=
nOfHtmlHyperlinks
> 2. http://w3future.com/weblog/2005/08/14.xml=23howToUseBaseUris
>
> On Thu, May 4, 2017 at 10:16 AM, John P. McCaskey
> <mailbox=40johnmccaskey.com <mailto:mailbox=40johnmccaskey.com>> wrote:=

>
>=C2=A0=C2=A0=C2=A0=C2=A0 I asked about this over at XML-DEV,
>=C2=A0=C2=A0=C2=A0=C2=A0 http://lists.xml.org/archives/xml-dev/201705/ms=
g00008.html
>=C2=A0=C2=A0=C2=A0=C2=A0 <http://lists.xml.org/archives/xml-dev/201705/m=
sg00008.html>.
>
>=C2=A0=C2=A0=C2=A0=C2=A0 Opinion there is with me and opposite the major=
ity here.
>
>=C2=A0=C2=A0=C2=A0=C2=A0 Readers there don=E2=80=99t seem to think there=
 is anything to debate. I
>=C2=A0=C2=A0=C2=A0=C2=A0 got one short answer and one dismissive comment=
 about =E2=80=9Dquestions
>=C2=A0=C2=A0=C2=A0=C2=A0 that simple.=E2=80=9D Someone take a look and b=
e sure I didn=E2=80=99t word my
>=C2=A0=C2=A0=C2=A0=C2=A0 question unfairly.
>
>=C2=A0=C2=A0=C2=A0=C2=A0 Whichever interpretation TEI adopts, sounds lik=
e it should be
>=C2=A0=C2=A0=C2=A0=C2=A0 documented in the Guidelines or a note somewher=
e.
>
>=C2=A0=C2=A0=C2=A0=C2=A0 If the McCaskey/XML-DEV interpretation is adopt=
ed, the GitHub
>=C2=A0=C2=A0=C2=A0=C2=A0 issue I posted about =40rendition stands. If no=
t, that issue goes away.
>
>=C2=A0=C2=A0=C2=A0=C2=A0 John
>
>
>
>=C2=A0=C2=A0=C2=A0=C2=A0 On 5/3/2017 8:03 PM, Hugh Cayless wrote:
>>=C2=A0=C2=A0=C2=A0=C2=A0 Those are the cards we've been dealt, yeah.
>>
>>=C2=A0=C2=A0=C2=A0=C2=A0 Sent from my phone.
>>
>>=C2=A0=C2=A0=C2=A0=C2=A0 On May 3, 2017, at 19:53, John P. McCaskey
>>=C2=A0=C2=A0=C2=A0=C2=A0 <mailbox=40JOHNMCCASKEY.COM <mailto:mailbox=40=
JOHNMCCASKEY.COM>> wrote:
>>
>>>=C2=A0=C2=A0=C2=A0=C2=A0 So, bottom line: A standalone fragment identi=
fier refers to the
>>>=C2=A0=C2=A0=C2=A0=C2=A0 loaded document and all the xml:base values a=
bove it in the
>>>=C2=A0=C2=A0=C2=A0=C2=A0 hierarchy are irrelevant. Encoders cannot use=
 xml:base to direct
>>>=C2=A0=C2=A0=C2=A0=C2=A0 a standalone =23fragment value to a location =
outside the loaded
>>>=C2=A0=C2=A0=C2=A0=C2=A0 document.
>>>
>>>=C2=A0=C2=A0=C2=A0=C2=A0 Is that right=3F
>>>
>>>=C2=A0=C2=A0=C2=A0=C2=A0 --
>>>
>>>
>>>=C2=A0=C2=A0=C2=A0=C2=A0 On 5/3/2017 6:20 PM, Hugh Cayless wrote:
>>>>=C2=A0=C2=A0=C2=A0=C2=A0 No. This clarifies the expected behavior whe=
n (e.g.) you have
>>>>=C2=A0=C2=A0=C2=A0=C2=A0 =40xml:base=3D=22=23frag=22. It says nothing=
 whatever about <p
>>>>=C2=A0=C2=A0=C2=A0=C2=A0 rendition=3D=22=23foo=22>. I don't think the=
se specifications interact
>>>>=C2=A0=C2=A0=C2=A0=C2=A0 in the way you're positing. Quite the opposi=
te.
>>>>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0 On Wed, May 3, 2017 at 6:13 PM, John P. McCa=
skey
>>>>=C2=A0=C2=A0=C2=A0=C2=A0 <mailbox=40johnmccaskey.com <mailto:mailbox=40=
johnmccaskey.com>> wrote:
>>>>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 The special behavior=
 of same-document references in R=46C
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 3986 is disavowed by=
 W3C:
>>>>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 4.4 Interpretation of same-document references
>>>>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 R=46C 3986 defines certain relative URI references, in
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 particular the empty string and those of the form
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 =23fragment, as same-document references. Dereferencing
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 of same-document references is handled specially.
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 However, their use as the value of an *xml:base
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 attribute does not* *involve *dereferencing, and XML
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 Base processors should resolve them in the usual way.
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 In particular, xml:base=3D=22=22 does not reset the base URI
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 to that of the containing document.
>>>>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 Note:
>>>>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 Some existing processors do treat these xml:base values
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 as resetting the base URI to that of the containing
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 document, so the use of such values is strongly
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 discouraged.
>>>>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 This says:
>>>>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 R=46C 3986 defines special =E2=80=9Cdereferencing=E2=80=9D of empty
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 strings and =23fragments. But over here in XML-land, we
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 don=E2=80=99t do =E2=80=9Cdereferencing.=E2=80=9C That=E2=80=99s not =
a word we use
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 here. Ignore that stuff about same-document references.
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 Just resolve empty strings and =23fragments as specified
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 in the W3C Recommendation above. And those of you who
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 did carry that stuff over from 3986 to XML-land, shame
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 on you. You messed things up for the rest of us.
>>>>
>>>>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 A note has:
>>>>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 5. The meanings of xml:base=3D=22=22 and xml:base=3D=22=23frag=22
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 have been clarified;
>>>>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 This says:
>>>>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 In this second version of this recommendation, we added
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 paragraph 4.4 specifically to get you 3986 people to
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 stop polluting our W3C with your special cases. Stop
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 doing that.
>>>>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 No=3F
>>>>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 -- John
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 On 5/3/2017 5:09 PM,=
 Hugh Cayless wrote:
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 That's exactly what=
 it does. It's just that the behavior
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 of same-document re=
ferences is prescribed in such a way
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 that they end up re=
solving to the current document
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 regardless of the v=
alue of the =40xml:base.
>>>>>
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 Put another way, =40=
xml:base has an influence on what the
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 client application =
will retrieve when it dereferences a
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 URI and retrieves t=
he referenced document, but in the case
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 of same-document re=
ferences no such retrieval is expected
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 to occur=E2=80=94it=
's assumed the client already has the document.
>>>>>
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 On Wed, May 3, 2017=
 at 4:29 PM, John P. McCaskey
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 <mailbox=40johnmcca=
skey.com
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 <mailto:mailbox=40j=
ohnmccaskey.com>> wrote:
>>>>>
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 Oh. I thought it obvious they all resolved the same.
>>>>>
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 I think the same about this one. You think otherwise=3F
>>>>>
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 <div xml:base=3D=22http://www.myteiproject.com/=22
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 <http://www.myteiproject.com/> <div
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 xml:base=3D=22images/=22> <div> <graphic url=3D=22logo.jpg=22>
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 </div> <div xml:base=3D=22http://www.dictionary.com/words/
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 <http://dictionary.com/words/>=22> <p xml:base=3D=22a.html=22>
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 <ref target=3D=22=23apple=22>apple</ref> </p> </div> </div>
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 </div>
>>>>>
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 I didn=E2=80=99t think xml:base had any necessary relation to
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 the current document. I thought in XML (not HTML), it
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 just sets a path for URIs below it.
>>>>>
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 --
>>>>>
>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 On 5/3/2017 3:42 PM, Hugh Cayless wrote:
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 Well, we don't know about the third one, except that
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 it points to whatever element *in the same document*
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 has the =40xml:id =22apple=22. When an application attempts
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 to dereference it, it should make the URI in =40ref
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 absolute and check it against the base
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 (http://www.dictionary.com/words/a.html
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 <http://www.dictionary.com/words/a.html>), discover
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 they are identical, decide it doesn't need to fetch
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 anything, and go looking in the current document for
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 the element with the id =22apple=22.
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 On Wed, May 3, 2017 at 3:27 PM, John P. McCaskey
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 <mailbox=40johnmccaskey.com
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0 <mailto:mailbox=40johnmccaskey.com>> wrote:
>>>>>>
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0 Are people proposing that these targets do no=
t
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0 all resolve to the same
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0 http://www.dictionary.com/words/a.html=23appl=
e
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0 <http://www.dictionary.com/words/a.html=23app=
le>=3F
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0 <div xml:base=3D=22http://www.dictionary.com/=
=22
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0 <http://www.dictionary.com/>> <p
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0 xml:base=3D=22words/=22> <ref
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0 target=3D=22a.html=23apple=22>apple</ref> </p=
> </div>
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0 <div xml:base=3D=22http://www.dictionary.com/=
words/=22
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0 <http://www.dictionary.com/words/>> <p
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0 xml:base=3D=22a.html=22> <ref
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0 target=3D=22=23apple=22>apple</ref> </p> </di=
v> <div
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0 xml:base=3D=22http://www.dictionary.com/words=
/a.html=22
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0 <http://www.dictionary.com/words/a.html>> <p>=

>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0 <ref target=3D=22=23apple=22>apple</ref> </p>=
 </div> <div
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0 xml:base=3D=22http://www.dictionary.com/=22
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0 <http://www.dictionary.com/>> <p> <ref
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0 target=3D=22words/a.html=23apple=22>apple</re=
f> </p>
>>>>>>=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=
=A0=C2=A0=C2=A0=C2=A0=C2=A0 </div>--
>>>>>>

------------------------------

Date:=C2=A0=C2=A0=C2=A0 Thu, 4 May 2017 14:32:37 -0600
=46rom:=C2=A0=C2=A0=C2=A0 =22C. M. Sperberg-McQueen=22 <cmsmcq=40BLACKMES=
ATECH.COM>
Subject: Re: =40xml:base with =40rendition (and maybe other pointers)

> On May 4, 2017, at 12:28 PM, Hugh Cayless <philomousos=40GMAIL.COM> wro=
te:
>
> That's a rather favorable interpretation on your part. One person agree=
s with you without elucidating,

I=E2=80=99m not sure this is true.=C2=A0 What Eliot Kimber said is that i=
n the context given, =E2=80=98=23apple=E2=80=99 identifies the same thing=
 as =E2=80=98http://www.dictionary.com/a.html=23apple'.=C2=A0 Does that d=
istinguish between the two (=3F are there just two=3F) =E2=80=94 does tha=
t distinguish among the various interpretations of the situation offered =
so far=3F

I don=E2=80=99t think so.=C2=A0 I think the disagreement we have is not o=
ver the statement=C2=A0 affirmed by Eliot Kimber, but over the following =
two claims:

=C2=A0 C1 In the context described, =E2=80=98=23apple=E2=80=99 is a same-=
document reference and
=C2=A0 can therefore by definition be dereferenced without a new retrieva=
l action.

=C2=A0 C2 In the context described, =E2=80=98=23apple=E2=80=99 does not r=
efer to the element in the
=C2=A0 current document with xml:id=3D=E2=80=9Capple=E2=80=9D (if any); i=
t=C2=A0 cannot be dereferenced
=C2=A0 without a new retrieval action.

I intend C1 as a representation of the interpretation of 3986 I=E2=80=99v=
e been offering, and C2 as a representation of the interpretation offered=
 by John McCaskey. (SRCBS and XVAI, in my note of earlier today.

> one says this discussion has jumped the shark (which is fair), and the =
third (Michael Kay) gives a fuller answer which adds up to =22it depends=22=
. Michael Kay is quite correct that in the context where a document retri=
eval is *expected* to occur, the URI would indeed be computed with refere=
nce to its base and fetched.

The URI is *always* computed with reference to its base.=C2=A0

Optimizations which produce the same result are, of course, allowed.=C2=A0=
 The preceding paragraph is a claim about the meaning of certain language=
 constructs, not a claim about what the CPU and network controller do dur=
ing evaluation of an expression by a conforming processor.

If it is then determined to be a same-document URI, the resource identifi=
ed by that URI is then =E2=80=9Cdefined to be within=E2=80=9D the current=
 document; in consequence no new retrieval is necessary and a new retriev=
al should be avoided.=C2=A0 The 'should' here means that 3986 recommends =
that new retrievals be avoided, but does not forbid new retrievals; if co=
nforming processors or specs have good reason for launching new retrieval=
s, that's not a violation of the rules of 3986.=C2=A0 The XSLT spec defin=
es the document() function as always launching a new retrieval.=C2=A0 (No=
te that this does not amount to any claim by the XSLT spec that the relev=
ant resource is not within the current document.)

>
> The thing is, I'm not aware of any TEI attributes or element/attribute =
combinations which are defined as *forcing* a retrieval action. I'd be ha=
ppy to be corrected if I'm missing any, of course.
>
> It's fair to ask not just how one might expect them to behave, but what=
 same-document references *mean* in the context of TEI documents with =40=
xml:base. I agree this is something we ought to make clear. I think there=
 is some possibility of wiggle room, given that TEI has its own media typ=
e. But I also think that we'd be better off adhering to the letter of R=46=
C 3986. The use of same-document references in TEI documents is ubiquitou=
s, and I'm firmly against anything that might break them.

Is it clear what counts as breakage, here=3F

If a given interpretation of the URI specs causes some URI references to =
break (by which I mean: to have an interpretation different from what the=
 encoders intended), can we be confident that a contrary interpretation w=
ill not break any=3F=C2=A0 Or is it the case that one interpretation will=
 break some URI references, and a different interpretation will break oth=
ers=3F

When there are two possible interpreations of a given rule in a spec, it=E2=
=80=99s seldom the case that everyone interprets it the same way.=C2=A0 T=
here is some risk that your choice does not lie between breaking things i=
n TEI documents and not breaking them, but between breaking those belongi=
ng to one project and breaking those belonging to another project.

> =46or what it's worth, modern web browsers seem to agree with your inte=
rpretation (mutatis mutandis=E2=80=94HTML base is not =40xml:base). As fa=
r as I can tell, probably because of a desire on the part of the Mozilla =
developers back in the day to maintain compatibility with IE 4(=21).=5B1=5D=


Can you expound=3F=C2=A0

Do you mean that in an HTML 4.01 or XHTML 1.0 document at http://example.=
com/doc.html containing <base href=3D=E2=80=9Chttp://example.org/base.htm=
l=E2=80=9D>, a link of the form <a href=3D=E2=80=9C=23quovadis=E2=80=9D>w=
here does this connect to=3F</a> will go to =E2=80=A6 where=3F=C2=A0

I think the two obvious things one might expect to see in the address bar=
 after traversing that link are

=C2=A0=C2=A0=C2=A0 http://example.com/doc.html=23quovadis
=C2=A0=C2=A0=C2=A0 http://example.org/base.html=23quovaids

In the first case, this will happen either after a document refresh or wi=
thout a refresh (just scrolling to the place).

If a browser shows one of these addresses, does that constitute a claim t=
hat the other identifies a different resource=3F=C2=A0 Or does it only co=
nstitute a decision on the part of the browser about which of the two pos=
sible paths to a given resource it should take=3F


> To further complicate matters, the author of R=46C 3986, Roy =46ielding=
, has said that using =40xml:base in the way you propose, i.e. to enable =
shorthand references rather than to set a canonical URI for the current d=
ocument, is abusive.=5B2=5D

Thank you for that reference.

I think the rules in 3986 make perfect sense if one assumes Roy =46ieldin=
g=E2=80=99s principle that the base URI within a document should be the b=
ase URI of the document, and that using xml:base or html:base to shorten =
references which would otherwise be long is not a scenario worth bending =
over backwards for.=C2=A0 (That is, it seems to me to simplify matters, r=
ather than complicating them.)

If one doesn=E2=80=99t want to make that assumption, some relatively simp=
le rules like the following might go some distance towards reducing the l=
ikelihood of unpleasant surprises:

=C2=A0 - Use the form =E2=80=9C=23fragment=E2=80=9D only for references t=
o locations in the current document.=C2=A0 These will always be same-docu=
ment references within the meaning of R=46C 3986.
=C2=A0 - Use xml:base to set all but the last bit of the URI, but not to =
set a full document URI:=C2=A0 xml:base=3D=E2=80=9Chttp://dictionary.exam=
ple.org/entries/=E2=80=9C with relative references to =E2=80=9Ca.html=23a=
pple=E2=80=9D and =E2=80=9Ca.html=23anodyne=E2=80=9D will be better than =
xml:base=3D=E2=80=9Chttp://dictionary.example.org/entries/a.html=E2=80=9D=
 with relative referencs to =E2=80=9C=23apple=E2=80=9D and =E2=80=9C=23an=
odyne=E2=80=9D.=C2=A0 These will never be taken to be same-document refer=
ences.

Note, however, that while =E2=80=9C=23fragment=E2=80=9D will always be a =
same-document reference, it will ALSO always be a reference to the given =
fragment in the resource identified by the base URI.=C2=A0 If that=E2=80=99=
s not logically the same as the resource within which the reference occur=
s, you=E2=80=99re playing with fire:=C2=A0 R=46C 3986 says =E2=80=9C=23fr=
agment=E2=80=9D =E2=80=9Cshould=E2=80=9D be dereferenced without a new re=
trieval, not that it must be.=C2=A0 Any software will be perfectly within=
 its rights to retrieve the base URI and look for the fragment there.
=C2=A0
>
> References:
> 1. http://w3future.com/weblog/2005/01/13.xml=23stillBugsInTheImplementa=
tionOfHtmlHyperlinks
> 2. http://w3future.com/weblog/2005/08/14.xml=23howToUseBaseUris


********************************************
C. M. Sperberg-McQueen
Black Mesa Technologies LLC
cmsmcq=40blackmesatech.com
http://www.blackmesatech.com
********************************************

------------------------------

Date:=C2=A0=C2=A0=C2=A0 Thu, 4 May 2017 14:42:11 -0600
=46rom:=C2=A0=C2=A0=C2=A0 =22C. M. Sperberg-McQueen=22 <cmsmcq=40BLACKMES=
ATECH.COM>
Subject: Re: =40xml:base with =40rendition (and maybe other pointers)

> On May 4, 2017, at 1:31 PM, John P. McCaskey <mailbox=40JOHNMCCASKEY.CO=
M> wrote:
>
> What is the short bottom-line guidance for someone trying to encode a d=
ocument=3F Is it this=3F
>
> To point inside a TEI document, as for =40rendition, use pointers of th=
e form =23destination. Any later changes to xml:base values will not mess=
 up your internal =23 pointers.
>
> Don=E2=80=99t try to point outside your TEI document by using xml:base =
plus a =23destination in the pointer attribute. No xml:base attribute wil=
l be prepended to a pointer that begins with a =23.

No.=C2=A0=C2=A0 Sorry.=C2=A0 That=E2=80=99s not the way it works, and I t=
hink this discussion has already demonstrated that it=E2=80=99s a dangero=
us way to describe the behavior.=C2=A0

If you want to keep things simple for the encoders, I=E2=80=99d rephrase =
this as something like:=C2=A0 Don=E2=80=99t try to point outside your TEI=
 document by using xml:base plus a =23destination in the pointer attribut=
e.=C2=A0 It does not have the desired meaning.=C2=A0 Point outside your T=
EI document either using an absolute URI or an xml:base attribute plus th=
e final part of the path (the =E2=80=98file name=E2=80=99 part) and the f=
ragment identifier.=C2=A0 So NOT <ptr xml:base=3D=E2=80=9Clib/foo.xml=E2=80=
=9D target=3D=E2=80=9C=23bar=E2=80=9D/> but <ptr xml:base=3D=E2=80=9Clib/=
=E2=80=9D target=3D=E2=80=9Cfoo.xml=23bar=E2=80=9D/>.


********************************************
C. M. Sperberg-McQueen
Black Mesa Technologies LLC
cmsmcq=40blackmesatech.com
http://www.blackmesatech.com
********************************************

------------------------------

Date:=C2=A0=C2=A0=C2=A0 Thu, 4 May 2017 20:44:29 +0000
=46rom:=C2=A0=C2=A0=C2=A0 Martin Mueller <martinmueller=40NORTHWESTERN.ED=
U>
Subject: Re: =40xml:base with =40rendition (and maybe other pointers)

I=E2=80=99m following this thread from a distance (much of it is way over=
 my head), but it reminds me of something I=E2=80=99ve said before. Every=
 few months or so there is an extended discussion on the TEI list that ra=
ises a tough issue.=C2=A0 It would be helpful to the members at large if =
the Council assumed the responsibility for digesting such a discussion in=
to a one or two-page summary.=C2=A0 There might a special place on the si=
te for such papers. Sometimes these discussion reach agreement, sometimes=
 they just help clarify positions.

The obvious objection to my suggestion is that the members of the Council=
 have other and more pressing things to do. On the other hand, documentat=
ion is clearly a responsibility of the Council, and position papers of th=
is type are a kind of documentation. Or one could say =E2=80=9CWhy don=E2=
=80=99t the readers of the list take or make the time to read the extende=
d discussion=3F=E2=80=9D A good question, but there is a very high time c=
ost involved in tracking a discussion that has half a dozen participants =
and dozens of entries.

I can=E2=80=99t be the only follower of this list who wouldn=E2=80=99t be=
 grateful for a succinct account of =E2=80=9Cwhat was all this about=3F=E2=
=80=9D

On 5/4/17, 3:32 PM, =22TEI (Text Encoding Initiative) public discussion l=
ist on behalf of C. M. Sperberg-McQueen=22 <TEI-L=40LISTSERV.BROWN.EDU on=
 behalf of cmsmcq=40BLACKMESATECH.COM> wrote:

=C2=A0=C2=A0=C2=A0 > On May 4, 2017, at 12:28 PM, Hugh Cayless <philomous=
os=40GMAIL.COM> wrote:
=C2=A0=C2=A0=C2=A0 >
=C2=A0=C2=A0=C2=A0 > That's a rather favorable interpretation on your par=
t. One person agrees with you without elucidating,
=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0 I=E2=80=99m not sure this is true.=C2=A0 What Eliot Ki=
mber said is that in the context given, =E2=80=98=23apple=E2=80=99 identi=
fies the same thing as =E2=80=98https://urldefense.proofpoint.com/v2/url=3F=
u=3Dhttp-3A=5F=5Fwww.dictionary.com=5Fa.html-23apple&d=3DDwI=46aQ&c=3DyHl=
S04HhBraes5BQ9ueu5zKhE7rtNXt=5Fd012z2PA6ws&r=3DrG8zxOdssqSzDRz4x1GLlmLOW6=
0xyVXydxwnJZpkxbk&m=3DbLi7T2xrOg4vV=46Rr7hE1SVQa7jmldPgLzCrevoRJWI4&s=3DO=
lhWZ8ch2cUtq8qgCtAvoznqSamkaD9OBrjp2Js-eJs&e=3D '.=C2=A0 Does that distin=
guish between the two (=3F are there just two=3F) =E2=80=94 does that dis=
tinguish among the various interpretations of the situation offered so fa=
r=3F
=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0 I don=E2=80=99t think so.=C2=A0 I think the disagreeme=
nt we have is not over the statement=C2=A0 affirmed by Eliot Kimber, but =
over the following two claims:
=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 C1 In the context described, =E2=80=98=23a=
pple=E2=80=99 is a same-document reference and
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 can therefore by definition be dereference=
d without a new retrieval action.
=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 C2 In the context described, =E2=80=98=23a=
pple=E2=80=99 does not refer to the element in the
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 current document with xml:id=3D=E2=80=9Cap=
ple=E2=80=9D (if any); it=C2=A0 cannot be dereferenced
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 without a new retrieval action.
=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0 I intend C1 as a representation of the interpretation =
of 3986 I=E2=80=99ve been offering, and C2 as a representation of the int=
erpretation offered by John McCaskey. (SRCBS and XVAI, in my note of earl=
ier today.
=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0 > one says this discussion has jumped the shark (which=
 is fair), and the third (Michael Kay) gives a fuller answer which adds u=
p to =22it depends=22. Michael Kay is quite correct that in the context w=
here a document retrieval is *expected* to occur, the URI would indeed be=
 computed with reference to its base and fetched.
=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0 The URI is *always* computed with reference to its bas=
e.=C2=A0
=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0 Optimizations which produce the same result are, of co=
urse, allowed.=C2=A0 The preceding paragraph is a claim about the meaning=
 of certain language constructs, not a claim about what the CPU and netwo=
rk controller do during evaluation of an expression by a conforming proce=
ssor.
=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0 If it is then determined to be a same-document URI, th=
e resource identified by that URI is then =E2=80=9Cdefined to be within=E2=
=80=9D the current document; in consequence no new retrieval is necessary=
 and a new retrieval should be avoided.=C2=A0 The 'should' here means tha=
t 3986 recommends that new retrievals be avoided, but does not forbid new=
 retrievals; if conforming processors or specs have good reason for launc=
hing new retrievals, that's not a violation of the rules of 3986.=C2=A0 T=
he XSLT spec defines the document() function as always launching a new re=
trieval.=C2=A0 (Note that this does not amount to any claim by the XSLT s=
pec that the relevant resource is not within the current document.)
=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0 >
=C2=A0=C2=A0=C2=A0 > The thing is, I'm not aware of any TEI attributes or=
 element/attribute combinations which are defined as *forcing* a retrieva=
l action. I'd be happy to be corrected if I'm missing any, of course.
=C2=A0=C2=A0=C2=A0 >
=C2=A0=C2=A0=C2=A0 > It's fair to ask not just how one might expect them =
to behave, but what same-document references *mean* in the context of TEI=
 documents with =40xml:base. I agree this is something we ought to make c=
lear. I think there is some possibility of wiggle room, given that TEI ha=
s its own media type. But I also think that we'd be better off adhering t=
o the letter of R=46C 3986. The use of same-document references in TEI do=
cuments is ubiquitous, and I'm firmly against anything that might break t=
hem.
=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0 Is it clear what counts as breakage, here=3F
=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0 If a given interpretation of the URI specs causes some=
 URI references to break (by which I mean: to have an interpretation diff=
erent from what the encoders intended), can we be confident that a contra=
ry interpretation will not break any=3F=C2=A0 Or is it the case that one =
interpretation will break some URI references, and a different interpreta=
tion will break others=3F
=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0 When there are two possible interpreations of a given =
rule in a spec, it=E2=80=99s seldom the case that everyone interprets it =
the same way.=C2=A0 There is some risk that your choice does not lie betw=
een breaking things in TEI documents and not breaking them, but between b=
reaking those belonging to one project and breaking those belonging to an=
other project.
=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0 > =46or what it's worth, modern web browsers seem to a=
gree with your interpretation (mutatis mutandis=E2=80=94HTML base is not =
=40xml:base). As far as I can tell, probably because of a desire on the p=
art of the Mozilla developers back in the day to maintain compatibility w=
ith IE 4(=21).=5B1=5D
=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0 Can you expound=3F=C2=A0
=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0 Do you mean that in an HTML 4.01 or XHTML 1.0 document=
 at https://urldefense.proofpoint.com/v2/url=3Fu=3Dhttp-3A=5F=5Fexample.c=
om=5Fdoc.html&d=3DDwI=46aQ&c=3DyHlS04HhBraes5BQ9ueu5zKhE7rtNXt=5Fd012z2PA=
6ws&r=3DrG8zxOdssqSzDRz4x1GLlmLOW60xyVXydxwnJZpkxbk&m=3DbLi7T2xrOg4vV=46R=
r7hE1SVQa7jmldPgLzCrevoRJWI4&s=3D29y8v-TK44tIIjTyqctxB4Lf9WPymQrVdzrLIvEW=
6qI&e=3D=C2=A0 containing <base href=3D=E2=80=9Chttps://urldefense.proofp=
oint.com/v2/url=3Fu=3Dhttp-3A=5F=5Fexample.org=5Fbase.html-25E=46-25B=46-=
25BD&d=3DDwI=46aQ&c=3DyHlS04HhBraes5BQ9ueu5zKhE7rtNXt=5Fd012z2PA6ws&r=3Dr=
G8zxOdssqSzDRz4x1GLlmLOW60xyVXydxwnJZpkxbk&m=3DbLi7T2xrOg4vV=46Rr7hE1SVQa=
7jmldPgLzCrevoRJWI4&s=3DlknBDjsTtVRI870bNMIn9dPdWUvd3whJSBhji92091Q&e=3D =
=EF=BF=BD=EF=BF=BD>, a link of the form <a href=3D=E2=80=9C=23quovadis=E2=
=80=9D>where does this connect to=3F</a> will go to =E2=80=A6 where=3F=C2=
=A0
=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0 I think the two obvious things one might expect to see=
 in the address bar after traversing that link are
=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 https://urldefense.proofpoint.=
com/v2/url=3Fu=3Dhttp-3A=5F=5Fexample.com=5Fdoc.html-23quovadis&d=3DDwI=46=
aQ&c=3DyHlS04HhBraes5BQ9ueu5zKhE7rtNXt=5Fd012z2PA6ws&r=3DrG8zxOdssqSzDRz4=
x1GLlmLOW60xyVXydxwnJZpkxbk&m=3DbLi7T2xrOg4vV=46Rr7hE1SVQa7jmldPgLzCrevoR=
JWI4&s=3D1hCsliHZ=46QuB2OctN=5F76mP=46bm9rUqdBn1A2SVJccT=46c&e=3D
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 https://urldefense.proofpoint.=
com/v2/url=3Fu=3Dhttp-3A=5F=5Fexample.org=5Fbase.html-23quovaids&d=3DDwI=46=
aQ&c=3DyHlS04HhBraes5BQ9ueu5zKhE7rtNXt=5Fd012z2PA6ws&r=3DrG8zxOdssqSzDRz4=
x1GLlmLOW60xyVXydxwnJZpkxbk&m=3DbLi7T2xrOg4vV=46Rr7hE1SVQa7jmldPgLzCrevoR=
JWI4&s=3DCOMDCJ41VanHpbGX9WMOSGuIRx38Szt4ldxlhdtB8vs&e=3D
=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0 In the first case, this will happen either after a doc=
ument refresh or without a refresh (just scrolling to the place).
=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0 If a browser shows one of these addresses, does that c=
onstitute a claim that the other identifies a different resource=3F=C2=A0=
 Or does it only constitute a decision on the part of the browser about w=
hich of the two possible paths to a given resource it should take=3F
=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0 > To further complicate matters, the author of R=46C 3=
986, Roy =46ielding, has said that using =40xml:base in the way you propo=
se, i.e. to enable shorthand references rather than to set a canonical UR=
I for the current document, is abusive.=5B2=5D
=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0 Thank you for that reference.
=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0 I think the rules in 3986 make perfect sense if one as=
sumes Roy =46ielding=E2=80=99s principle that the base URI within a docum=
ent should be the base URI of the document, and that using xml:base or ht=
ml:base to shorten references which would otherwise be long is not a scen=
ario worth bending over backwards for.=C2=A0 (That is, it seems to me to =
simplify matters, rather than complicating them.)
=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0 If one doesn=E2=80=99t want to make that assumption, s=
ome relatively simple rules like the following might go some distance tow=
ards reducing the likelihood of unpleasant surprises:
=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 - Use the form =E2=80=9C=23fragment=E2=80=9D=
 only for references to locations in the current document.=C2=A0 These wi=
ll always be same-document references within the meaning of R=46C 3986.
=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0 - Use xml:base to set all but the last bit=
 of the URI, but not to set a full document URI:=C2=A0 xml:base=3D=E2=80=9C=
https://urldefense.proofpoint.com/v2/url=3Fu=3Dhttp-3A=5F=5Fdictionary.ex=
ample.org=5Fentries=5F&d=3DDwI=46aQ&c=3DyHlS04HhBraes5BQ9ueu5zKhE7rtNXt=5F=
d012z2PA6ws&r=3DrG8zxOdssqSzDRz4x1GLlmLOW60xyVXydxwnJZpkxbk&m=3DbLi7T2xrO=
g4vV=46Rr7hE1SVQa7jmldPgLzCrevoRJWI4&s=3Dui=46Zhf2Cm=46nQIHzkLrsBKwvBnPTf=
2AtYINKYtwKyWMI&e=3D =E2=80=9C with relative references to =E2=80=9Ca.htm=
l=23apple=E2=80=9D and =E2=80=9Ca.html=23anodyne=E2=80=9D will be better =
than xml:base=3D=E2=80=9Chttps://urldefense.proofpoint.com/v2/url=3Fu=3Dh=
ttp-3A=5F=5Fdictionary.example.org=5Fentries=5Fa.html&d=3DDwI=46aQ&c=3DyH=
lS04HhBraes5BQ9ueu5zKhE7rtNXt=5Fd012z2PA6ws&r=3DrG8zxOdssqSzDRz4x1GLlmLOW=
60xyVXydxwnJZpkxbk&m=3DbLi7T2xrOg4vV=46Rr7hE1SVQa7jmldPgLzCrevoRJWI4&s=3D=
pNUfEVwly=46PDAjVMWdE4ChwxmVlqytmyd6fi54Odw5A&e=3D =E2=80=9D with relativ=
e referencs to =E2=80=9C=23apple=E2=80=9D and =E2=80=9C=23anodyne=E2=80=9D=
.=C2=A0 These will never be taken to be same-document references.
=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0 Note, however, that while =E2=80=9C=23fragment=E2=80=9D=
 will always be a same-document reference, it will ALSO always be a refer=
ence to the given fragment in the resource identified by the base URI.=C2=
=A0 If that=E2=80=99s not logically the same as the resource within which=
 the reference occurs, you=E2=80=99re playing with fire:=C2=A0 R=46C 3986=
 says =E2=80=9C=23fragment=E2=80=9D =E2=80=9Cshould=E2=80=9D be dereferen=
ced without a new retrieval, not that it must be.=C2=A0 Any software will=
 be perfectly within its rights to retrieve the base URI and look for the=
 fragment there.
=C2=A0=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0 >
=C2=A0=C2=A0=C2=A0 > References:
=C2=A0=C2=A0=C2=A0 > 1. https://urldefense.proofpoint.com/v2/url=3Fu=3Dht=
tp-3A=5F=5Fw3future.com=5Fweblog=5F2005=5F01=5F13.xml-23stillBugsInTheImp=
lementationOfHtmlHyperlinks&d=3DDwI=46aQ&c=3DyHlS04HhBraes5BQ9ueu5zKhE7rt=
NXt=5Fd012z2PA6ws&r=3DrG8zxOdssqSzDRz4x1GLlmLOW60xyVXydxwnJZpkxbk&m=3DbLi=
7T2xrOg4vV=46Rr7hE1SVQa7jmldPgLzCrevoRJWI4&s=3DEOYAXmorJnWNneuoby45ZricHw=
P8Ll9=5Fub7gKR2Ck-4&e=3D
=C2=A0=C2=A0=C2=A0 > 2. https://urldefense.proofpoint.com/v2/url=3Fu=3Dht=
tp-3A=5F=5Fw3future.com=5Fweblog=5F2005=5F08=5F14.xml-23howToUseBaseUris&=
d=3DDwI=46aQ&c=3DyHlS04HhBraes5BQ9ueu5zKhE7rtNXt=5Fd012z2PA6ws&r=3DrG8zxO=
dssqSzDRz4x1GLlmLOW60xyVXydxwnJZpkxbk&m=3DbLi7T2xrOg4vV=46Rr7hE1SVQa7jmld=
PgLzCrevoRJWI4&s=3DpUcTcRc=5FmAHAy8r8nPCz8cD8nuQBDOqDvJ-37BCwJJQ&e=3D=C2=A0=

=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0
=C2=A0=C2=A0=C2=A0 ********************************************
=C2=A0=C2=A0=C2=A0 C. M. Sperberg-McQueen
=C2=A0=C2=A0=C2=A0 Black Mesa Technologies LLC
=C2=A0=C2=A0=C2=A0 cmsmcq=40blackmesatech.com
=C2=A0=C2=A0=C2=A0 https://urldefense.proofpoint.com/v2/url=3Fu=3Dhttp-3A=
=5F=5Fwww.blackmesatech.com&d=3DDwI=46aQ&c=3DyHlS04HhBraes5BQ9ueu5zKhE7rt=
NXt=5Fd012z2PA6ws&r=3DrG8zxOdssqSzDRz4x1GLlmLOW60xyVXydxwnJZpkxbk&m=3DbLi=
7T2xrOg4vV=46Rr7hE1SVQa7jmldPgLzCrevoRJWI4&s=3DceMDgmbvrnWoyJVZ8VidJIp4ud=
gcdq0tzNL8qVw=5FitM&e=3D
=C2=A0=C2=A0=C2=A0 ********************************************
=C2=A0=C2=A0=C2=A0

------------------------------

Date:=C2=A0=C2=A0=C2=A0 Thu, 4 May 2017 17:07:00 -0400
=46rom:=C2=A0=C2=A0=C2=A0 Hugh Cayless <philomousos=40GMAIL.COM>
Subject: Re: =40xml:base with =40rendition (and maybe other pointers)

On Thu, May 4, 2017 at 4:32 PM, C. M. Sperberg-McQueen <
cmsmcq=40blackmesatech.com> wrote:

>
>
> > one says this discussion has jumped the shark (which is fair), and th=
e
> third (Michael Kay) gives a fuller answer which adds up to =22it depend=
s=22.
> Michael Kay is quite correct that in the context where a document retri=
eval
> is *expected* to occur, the URI would indeed be computed with reference=
 to
> its base and fetched.
>
> The URI is *always* computed with reference to its base.
>

Right. It's the =22and fetched=22 part that's crucial. In a context where=

retrieval is forced, if the base is different from the document URI you'd=

expect to end up somewhere else.

>
> Optimizations which produce the same result are, of course, allowed.=C2=
=A0 The
> preceding paragraph is a claim about the meaning of certain language
> constructs, not a claim about what the CPU and network controller do du=
ring
> evaluation of an expression by a conforming processor.
>
> If it is then determined to be a same-document URI, the resource
> identified by that URI is then =E2=80=9Cdefined to be within=E2=80=9D t=
he current document;
> in consequence no new retrieval is necessary and a new retrieval should=
 be
> avoided.=C2=A0 The 'should' here means that 3986 recommends that new re=
trievals
> be avoided, but does not forbid new retrievals; if conforming processor=
s or
> specs have good reason for launching new retrievals, that's not a viola=
tion
> of the rules of 3986.=C2=A0 The XSLT spec defines the document() functi=
on as
> always launching a new retrieval.=C2=A0 (Note that this does not amount=
 to any
> claim by the XSLT spec that the relevant resource is not within the cur=
rent
> document.)
>
> >
> > The thing is, I'm not aware of any TEI attributes or element/attribut=
e
> combinations which are defined as *forcing* a retrieval action. I'd be
> happy to be corrected if I'm missing any, of course.
> >
> > It's fair to ask not just how one might expect them to behave, but wh=
at
> same-document references *mean* in the context of TEI documents with
> =40xml:base. I agree this is something we ought to make clear. I think =
there
> is some possibility of wiggle room, given that TEI has its own media ty=
pe.
> But I also think that we'd be better off adhering to the letter of R=46=
C
> 3986. The use of same-document references in TEI documents is ubiquitou=
s,
> and I'm firmly against anything that might break them.
>
> Is it clear what counts as breakage, here=3F
>

Well, from my selfish perspective, =22breakage=22 mainly means we've got =
to go
all over the Guidelines and add notes to the effect that '=23fragment'
pointers may behave differently if =40xml:base is set. To an extent, want=
ing
to avoid this is sheer laziness on my part. But I'm also convinced that
<tei:ref target=3D=22=23foo=22> in P5 is intended to mean the same thing =
that <ref
target=3D=22foo=22> did in P4, when the value of =40target was IDRE=46S r=
ather than
teidata.pointer+ (I believe you mentioned this earlier), and that the
introduction of =40xml:base was not intended to affect that meaning. I th=
ink
we're better off assuming that same-document references are referring to
the document that contains them.


>
> If a given interpretation of the URI specs causes some URI references t=
o
> break (by which I mean: to have an interpretation different from what t=
he
> encoders intended), can we be confident that a contrary interpretation =
will
> not break any=3F=C2=A0 Or is it the case that one interpretation will b=
reak some
> URI references, and a different interpretation will break others=3F
>
> When there are two possible interpreations of a given rule in a spec, i=
t=E2=80=99s
> seldom the case that everyone interprets it the same way.=C2=A0 There i=
s some
> risk that your choice does not lie between breaking things in TEI docum=
ents
> and not breaking them, but between breaking those belonging to one proj=
ect
> and breaking those belonging to another project.
>
> > =46or what it's worth, modern web browsers seem to agree with your
> interpretation (mutatis mutandis=E2=80=94HTML base is not =40xml:base).=
 As far as I
> can tell, probably because of a desire on the part of the Mozilla
> developers back in the day to maintain compatibility with IE 4(=21).=5B=
1=5D
>
> Can you expound=3F
>
> Do you mean that in an HTML 4.01 or XHTML 1.0 document at
> http://example.com/doc.html containing <base href=3D=E2=80=9C
> http://example.org/base.html=E2=80=9D>, a link of the form <a
> href=3D=E2=80=9C=23quovadis=E2=80=9D>where does this connect to=3F</a> =
will go to =E2=80=A6 where=3F
>

In my brief experimentation, Chrome, =46irefox, and Safari will all load
http://example.org/base.html=23quovadis in this case, rather than scroll =
to
the element with id=3D=22quovadis=22 in the current page.

>
> I think the two obvious things one might expect to see in the address b=
ar
> after traversing that link are
>
>=C2=A0=C2=A0=C2=A0=C2=A0 http://example.com/doc.html=23quovadis
>=C2=A0=C2=A0=C2=A0=C2=A0 http://example.org/base.html=23quovaids
>
> In the first case, this will happen either after a document refresh or
> without a refresh (just scrolling to the place).
>
> If a browser shows one of these addresses, does that constitute a claim=

> that the other identifies a different resource=3F=C2=A0 Or does it only=
 constitute
> a decision on the part of the browser about which of the two possible p=
aths
> to a given resource it should take=3F
>
> I think it means the browser implementers decided to favor stability ov=
er
technical correctness.


>
> > To further complicate matters, the author of R=46C 3986, Roy =46ieldi=
ng, has
> said that using =40xml:base in the way you propose, i.e. to enable shor=
thand
> references rather than to set a canonical URI for the current document,=
 is
> abusive.=5B2=5D
>
> Thank you for that reference.
>
> I think the rules in 3986 make perfect sense if one assumes Roy =46ield=
ing=E2=80=99s
> principle that the base URI within a document should be the base URI of=
 the
> document, and that using xml:base or html:base to shorten references wh=
ich
> would otherwise be long is not a scenario worth bending over backwards
> for.=C2=A0 (That is, it seems to me to simplify matters, rather than
> complicating them.)
>
> If one doesn=E2=80=99t want to make that assumption, some relatively si=
mple rules
> like the following might go some distance towards reducing the likeliho=
od
> of unpleasant surprises:
>
>=C2=A0=C2=A0 - Use the form =E2=80=9C=23fragment=E2=80=9D only for refer=
ences to locations in the
> current document.=C2=A0 These will always be same-document references w=
ithin the
> meaning of R=46C 3986.
>=C2=A0=C2=A0 - Use xml:base to set all but the last bit of the URI, but =
not to set a
> full document URI:=C2=A0 xml:base=3D=E2=80=9Chttp://dictionary.example.=
org/entries/=E2=80=9C
> with relative references to =E2=80=9Ca.html=23apple=E2=80=9D and =E2=80=
=9Ca.html=23anodyne=E2=80=9D will be
> better than xml:base=3D=E2=80=9Chttp://dictionary.example.org/entries/a=
.html=E2=80=9D with
> relative referencs to =E2=80=9C=23apple=E2=80=9D and =E2=80=9C=23anodyn=
e=E2=80=9D.=C2=A0 These will never be taken
> to be same-document references.
>
> Note, however, that while =E2=80=9C=23fragment=E2=80=9D will always be =
a same-document
> reference, it will ALSO always be a reference to the given fragment in =
the
> resource identified by the base URI.=C2=A0 If that=E2=80=99s not logica=
lly the same as
> the resource within which the reference occurs, you=E2=80=99re playing =
with fire:
> R=46C 3986 says =E2=80=9C=23fragment=E2=80=9D =E2=80=9Cshould=E2=80=9D =
be dereferenced without a new retrieval,
> not that it must be.=C2=A0 Any software will be perfectly within its ri=
ghts to
> retrieve the base URI and look for the fragment there.
>

I think this is perfectly reasonable. I believe we're justified in saying=

what the TEI expects =22=23fragment=22 to mean in the context of a TEI do=
cument,
but we can't guarantee that some piece of software that processes your
document won't make different decisions than we expect.

>
> >
> > References:
> > 1. http://w3future.com/weblog/2005/01/13.xml=23
> stillBugsInTheImplementationOfHtmlHyperlinks
> > 2. http://w3future.com/weblog/2005/08/14.xml=23howToUseBaseUris
>
>
> ********************************************
> C. M. Sperberg-McQueen
> Black Mesa Technologies LLC
> cmsmcq=40blackmesatech.com
> http://www.blackmesatech.com
> ********************************************
>
>

------------------------------

End of TEI-L Digest - 3 May 2017 to 4 May 2017 (=232017-101)
**********************************************************

--590c713e_4691fe5f_d8c
Content-Type: text/html; charset="utf-8"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

<html><head><style>body=7Bfont-family:Helvetica,Arial;font-size:13px=7D</=
style></head><body style=3D=22word-wrap: break-word; -webkit-nbsp-mode: s=
pace; -webkit-line-break: after-white-space;=22><div id=3D=22bloop=5Fcust=
omfont=22 style=3D=22font-family:Helvetica,Arial;font-size:13px; color: r=
gba(0,0,0,1.0); margin: 0px; line-height: auto;=22>Thank you Matthew and =
=46ranz,</div><div id=3D=22bloop=5Fcustomfont=22 style=3D=22font-family:H=
elvetica,Arial;font-size:13px; color: rgba(0,0,0,1.0); margin: 0px; line-=
height: auto;=22>my mistake=21 I should have read the GL more carefully</=
div><div id=3D=22bloop=5Fcustomfont=22 style=3D=22font-family:Helvetica,A=
rial;font-size:13px; color: rgba(0,0,0,1.0); margin: 0px; line-height: au=
to;=22>Thanks a lot for your explanations.</div> Best,<div>Emmanuelle<br>=
 <div id=3D=22bloop=5Fsign=5F1493987560623540992=22 class=3D=22bloop=5Fsi=
gn=22><pre class=3D=22moz-signature=22 cols=3D=2272=22><br></pre></div> <=
div class=3D=22airmail=5Fext=5Fon=22 style=3D=22color:black=22><br>De:&nb=
sp;<span style=3D=22color:black=22>MLH</span> <a href=3D=22mailto:matthew=
holford=40hotmail.com=22>&lt;matthewholford=40hotmail.com&gt;</a><br>R=C3=
=A9pondre:&nbsp;<span style=3D=22color:black=22>MLH</span> <a href=3D=22m=
ailto:matthewholford=40hotmail.com=22>&lt;matthewholford=40hotmail.com&gt=
;</a><br>Date:&nbsp;<span style=3D=22color:black=22>5 mai 2017 at 11:05:3=
1</span><br>=C3=80:&nbsp;<span style=3D=22color:black=22>(Text Encoding I=
nitiative) public discussion list TEI</span> <a href=3D=22mailto:tei-l=40=
listserv.brown.edu=22>&lt;tei-l=40listserv.brown.edu&gt;</a>, <span style=
=3D=22color:black=22>emmanuelle.morlock=40MOM.=46R</span> <a href=3D=22ma=
ilto:emmanuelle.morlock=40mom.fr=22>&lt;emmanuelle.morlock=40mom.fr&gt;</=
a><br>Sujet:&nbsp;<span style=3D=22color:black=22> Re: TEI-L Digest - 3 M=
ay 2017 to 4 May 2017 (=232017-101) <br></span></div><br> <blockquote typ=
e=3D=22cite=22 class=3D=22clean=5Fbq=22><span><div dir=3D=22ltr=22><div><=
/div><div>




<title></title>


<div id=3D=22divtagdefaultwrapper=22 style=3D=22font-size:12pt;color:=230=
00000;font-family:Calibri,Arial,Helvetica,sans-serif;=22 dir=3D=22ltr=22>=

<p>Dear Emmanuelle,<br></p>
<p>My understanding is that the two numbers in =40writtenLines
(also&nbsp;=40ruledLines), is that the first number gives the minimum
number of lines per page or column in the codex as a whole, and the
second gives the maximum number of lines per page or column in the
codex as a whole.<br></p>
<p>e.g.<br></p>
<p>columns=3D=222=22 writtenLines=3D=2220 30=22<br></p>
<p>would not mean that column a had 20 lines and column b 30 lines,
but that the codex as a whole had 2 columns per page and between 20
and 30 lines per column throughout.</p>
<p>similarly</p>
<p><span>columns=3D=221=22 writtenLines=3D=2220 30=22</span><br></p>
<p>would mean a codex in long lines with between 20 and 30 lines
per page throughout.</p>
<p>Best wishes,</p>
<p>Matthew<br></p>
<br>
<br>
<div style=3D=22color: rgb(0, 0, 0);=22>
<div>
<hr tabindex=3D=22-1=22 style=3D=22display:inline-block; width:98%=22>
<div id=3D=22x=5FdivRply=46wdMsg=22 dir=3D=22ltr=22><font style=3D=22font=
-size:11pt=22 color=3D=22=23000000=22 face=3D=22Calibri, sans-serif=22><b=
>=46rom:</b> TEI (Text
Encoding Initiative) public discussion list
&lt;TEI-L=40LISTSERV.BROWN.EDU&gt; on behalf of TEI-L automatic
digest system &lt;LISTSERV=40LISTSERV.BROWN.EDU&gt;<br>
<b>Sent:</b> 05 May 2017 04:00<br>
<b>To:</b> TEI-L=40LISTSERV.BROWN.EDU<br>
<b>Subject:</b> TEI-L Digest - 3 May 2017 to 4 May 2017
(=232017-101)</font>
<div>&nbsp;</div>
</div>
</div>
<div class=3D=22PlainText=22><font size=3D=222=22><span style=3D=22font-s=
ize:10pt;=22>There are 10 messages totaling 2938 lines in this
issue.<br>
<br>
Topics of the day:<br>
<br>
&nbsp; 1. =22May contain: Empty element=22<br>
&nbsp; 2. question about attribute =40writtenLines (metadata)<br>
&nbsp; 3. =40xml:base with =40rendition (and maybe other pointers)
(8)<br>
<br>
----------------------------------------------------------------------<br=
>

<br>
Date:&nbsp;&nbsp;&nbsp; Thu, 4 May 2017 11:28:58 +0100<br>
=46rom:&nbsp;&nbsp;&nbsp; Lou Burnard
&lt;lou.burnard=40RETIRED.OX.AC.UK&gt;<br>
Subject: Re: =22May contain: Empty element=22<br>
<br>
On 03/05/17 16:19, John P. McCaskey wrote:<br>
&gt; When the spec for an element says =E2=80=9CMay contain: Empty elemen=
t=E2=80=9C
that<br>
&gt; does not actually mean the spec=E2=80=99d element can contain any
other<br>
&gt; element as long as the contained one is empty, right=3F It
actually<br>
&gt; means =E2=80=9CMay contain: No other element,=E2=80=9C right=3F<br>
&gt;<br>
&gt; Should that be changed=3F<br>
&gt;<br>
&gt;<br>
<br>
The =22may contain=22 part is boiler plate text provided by the
stylesheet<br>
in various languages: it might be tricky to change it. The other
part is<br>
meant to imply that the element in question cannot contain
anything,<br>
i.e. it is empty. Since there is now a proposal in the works
(see<br>
<a href=3D=22https://github.com/TEIC/TEI/issues/1596=22 id=3D=22LPlnk5618=
22=22 previewremoved=3D=22true=22>https://github.com/TEIC/TEI/issues/1596=
</a>)&nbsp; to allow
for an empty content</span></font>
<div id=3D=22LPBorder=5FGT=5F14939752427340.8912108702195994=22 style=3D=22=
margin-bottom: 20px; overflow: auto; width: 100%; text-indent: 0px;=22>
<table id=3D=22LPContainer=5F14939752427290.7457787929213734=22 style=3D=22=
width: 90%; background-color: rgb(255, 255, 255); position: relative; ove=
rflow: auto; padding-top: 20px; padding-bottom: 20px; margin-top: 20px; b=
order-top: 1px dotted rgb(200, 200, 200); border-bottom: 1px dotted rgb(2=
00, 200, 200);=22 role=3D=22presentation=22 cellspacing=3D=220=22>
<tbody>
<tr style=3D=22border-spacing: 0px;=22 valign=3D=22top=22>
<td id=3D=22ImageCell=5F14939752427300.11624356572982253=22 style=3D=22wi=
dth: 250px; position: relative; display: table-cell; padding-right: 20px;=
=22 colspan=3D=221=22>
<div id=3D=22LPImageContainer=5F14939752427300.333276741717902=22 style=3D=
=22background-color: rgb(255, 255, 255); height: 250px; position: relativ=
e; margin: auto; display: table; width: 250px;=22>
<a id=3D=22LPImageAnchor=5F14939752427310.4235254913032114=22 style=3D=22=
display: table-cell; text-align: center;=22 href=3D=22https://github.com/=
TEIC/TEI/issues/1596=22 target=3D=22=5Fblank=22><img style=3D=22display: =
inline-block; max-width: 250px; max-height: 250px; height: 250px; width: =
250px; border-width: 0px; vertical-align: bottom;=22 id=3D=22LPThumbnailI=
mageID=5F14939752427310.027161477749538365=22 width=3D=22250=22 height=3D=
=22250=22 src=3D=22https://avatars2.githubusercontent.com/u/4857689=3Fv=3D=
3&amp;s=3D400=22></a></div>
</td>
<td id=3D=22TextCell=5F14939752427320.9678641718239404=22 style=3D=22vert=
ical-align: top; position: relative; padding: 0px; display: table-cell;=22=
 colspan=3D=222=22>
<div id=3D=22LPRemovePreviewContainer=5F14939752427320.9486097071502633=22=
>
</div>
<div id=3D=22LPTitle=5F14939752427320.10952548567936904=22 style=3D=22top=
: 0px; color: rgb(0, 120, 215); font-weight: 400; font-size: 21px; font-f=
amily: &quot;wf=5Fsegoe-ui=5Flight&quot;,&quot;Segoe UI Light&quot;,&quot=
;Segoe WP Light&quot;,&quot;Segoe UI&quot;,&quot;Segoe WP&quot;,Tahoma,Ar=
ial,sans-serif; line-height: 21px;=22>
<a id=3D=22LPUrlAnchor=5F14939752427320.13335782404330554=22 style=3D=22t=
ext-decoration: none;=22 href=3D=22https://github.com/TEIC/TEI/issues/159=
6=22 target=3D=22=5Fblank=22>Pure-ODD
content model elements appear to be strictly weaker than regular
languages -- intended=3F =C2=B7 Issue =231596 =C2=B7 TEIC/TEI</a></div>
<div id=3D=22LPMetadata=5F14939752427330.4057287606307245=22 style=3D=22m=
argin: 10px 0px 16px; color: rgb(102, 102, 102); font-weight: 400; font-f=
amily: &quot;wf=5Fsegoe-ui=5Fnormal&quot;,&quot;Segoe UI&quot;,&quot;Sego=
e WP&quot;,Tahoma,Arial,sans-serif; font-size: 14px; line-height: 14px;=22=
>
github.com</div>
<div id=3D=22LPDescription=5F14939752427330.8268234060783209=22 style=3D=22=
display: block; color: rgb(102, 102, 102); font-weight: 400; font-family:=
 &quot;wf=5Fsegoe-ui=5Fnormal&quot;,&quot;Segoe UI&quot;,&quot;Segoe WP&q=
uot;,Tahoma,Arial,sans-serif; font-size: 14px; line-height: 20px; max-hei=
ght: 100px; overflow: hidden;=22>
The elements used for content models in pure ODDs include, if I
understand chapter 22 correctly: Elements denoting atomic units of
the content model (anyElement, dataRef, elementRef, textNode,
val...</div>
</td>
</tr>
</tbody>
</table>
</div>
<font size=3D=222=22><span style=3D=22font-size:10pt;=22><br>
model to be represented in an ODD by using an explicit element
called<br>
=22&lt;empty/&gt;=22&nbsp;&nbsp; that text maybe needs to be revisited.
You might like to<br>
raise a Stylesheets ticket to remind someone to review it.<br>
<br>
<br>
------------------------------<br>
<br>
Date:&nbsp;&nbsp;&nbsp; Thu, 4 May 2017 12:45:53 +0200<br>
=46rom:&nbsp;&nbsp;&nbsp; Emmanuelle Morlock
&lt;emmanuelle.morlock=40MOM.=46R&gt;<br>
Subject: question about attribute =40writtenLines (metadata)<br>
<br>
Dear list,<br>
<br>
I was just wondering why the attribute&nbsp;=40writtenLines on layout
can contain either one or two numbers (to represent the number of
lines of one or two columns), but not more.<br>
<br>
My question is mainly out of curiosity, I don=E2=80=99t have any use case=
,
but it just seems slightly odd (the I may assume that in the
manuscript world it=E2=80=99s very rare to have more than two columns=E2=80=
=A6) but
what if =3F How would you do if you had more than two columns
=3F&nbsp;<br>
<br>
thanks =21<br>
Best,&nbsp;<br>
--&nbsp;<br>
Emmanuelle Morlock<br>
IE CNRS - Humanit=C3=A9s num=C3=A9riques &amp; TEI (Text Encoding
Initiative)<br>
UMR 5189 HISoMA (Histoire et Sources des Mondes antiques) -
Lyon<br>
<a href=3D=22http://www.hisoma.mom.fr/annuaire/morlock-emmanuelle=22 id=3D=
=22LPlnk779897=22 previewremoved=3D=22true=22>http://www.hisoma.mom.fr/an=
nuaire/morlock-emmanuelle</a></span></font>
<div id=3D=22LPBorder=5FGT=5F14939752994260.8257561761230685=22 style=3D=22=
margin-bottom: 20px; overflow: auto; width: 100%; text-indent: 0px;=22>
<table id=3D=22LPContainer=5F14939752994200.31993924872024293=22 style=3D=
=22width: 90%; background-color: rgb(255, 255, 255); position: relative; =
overflow: auto; padding-top: 20px; padding-bottom: 20px; margin-top: 20px=
; border-top: 1px dotted rgb(200, 200, 200); border-bottom: 1px dotted rg=
b(200, 200, 200);=22 role=3D=22presentation=22 cellspacing=3D=220=22>
<tbody>
<tr style=3D=22border-spacing: 0px;=22 valign=3D=22top=22>
<td id=3D=22TextCell=5F14939752994220.07261932311049946=22 style=3D=22ver=
tical-align: top; position: relative; padding: 0px; display: table-cell;=22=
 colspan=3D=222=22>
<div id=3D=22LPRemovePreviewContainer=5F14939752994230.9296336326709831=22=
>
</div>
<div id=3D=22LPTitle=5F14939752994230.8155300743021431=22 style=3D=22top:=
 0px; color: rgb(0, 120, 215); font-weight: 400; font-size: 21px; font-fa=
mily: &quot;wf=5Fsegoe-ui=5Flight&quot;,&quot;Segoe UI Light&quot;,&quot;=
Segoe WP Light&quot;,&quot;Segoe UI&quot;,&quot;Segoe WP&quot;,Tahoma,Ari=
al,sans-serif; line-height: 21px;=22>
<a id=3D=22LPUrlAnchor=5F14939752994240.0042646812302495896=22 style=3D=22=
text-decoration: none;=22 href=3D=22http://www.hisoma.mom.fr/annuaire/mor=
lock-emmanuelle=22 target=3D=22=5Fblank=22>MORLOCK Emmanuelle =7C Laborat=
oire HISOMA</a></div>
<div id=3D=22LPMetadata=5F14939752994240.29887269272311834=22 style=3D=22=
margin: 10px 0px 16px; color: rgb(102, 102, 102); font-weight: 400; font-=
family: &quot;wf=5Fsegoe-ui=5Fnormal&quot;,&quot;Segoe UI&quot;,&quot;Seg=
oe WP&quot;,Tahoma,Arial,sans-serif; font-size: 14px; line-height: 14px;=22=
>
www.hisoma.mom.fr</div>
<div id=3D=22LPDescription=5F14939752994250.022526769284129444=22 style=3D=
=22display: block; color: rgb(102, 102, 102); font-weight: 400; font-fami=
ly: &quot;wf=5Fsegoe-ui=5Fnormal&quot;,&quot;Segoe UI&quot;,&quot;Segoe W=
P&quot;,Tahoma,Arial,sans-serif; font-size: 14px; line-height: 20px; max-=
height: 100px; overflow: hidden;=22>
Activit=C3=A9s principalesSoutien au projets de d'humanit=C3=A9s num=C3=A9=
riques
et d'=C3=A9dition de sources anciennes en XML/TEI ;Conception et
=C3=A9laboration de proc=C3=A9dures de ...</div>
</td>
</tr>
</tbody>
</table>
</div>
<font size=3D=222=22><span style=3D=22font-size:10pt;=22><br>
06 85 84 69 16<br>
=40emma=5Fmorlock<br>
<br>
------------------------------<br>
<br>
Date:&nbsp;&nbsp;&nbsp; Thu, 4 May 2017 10:16:19 -0400<br>
=46rom:&nbsp;&nbsp;&nbsp; =22John P. McCaskey=22
&lt;mailbox=40JOHNMCCASKEY.COM&gt;<br>
Subject: Re: =40xml:base with =40rendition (and maybe other
pointers)<br>
<br>
I asked about this over at XML-DEV,<br>
<a href=3D=22http://lists.xml.org/archives/xml-dev/201705/msg00008.html=22=
 id=3D=22LPlnk403392=22 previewremoved=3D=22true=22>http://lists.xml.org/=
archives/xml-dev/201705/msg00008.html</a>.<br>

<br>
Opinion there is with me and opposite the majority here.<br>
<br>
Readers there don=E2=80=99t seem to think there is anything to debate. I
got one<br>
short answer and one dismissive comment about =E2=80=9Dquestions that
simple.=E2=80=9D<br>
Someone take a look and be sure I didn=E2=80=99t word my question
unfairly.<br>
<br>
Whichever interpretation TEI adopts, sounds like it should be
documented<br>
in the Guidelines or a note somewhere.<br>
<br>
If the McCaskey/XML-DEV interpretation is adopted, the GitHub issue
I<br>
posted about =40rendition stands. If not, that issue goes away.<br>
<br>
John<br>
<br>
<br>
<br>
On 5/3/2017 8:03 PM, Hugh Cayless wrote:<br>
&gt; Those are the cards we've been dealt, yeah.<br>
&gt;<br>
&gt; Sent from my phone.<br>
&gt;<br>
&gt; On May 3, 2017, at 19:53, John P. McCaskey
&lt;mailbox=40JOHNMCCASKEY.COM<br>
&gt; &lt;<a href=3D=22mailto:mailbox=40JOHNMCCASKEY.COM=22>mailto:mailbox=
=40JOHNMCCASKEY.COM</a>&gt;&gt;
wrote:<br>
&gt;<br>
&gt;&gt; So, bottom line: A standalone fragment identifier refers
to the<br>
&gt;&gt; loaded document and all the xml:base values above it in
the hierarchy<br>
&gt;&gt; are irrelevant. Encoders cannot use xml:base to direct a
standalone<br>
&gt;&gt; =23fragment value to a location outside the loaded
document.<br>
&gt;&gt;<br>
&gt;&gt; Is that right=3F<br>
&gt;&gt;<br>
&gt;&gt; --<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; On 5/3/2017 6:20 PM, Hugh Cayless wrote:<br>
&gt;&gt;&gt; No. This clarifies the expected behavior when (e.g.)
you have<br>
&gt;&gt;&gt; =40xml:base=3D=22=23frag=22. It says nothing whatever about
&lt;p<br>
&gt;&gt;&gt; rendition=3D=22=23foo=22&gt;. I don't think these
specifications interact in<br>
&gt;&gt;&gt; the way you're positing. Quite the opposite.<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt; On Wed, May 3, 2017 at 6:13 PM, John P. McCaskey<br>
&gt;&gt;&gt; &lt;mailbox=40johnmccaskey.com &lt;<a href=3D=22mailto:mailb=
ox=40johnmccaskey.com=22>mailto:mailbox=40johnmccaskey.com</a>&gt;&gt;
wrote:<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; The special behavior of
same-document references in R=46C 3986 is<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; disavowed by W3C:<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 4.4
Interpretation of same-document references<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; R=46C
3986 defines certain relative URI references, in<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
particular the empty string and those of the form =23fragment,<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; as
same-document references. Dereferencing of same-document<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
references is handled specially. However, their use as the<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; value
of an *xml:base attribute does not* *involve<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
*dereferencing, and XML Base processors should resolve them<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; in the
usual way. In particular, xml:base=3D=22=22 does not reset<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; the
base URI to that of the containing document.<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
Note:<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Some
existing processors do treat these xml:base values as<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
resetting the base URI to that of the containing document,<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; so the
use of such values is strongly discouraged.<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; This says:<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; R=46C
3986 defines special =E2=80=9Cdereferencing=E2=80=9D of empty strings<br>=

&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; and
=23fragments. But over here in XML-land, we don=E2=80=99t do<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
=E2=80=9Cdereferencing.=E2=80=9C That=E2=80=99s not a word we use here. I=
gnore that<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; stuff
about same-document references. Just resolve empty<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
strings and =23fragments as specified in the W3C<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
Recommendation above. And those of you who did carry that<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; stuff
over from 3986 to XML-land, shame on you. You messed<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; things
up for the rest of us.<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; A note has:<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 5. The
meanings of xml:base=3D=22=22 and xml:base=3D=22=23frag=22 have<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; been
clarified;<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; This says:<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; In
this second version of this recommendation, we added<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
paragraph 4.4 specifically to get you 3986 people to stop<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
polluting our W3C with your special cases. Stop doing that.<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; No=3F<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; -- John<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; On 5/3/2017 5:09 PM, Hugh
Cayless wrote:<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; That's exactly what it
does. It's just that the behavior of<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; same-document references
is prescribed in such a way that they<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; end up resolving to the
current document regardless of the<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; value of the
=40xml:base.<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; Put another way, =40xml:base
has an influence on what the client<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; application will retrieve
when it dereferences a URI and<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; retrieves the referenced
document, but in the case of<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; same-document references
no such retrieval is expected to<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; occur=E2=80=94it's assumed the
client already has the document.<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; On Wed, May 3, 2017 at
4:29 PM, John P. McCaskey<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;
&lt;mailbox=40johnmccaskey.com &lt;<a href=3D=22mailto:mailbox=40johnmcca=
skey.com=22>mailto:mailbox=40johnmccaskey.com</a>&gt;&gt;
wrote:<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
Oh. I thought it obvious they all resolved the same.<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; I
think the same about this one. You think otherwise=3F<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
&lt;div xml:base=3D=22<a href=3D=22http://www.myteiproject.com/=22>http:/=
/www.myteiproject.com/</a>=22<br>

&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
&lt;<a href=3D=22http://www.myteiproject.com/=22>http://www.myteiproject.=
com/</a>&gt;
&lt;div xml:base=3D=22images/=22&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
&lt;div&gt; &lt;graphic url=3D=22logo.jpg=22&gt; &lt;/div&gt;
&lt;div<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
xml:base=3D=22<a href=3D=22=22></a>http://www.dictionary.com/words/<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
&lt;<a href=3D=22http://dictionary.com/words/=22>http://dictionary.com/wo=
rds/</a>&gt;=22&gt;
&lt;p xml:base=3D=22a.html=22&gt; &lt;ref<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
target=3D=22=23apple=22&gt;apple&lt;/ref&gt; &lt;/p&gt; &lt;/div&gt;
&lt;/div&gt; &lt;/div&gt;<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; I
didn=E2=80=99t think xml:base had any necessary relation to the<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
current document. I thought in XML (not HTML), it just sets<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; a
path for URIs below it.<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
--<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; On
5/3/2017 3:42 PM, Hugh Cayless wrote:<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
Well, we don't know about the third one, except that it<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
points to whatever element *in the same document* has the<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
=40xml:id =22apple=22. When an application attempts to<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
dereference it, it should make the URI in =40ref absolute<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
and check it against the base<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
(<a href=3D=22=22></a>http://www.dictionary.com/words/a.html<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
&lt;<a href=3D=22http://www.dictionary.com/words/a.html=22>http://www.dic=
tionary.com/words/a.html</a>&gt;),
discover they<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
are identical, decide it doesn't need to fetch anything,<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
and go looking in the current document for the element<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
with the id =22apple=22.<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
On Wed, May 3, 2017 at 3:27 PM, John P. McCaskey<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
&lt;mailbox=40johnmccaskey.com<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
&lt;<a href=3D=22mailto:mailbox=40johnmccaskey.com=22>mailto:mailbox=40jo=
hnmccaskey.com</a>&gt;&gt;
wrote:<br>
&gt;&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
Are people proposing that these targets do not all<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
resolve to the same<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
<a href=3D=22http://www.dictionary.com/words/a.html=23apple=22>http://www=
.dictionary.com/words/a.html=23apple</a><br>

&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
&lt;<a href=3D=22http://www.dictionary.com/words/a.html=23apple=22>http:/=
/www.dictionary.com/words/a.html=23apple</a>&gt;=3F
&lt;div<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
xml:base=3D=22<a href=3D=22http://www.dictionary.com/=22>http://www.dicti=
onary.com/</a>=22<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
&lt;<a href=3D=22http://www.dictionary.com/=22>http://www.dictionary.com/=
</a>&gt;&gt;
&lt;p xml:base=3D=22words/=22&gt;<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
&lt;ref target=3D=22a.html=23apple=22&gt;apple&lt;/ref&gt; &lt;/p&gt;
&lt;/div&gt;<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
&lt;div xml:base=3D=22<a href=3D=22http://www.dictionary.com/words/=22>ht=
tp://www.dictionary.com/words/</a>=22<br>

&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
&lt;<a href=3D=22http://www.dictionary.com/words/=22>http://www.dictionar=
y.com/words/</a>&gt;&gt;
&lt;p<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
xml:base=3D=22a.html=22&gt; &lt;ref
target=3D=22=23apple=22&gt;apple&lt;/ref&gt;<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
&lt;/p&gt; &lt;/div&gt; &lt;div<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
xml:base=3D=22<a href=3D=22http://www.dictionary.com/words/a.html=22>http=
://www.dictionary.com/words/a.html</a>=22<br>

&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
&lt;<a href=3D=22http://www.dictionary.com/words/a.html=22>http://www.dic=
tionary.com/words/a.html</a>&gt;&gt;
&lt;p&gt; &lt;ref<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
target=3D=22=23apple=22&gt;apple&lt;/ref&gt; &lt;/p&gt; &lt;/div&gt;
&lt;div<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
xml:base=3D=22<a href=3D=22http://www.dictionary.com/=22>http://www.dicti=
onary.com/</a>=22<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
&lt;<a href=3D=22http://www.dictionary.com/=22>http://www.dictionary.com/=
</a>&gt;&gt;
&lt;p&gt; &lt;ref<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
target=3D=22words/a.html=23apple=22&gt;apple&lt;/ref&gt; &lt;/p&gt;
&lt;/div&gt;--<br>
&gt;&gt;&gt;&gt;&gt;<br>
<br>
------------------------------<br>
<br>
Date:&nbsp;&nbsp;&nbsp; Thu, 4 May 2017 11:37:19 -0600<br>
=46rom:&nbsp;&nbsp;&nbsp; =22C. M. Sperberg-McQueen=22
&lt;cmsmcq=40BLACKMESATECH.COM&gt;<br>
Subject: Re: =40xml:base with =40rendition (and maybe other
pointers)<br>
<br>
&gt; On May 3, 2017, at 5:53 PM, John P. McCaskey
&lt;mailbox=40JOHNMCCASKEY.COM&gt; wrote:<br>
&gt;<br>
&gt; So, bottom line: A standalone fragment identifier refers to
the loaded document and all the xml:base values above it in the
hierarchy are irrelevant. Encoders cannot use xml:base to direct a
standalone =23fragment value to a location outside the loaded
document.<br>
&gt;<br>
&gt; Is that right=3F<br>
<br>
Yes and no.&nbsp; =46or some purposes, the ways in which the answer
is 'no'<br>
are pedantic and can be ignored; for others they seem
important.<br>
I did not respond to your summary yesterday, because objecting
to<br>
the wording you used seemed unnecessarily pedantic.&nbsp; Also,
I<br>
overlooked the second sentence, which I think is the wrong
conclusion<br>
to draw.<br>
<br>
Given the following fragment of resource <a href=3D=22http://example.org/=
eg.xml=22>http://example.org/eg.xml</a><br>
<br>
&nbsp; &lt;div xml:base=3D=22<a href=3D=22http://www.dictionary.com/a.htm=
l=22>http://www.dictionary.com/a.html</a>=22&gt;<br>

&nbsp;&nbsp;&nbsp; &lt;p&gt;<br>
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; &lt;ref
target=3D=22=23apple=22&gt;Apple&lt;/ref&gt;<br>
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; &lt;ref
target=3D=22a.html=23avocado=22&gt;Avocado&lt;/ref&gt;<br>
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; &lt;ref target=3D=22<a href=3D=22http://ww=
w.dictionary.com/a.html=23anise&quot;%3EAnise%3C/ref=22>http://www.dictio=
nary.com/a.html=23anise=22&gt;Anise&lt;/ref</a>&gt;<br>

&nbsp;&nbsp;&nbsp; &lt;/p&gt;<br>
&nbsp; &lt;/div&gt;<br>
<br>
we can consider two ways of interpreting the target
attributes.<br>
<br>
Note that the discussion below ignores some possibly salient
facts:<br>
<br>
- URIs can denote different resources at different moments.<br>
&nbsp;&nbsp;&nbsp; (The discussion assumes the URI-resource mapping
is not changing.)<br>
&nbsp;&nbsp;&nbsp;<br>
&nbsp; - A given resource can have multiple representations.&nbsp;
(The<br>
&nbsp;&nbsp;&nbsp; discussion ignores any resulting
complications.)<br>
&nbsp;&nbsp;&nbsp;<br>
&nbsp; - URIs whose path component ends in .xml and .html do
not<br>
&nbsp;&nbsp;&nbsp;&nbsp; necessarily have particular MIME types, so
there is no guarantee<br>
&nbsp;&nbsp;&nbsp;&nbsp; that a fragment identifier like =23apple
will have similar<br>
&nbsp;&nbsp;&nbsp;&nbsp; meanings.&nbsp; (The discussion assumes
that fragment identifiers<br>
&nbsp;&nbsp;&nbsp;&nbsp; point to elements assigned IDs by the HTML
'id' attribute<br>
&nbsp;&nbsp;&nbsp;&nbsp; and/or the xml:id
attribute.)&nbsp;&nbsp;&nbsp;&nbsp;<br>
<br>
Interpretation 1 ('xml:base values are irrelevant').&nbsp; Ignore
xml:base<br>
when resolving '=23apple' =5Bbut not when resolving other
relative<br>
references=5D.&nbsp; The target attributes are interpreted as
denoting<br>
<br>
&nbsp;&nbsp;&nbsp; (a) <a href=3D=22http://example.org/eg.xml=23apple=22>=
http://example.org/eg.xml=23apple</a><br>

&nbsp;&nbsp;&nbsp; (b) <a href=3D=22http://www.dictionary.com/a.html=23av=
ocado=22>http://www.dictionary.com/a.html=23avocado</a><br>

&nbsp;&nbsp;&nbsp; (c) <a href=3D=22http://www.dictionary.com/a.html=23an=
ise=22>http://www.dictionary.com/a.html=23anise</a><br>

<br>
None of these have anything to do with any of:<br>
<br>
&nbsp;&nbsp;&nbsp; (d) <a href=3D=22http://www.dictionary.com/a.html=23ap=
ple=22>http://www.dictionary.com/a.html=23apple</a><br>

&nbsp;&nbsp;&nbsp; (e) <a href=3D=22http://example.org/eg.xml=23avocado=22=
>http://example.org/eg.xml=23avocado</a><br>

&nbsp;&nbsp;&nbsp; (f) <a href=3D=22http://example.org/eg.xml=23anise=22>=
http://example.org/eg.xml=23anise</a>&nbsp;&nbsp;&nbsp;<br>

<br>
Of these, (a) is a same-document reference and R=46C 3986 says
it<br>
=22should=22 be dereferenced without a new retrieval action.&nbsp; If a
new<br>
retrieval action is nevertheless launched, the resource retrieved
is<br>
(a).<br>
<br>
=5BIt is not clear to me whether the XVAI ('xml:base values are<br>
irrelevant') interpretation takes a position on whether any of
these<br>
other than (a) are same-document references.=5D<br>
<br>
Interpretation 2 ('same-document references can be surprising',
or<br>
SRCBS).&nbsp; Resolve all relative references against the base URI
in the<br>
usual way.&nbsp; Dereference same-document references either by
looking in<br>
the same document (as recommended by R=46C 3986) or by launching a
new<br>
retrieval operation.<br>
<br>
The target attributes are resolved to the absolute forms<br>
<br>
&nbsp;&nbsp;&nbsp; (d) <a href=3D=22http://www.dictionary.com/a.html=23ap=
ple=22>http://www.dictionary.com/a.html=23apple</a><br>

&nbsp;&nbsp;&nbsp; (b) <a href=3D=22http://www.dictionary.com/a.html=23av=
ocado=22>http://www.dictionary.com/a.html=23avocado</a><br>

&nbsp;&nbsp;&nbsp; (c) <a href=3D=22http://www.dictionary.com/a.html=23an=
ise=22>http://www.dictionary.com/a.html=23anise</a><br>

<br>
All of these are same-document references, so according to R=46C
3986,<br>
they should be dereferenced without a new retrieval action.&nbsp;
If a<br>
retrieval action is nevertheless launched, it will go to URIs
(d),<br>
(b), (c) respectively, not (a), (e), (f).&nbsp; =46rom the fact that
(d),<br>
(b), and (c) can be dereferenced without a new retrieval, it
follows<br>
(as far as I can tell) that these three resources can also be
denoted<br>
by URIs (a), (e), (f).<br>
<br>
The XVAI and SRCBS interpretations agree on the following
proposition,<br>
which has important relevance for operations on the data:<br>
<br>
P1 The relative reference target=3D=22=23apple=22 can be dereferenced
by<br>
locating the element in the current document with xml:id=3D=22apple=22,
if<br>
such an element exists.<br>
<br>
=46or people whose main interest is the truth or falsity of
that<br>
proposition, then, the answer is =22yes, that's right=22 -- the effect
is<br>
the same, and all else is just pilpul.<br>
<br>
=46rom P1, it follows (I think) that<br>
<br>
P2 The resource identified by target=3D=22=23apple=22 is identified by
the<br>
absolute URI <a href=3D=22http://example.org/eg.xml=23apple=22>http://exa=
mple.org/eg.xml=23apple</a>
((a) above).<br>
<br>
The two interpretations disagree, or seem to disagree, on a number
of<br>
other propositions, most obviously:<br>
<br>
P3 The relative reference '=23apple' does not identify the
resource<br>
identifed by <a href=3D=22http://www.dictionary.com/a.html=23apple=22>htt=
p://www.dictionary.com/a.html=23apple</a>.<br>

<br>
XVAI does not actually entail P2, but it is compatible with
P3.&nbsp; (To<br>
reach P3 it is necessary to assume some rule like =22No two
URIs<br>
identify the same thing=22 or =22If we don't know that a URI identifies
a<br>
thing, then it does not identify that thing.=22)&nbsp; SRCBS entails
the<br>
negation of P3.<br>
<br>
They also prescribe different URIs for the case that software<br>
determines to perform a fresh retrieval action for the
relative<br>
reference =23apple: XVAI prescribes the absolute URI (a), SRCBS<br>
prescribes URI (d).<br>
<br>
I don=E2=80=99t believe anyone has seriously suggested XVAI as the<br>
relevant rule of interpretation for examples like the one
given;<br>
what I have suggested (and I have understood Hugh Cayless<br>
to be agreeing with) is SRCBS.&nbsp; Operationally, they can
have<br>
similar results in some circumstances (specifically:&nbsp; they
can<br>
both result in no new retrieval action being undertaken in
order<br>
to dereference =E2=80=98=23apple=E2=80=99), but they differ in ways which=
 can
be<br>
critical.<br>
<br>
<br>
********************************************<br>
C. M. Sperberg-McQueen<br>
Black Mesa Technologies LLC<br>
cmsmcq=40blackmesatech.com<br>
<a href=3D=22http://www.blackmesatech.com=22>http://www.blackmesatech.com=
</a><br>

********************************************<br>
<br>
------------------------------<br>
<br>
Date:&nbsp;&nbsp;&nbsp; Thu, 4 May 2017 14:28:37 -0400<br>
=46rom:&nbsp;&nbsp;&nbsp; Hugh Cayless
&lt;philomousos=40GMAIL.COM&gt;<br>
Subject: Re: =40xml:base with =40rendition (and maybe other
pointers)<br>
<br>
That's a rather favorable interpretation on your part. One person
agrees<br>
with you without elucidating, one says this discussion has jumped
the shark<br>
(which is fair), and the third (Michael Kay) gives a fuller answer
which<br>
adds up to =22it depends=22. Michael Kay is quite correct that in the
context<br>
where a document retrieval is *expected* to occur, the URI would
indeed be<br>
computed with reference to its base and fetched.<br>
<br>
The thing is, I'm not aware of any TEI attributes or
element/attribute<br>
combinations which are defined as *forcing* a retrieval action. I'd
be<br>
happy to be corrected if I'm missing any, of course.<br>
<br>
It's fair to ask not just how one might expect them to behave, but
what<br>
same-document references *mean* in the context of TEI documents
with<br>
=40xml:base. I agree this is something we ought to make clear. I
think there<br>
is some possibility of wiggle room, given that TEI has its own
media type.<br>
But I also think that we'd be better off adhering to the letter of
R=46C<br>
3986. The use of same-document references in TEI documents is
ubiquitous,<br>
and I'm firmly against anything that might break them.<br>
<br>
=46or what it's worth, modern web browsers seem to agree with
your<br>
interpretation (mutatis mutandis=E2=80=94HTML base is not =40xml:base). A=
s
far as I<br>
can tell, probably because of a desire on the part of the
Mozilla<br>
developers back in the day to maintain compatibility with IE
4(=21).=5B1=5D<br>
<br>
To further complicate matters, the author of R=46C 3986, Roy
=46ielding, has<br>
said that using =40xml:base in the way you propose, i.e. to enable
shorthand<br>
references rather than to set a canonical URI for the current
document, is<br>
abusive.=5B2=5D<br>
<br>
Given all this, I still agree with Michael Sperberg-McQueen's
fuller<br>
explication of the issues at hand. I would interpret
same-document<br>
references as pointing to the current document, with the caveat
that there<br>
might be, now or in the future, certain pointer attributes or<br>
element/attribute combinations that mandate a retrieval action.
Such a<br>
retrieval would necessarily use whatever base was defined for the
URI in<br>
question.<br>
<br>
References:<br>
1.<br>
<a href=3D=22http://w3future.com/weblog/2005/01/13.xml=23stillBugsInTheIm=
plementationOfHtmlHyperlinks=22>
http://w3future.com/weblog/2005/01/13.xml=23stillBugsInTheImplementationO=
fHtmlHyperlinks</a><br>

2. <a href=3D=22http://w3future.com/weblog/2005/08/14.xml=23howToUseBaseU=
ris=22>http://w3future.com/weblog/2005/08/14.xml=23howToUseBaseUris</a><b=
r>

<br>
On Thu, May 4, 2017 at 10:16 AM, John P. McCaskey
&lt;mailbox=40johnmccaskey.com&gt;<br>
wrote:<br>
<br>
&gt; I asked about this over at XML-DEV, <a href=3D=22http://lists.xml.or=
g/archives/=22>http://lists.xml.org/archives/</a><br>

&gt; xml-dev/201705/msg00008.html.<br>
&gt;<br>
&gt; Opinion there is with me and opposite the majority here.<br>
&gt;<br>
&gt; Readers there don=E2=80=99t seem to think there is anything to debat=
e.
I got one<br>
&gt; short answer and one dismissive comment about =E2=80=9Dquestions tha=
t
simple.=E2=80=9D<br>
&gt; Someone take a look and be sure I didn=E2=80=99t word my question
unfairly.<br>
&gt; Whichever interpretation TEI adopts, sounds like it should be
documented<br>
&gt; in the Guidelines or a note somewhere.<br>
&gt;<br>
&gt; If the McCaskey/XML-DEV interpretation is adopted, the GitHub
issue I<br>
&gt; posted about =40rendition stands. If not, that issue goes
away.<br>
&gt;<br>
&gt; John<br>
&gt;<br>
&gt;<br>
&gt;<br>
&gt; On 5/3/2017 8:03 PM, Hugh Cayless wrote:<br>
&gt;<br>
&gt; Those are the cards we've been dealt, yeah.<br>
&gt;<br>
&gt; Sent from my phone.<br>
&gt;<br>
&gt; On May 3, 2017, at 19:53, John P. McCaskey
&lt;mailbox=40JOHNMCCASKEY.COM&gt;<br>
&gt; wrote:<br>
&gt;<br>
&gt; So, bottom line: A standalone fragment identifier refers to
the loaded<br>
&gt; document and all the xml:base values above it in the hierarchy
are<br>
&gt; irrelevant. Encoders cannot use xml:base to direct a
standalone =23fragment<br>
&gt; value to a location outside the loaded document.<br>
&gt;<br>
&gt; Is that right=3F<br>
&gt;<br>
&gt; --<br>
&gt;<br>
&gt;<br>
&gt; On 5/3/2017 6:20 PM, Hugh Cayless wrote:<br>
&gt;<br>
&gt; No. This clarifies the expected behavior when (e.g.) you have
=40xml:base=3D=22=23frag=22.<br>
&gt; It says nothing whatever about &lt;p rendition=3D=22=23foo=22&gt;. I=

don't think these<br>
&gt; specifications interact in the way you're positing. Quite the
opposite.<br>
&gt;<br>
&gt; On Wed, May 3, 2017 at 6:13 PM, John P. McCaskey
&lt;mailbox=40johnmccaskey.com<br>
&gt; &gt; wrote:<br>
&gt;<br>
&gt;&gt; The special behavior of same-document references in R=46C
3986 is disavowed<br>
&gt;&gt; by W3C:<br>
&gt;&gt;<br>
&gt;&gt; 4.4 Interpretation of same-document references<br>
&gt;&gt;<br>
&gt;&gt; R=46C 3986 defines certain relative URI references, in
particular the empty<br>
&gt;&gt; string and those of the form =23fragment, as same-document
references.<br>
&gt;&gt; Dereferencing of same-document references is handled
specially. However,<br>
&gt;&gt; their use as the value of an *xml:base attribute does not*
*involve *dereferencing,<br>
&gt;&gt; and XML Base processors should resolve them in the usual
way. In<br>
&gt;&gt; particular, xml:base=3D=22=22 does not reset the base URI to
that of the<br>
&gt;&gt; containing document.<br>
&gt;&gt;<br>
&gt;&gt; Note:<br>
&gt;&gt;<br>
&gt;&gt; Some existing processors do treat these xml:base values as
resetting the<br>
&gt;&gt; base URI to that of the containing document, so the use of
such values is<br>
&gt;&gt; strongly discouraged.<br>
&gt;&gt;<br>
&gt;&gt; This says:<br>
&gt;&gt;<br>
&gt;&gt; R=46C 3986 defines special =E2=80=9Cdereferencing=E2=80=9D of em=
pty strings
and =23fragments.<br>
&gt;&gt; But over here in XML-land, we don=E2=80=99t do =E2=80=9Cderefere=
ncing.=E2=80=9C
That=E2=80=99s not a word<br>
&gt;&gt; we use here. Ignore that stuff about same-document
references. Just resolve<br>
&gt;&gt; empty strings and =23fragments as specified in the W3C
Recommendation above.<br>
&gt;&gt; And those of you who did carry that stuff over from 3986
to XML-land, shame<br>
&gt;&gt; on you. You messed things up for the rest of us.<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; A note has:<br>
&gt;&gt;<br>
&gt;&gt; 5. The meanings of xml:base=3D=22=22 and xml:base=3D=22=23frag=22=
 have
been clarified;<br>
&gt;&gt;<br>
&gt;&gt; This says:<br>
&gt;&gt;<br>
&gt;&gt; In this second version of this recommendation, we added
paragraph 4.4<br>
&gt;&gt; specifically to get you 3986 people to stop polluting our
W3C with your<br>
&gt;&gt; special cases. Stop doing that.<br>
&gt;&gt;<br>
&gt;&gt; No=3F<br>
&gt;&gt;<br>
&gt;&gt; -- John<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt;<br>
&gt;&gt; On 5/3/2017 5:09 PM, Hugh Cayless wrote:<br>
&gt;&gt;<br>
&gt;&gt; That's exactly what it does. It's just that the behavior
of same-document<br>
&gt;&gt; references is prescribed in such a way that they end up
resolving to the<br>
&gt;&gt; current document regardless of the value of the
=40xml:base.<br>
&gt;&gt;<br>
&gt;&gt; Put another way, =40xml:base has an influence on what the
client<br>
&gt;&gt; application will retrieve when it dereferences a URI and
retrieves the<br>
&gt;&gt; referenced document, but in the case of same-document
references no such<br>
&gt;&gt; retrieval is expected to occur=E2=80=94it's assumed the client
already has the<br>
&gt;&gt; document.<br>
&gt;&gt;<br>
&gt;&gt; On Wed, May 3, 2017 at 4:29 PM, John P. McCaskey
&lt;<br>
&gt;&gt; mailbox=40johnmccaskey.com&gt; wrote:<br>
&gt;&gt;<br>
&gt;&gt;&gt; Oh. I thought it obvious they all resolved the
same.<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt; I think the same about this one. You think
otherwise=3F<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt; &lt;div xml:base=3D=22<a href=3D=22http://www.myteiproject.c=
om/=22>http://www.myteiproject.com/</a>=22
&lt;<a href=3D=22http://www.myteiproject.com/=22>http://www.myteiproject.=
com/</a>&gt;<br>

&gt;&gt;&gt;&nbsp;&nbsp; &lt;div xml:base=3D=22images/=22&gt;<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; &lt;div&gt;<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; &lt;graphic
url=3D=22logo.jpg=22&gt;<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; &lt;/div&gt;<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; &lt;div xml:base=3D=22<a href=3D=22h=
ttp://www.dictionary.com/words/=22>http://www.dictionary.com/words/</a>=22=
&gt;<br>

&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; &lt;p
xml:base=3D=22a.html=22&gt;<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
&lt;ref target=3D=22=23apple=22&gt;apple&lt;/ref&gt;<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; &lt;/p&gt;<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; &lt;/div&gt;<br>
&gt;&gt;&gt;&nbsp;&nbsp; &lt;/div&gt;<br>
&gt;&gt;&gt; &lt;/div&gt;<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt; I didn=E2=80=99t think xml:base had any necessary relation t=
o
the current<br>
&gt;&gt;&gt; document. I thought in XML (not HTML), it just sets a
path for URIs below<br>
&gt;&gt;&gt; it.<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt; --<br>
&gt;&gt;&gt; On 5/3/2017 3:42 PM, Hugh Cayless wrote:<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt; Well, we don't know about the third one, except that
it points to<br>
&gt;&gt;&gt; whatever element *in the same document* has the
=40xml:id =22apple=22. When<br>
&gt;&gt;&gt; an application attempts to dereference it, it should
make the URI in =40ref<br>
&gt;&gt;&gt; absolute and check it against the base (<a href=3D=22=22></a=
>http://www.dictionary.com/wor<br>
&gt;&gt;&gt; ds/a.html), discover they are identical, decide it
doesn't need to<br>
&gt;&gt;&gt; fetch anything, and go looking in the current document
for the element with<br>
&gt;&gt;&gt; the id =22apple=22.<br>
&gt;&gt;&gt; On Wed, May 3, 2017 at 3:27 PM, John P. McCaskey
&lt;<br>
&gt;&gt;&gt; mailbox=40johnmccaskey.com&gt; wrote:<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt; Are people proposing that these targets do not all
resolve<br>
&gt;&gt;&gt;&gt; to the same <a href=3D=22http://www.dictionary.com/words=
/a.html=23apple=3F=22>http://www.dictionary.com/words/a.html=23apple=3F</=
a><br>

&gt;&gt;&gt;&gt; &lt;div xml:base=3D=22<a href=3D=22http://www.dictionary=
.com/=22>http://www.dictionary.com/</a>=22
&lt;<a href=3D=22http://www.dictionary.com/=22>http://www.dictionary.com/=
</a>&gt;&gt;<br>

&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; &lt;p
xml:base=3D=22words/=22&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
&lt;ref target=3D=22a.html=23apple=22&gt;apple&lt;/ref&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; &lt;/p&gt;<br>
&gt;&gt;&gt;&gt; &lt;/div&gt;<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt; &lt;div xml:base=3D=22<a href=3D=22http://www.dictionary=
.com/words/=22>http://www.dictionary.com/words/</a>=22
&lt;<a href=3D=22http://www.dictionary.com/words/=22>http://www.dictionar=
y.com/words/</a>&gt;&gt;<br>

&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; &lt;p
xml:base=3D=22a.html=22&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
&lt;ref target=3D=22=23apple=22&gt;apple&lt;/ref&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; &lt;/p&gt;<br>
&gt;&gt;&gt;&gt; &lt;/div&gt;<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt; &lt;div xml:base=3D=22<a href=3D=22http://www.dictionary=
.com/words/a.html=22>http://www.dictionary.com/words/a.html</a>=22
&lt;<a href=3D=22http://www.dictionary.com/words/a.html=22>http://www.dic=
tionary.com/words/a.html</a>&gt;&gt;<br>

&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; &lt;p&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
&lt;ref target=3D=22=23apple=22&gt;apple&lt;/ref&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; &lt;/p&gt;<br>
&gt;&gt;&gt;&gt; &lt;/div&gt;<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt; &lt;div xml:base=3D=22<a href=3D=22http://www.dictionary=
.com/=22>http://www.dictionary.com/</a>=22
&lt;<a href=3D=22http://www.dictionary.com/=22>http://www.dictionary.com/=
</a>&gt;&gt;<br>

&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; &lt;p&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
&lt;ref target=3D=22words/a.html=23apple=22&gt;apple&lt;/ref&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; &lt;/p&gt;<br>
&gt;&gt;&gt;&gt; &lt;/div&gt;<br>
&gt;&gt;&gt;&gt; --<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;<br>
<br>
------------------------------<br>
<br>
Date:&nbsp;&nbsp;&nbsp; Thu, 4 May 2017 15:31:42 -0400<br>
=46rom:&nbsp;&nbsp;&nbsp; =22John P. McCaskey=22
&lt;mailbox=40JOHNMCCASKEY.COM&gt;<br>
Subject: Re: =40xml:base with =40rendition (and maybe other
pointers)<br>
<br>
What is the short bottom-line guidance for someone trying to encode
a<br>
document=3F Is it this=3F<br>
<br>
&nbsp;&nbsp;&nbsp; To point inside a TEI document, as for
=40rendition, use pointers of<br>
&nbsp;&nbsp;&nbsp; the form =23destination. Any later changes to
xml:base values will not<br>
&nbsp;&nbsp;&nbsp; mess up your internal =23 pointers.<br>
<br>
&nbsp;&nbsp;&nbsp; Don=E2=80=99t try to point outside your TEI document b=
y
using xml:base plus<br>
&nbsp;&nbsp;&nbsp; a =23destination in the pointer attribute. No
xml:base attribute will<br>
&nbsp;&nbsp;&nbsp; be prepended to a pointer that begins with a
=23.<br>
<br>
Even if I and others find that idiosyncratic and surprising,
it=E2=80=99s<br>
unambiguous, has practical benefits, requires no knowledge of R=46Cs
or<br>
W3C specs, is easy to articulate, and is easy to encode to.<br>
<br>
John<br>
<br>
<br>
<br>
On 5/4/2017 2:28 PM, Hugh Cayless wrote:<br>
&gt; That's a rather favorable interpretation on your part. One
person<br>
&gt; agrees with you without elucidating, one says this discussion
has<br>
&gt; jumped the shark (which is fair), and the third (Michael Kay)
gives a<br>
&gt; fuller answer which adds up to =22it depends=22. Michael Kay is
quite<br>
&gt; correct that in the context where a document retrieval is
*expected*<br>
&gt; to occur, the URI would indeed be computed with reference to
its base<br>
&gt; and fetched.<br>
&gt;<br>
&gt; The thing is, I'm not aware of any TEI attributes or
element/attribute<br>
&gt; combinations which are defined as *forcing* a retrieval
action. I'd be<br>
&gt; happy to be corrected if I'm missing any, of course.<br>
&gt;<br>
&gt; It's fair to ask not just how one might expect them to behave,
but<br>
&gt; what same-document references *mean* in the context of TEI
documents<br>
&gt; with =40xml:base. I agree this is something we ought to make
clear. I<br>
&gt; think there is some possibility of wiggle room, given that TEI
has its<br>
&gt; own media type. But I also think that we'd be better off
adhering to<br>
&gt; the letter of R=46C 3986. The use of same-document references in
TEI<br>
&gt; documents is ubiquitous, and I'm firmly against anything that
might<br>
&gt; break them.<br>
&gt;<br>
&gt; =46or what it's worth, modern web browsers seem to agree with
your<br>
&gt; interpretation (mutatis mutandis=E2=80=94HTML base is not =40xml:bas=
e).
As far<br>
&gt; as I can tell, probably because of a desire on the part of the
Mozilla<br>
&gt; developers back in the day to maintain compatibility with IE
4(=21).=5B1=5D<br>
&gt;<br>
&gt; To further complicate matters, the author of R=46C 3986, Roy
=46ielding,<br>
&gt; has said that using =40xml:base in the way you propose, i.e. to
enable<br>
&gt; shorthand references rather than to set a canonical URI for
the<br>
&gt; current document, is abusive.=5B2=5D<br>
&gt;<br>
&gt; Given all this, I still agree with Michael Sperberg-McQueen's
fuller<br>
&gt; explication of the issues at hand. I would interpret
same-document<br>
&gt; references as pointing to the current document, with the
caveat that<br>
&gt; there might be, now or in the future, certain pointer
attributes or<br>
&gt; element/attribute combinations that mandate a retrieval
action. Such a<br>
&gt; retrieval would necessarily use whatever base was defined for
the URI<br>
&gt; in question.<br>
&gt;<br>
&gt; References:<br>
&gt; 1.<br>
&gt; <a href=3D=22http://w3future.com/weblog/2005/01/13.xml=23stillBugsIn=
TheImplementationOfHtmlHyperlinks=22>
http://w3future.com/weblog/2005/01/13.xml=23stillBugsInTheImplementationO=
fHtmlHyperlinks</a><br>

&gt; 2. <a href=3D=22http://w3future.com/weblog/2005/08/14.xml=23howToUse=
BaseUris=22>http://w3future.com/weblog/2005/08/14.xml=23howToUseBaseUris<=
/a><br>

&gt;<br>
&gt; On Thu, May 4, 2017 at 10:16 AM, John P. McCaskey<br>
&gt; &lt;mailbox=40johnmccaskey.com &lt;<a href=3D=22mailto:mailbox=40joh=
nmccaskey.com=22>mailto:mailbox=40johnmccaskey.com</a>&gt;&gt;
wrote:<br>
&gt;<br>
&gt;&nbsp;&nbsp;&nbsp;&nbsp; I asked about this over at
XML-DEV,<br>
&gt;&nbsp;&nbsp;&nbsp;&nbsp; <a href=3D=22http://lists.xml.org/archives/x=
ml-dev/201705/msg00008.html=22>http://lists.xml.org/archives/xml-dev/2017=
05/msg00008.html</a><br>

&gt;&nbsp;&nbsp;&nbsp;&nbsp; &lt;<a href=3D=22http://lists.xml.org/archiv=
es/xml-dev/201705/msg00008.html=22>http://lists.xml.org/archives/xml-dev/=
201705/msg00008.html</a>&gt;.<br>

&gt;<br>
&gt;&nbsp;&nbsp;&nbsp;&nbsp; Opinion there is with me and opposite
the majority here.<br>
&gt;<br>
&gt;&nbsp;&nbsp;&nbsp;&nbsp; Readers there don=E2=80=99t seem to think
there is anything to debate. I<br>
&gt;&nbsp;&nbsp;&nbsp;&nbsp; got one short answer and one
dismissive comment about =E2=80=9Dquestions<br>
&gt;&nbsp;&nbsp;&nbsp;&nbsp; that simple.=E2=80=9D Someone take a look an=
d
be sure I didn=E2=80=99t word my<br>
&gt;&nbsp;&nbsp;&nbsp;&nbsp; question unfairly.<br>
&gt;<br>
&gt;&nbsp;&nbsp;&nbsp;&nbsp; Whichever interpretation TEI adopts,
sounds like it should be<br>
&gt;&nbsp;&nbsp;&nbsp;&nbsp; documented in the Guidelines or a note
somewhere.<br>
&gt;<br>
&gt;&nbsp;&nbsp;&nbsp;&nbsp; If the McCaskey/XML-DEV interpretation
is adopted, the GitHub<br>
&gt;&nbsp;&nbsp;&nbsp;&nbsp; issue I posted about =40rendition
stands. If not, that issue goes away.<br>
&gt;<br>
&gt;&nbsp;&nbsp;&nbsp;&nbsp; John<br>
&gt;<br>
&gt;<br>
&gt;<br>
&gt;&nbsp;&nbsp;&nbsp;&nbsp; On 5/3/2017 8:03 PM, Hugh Cayless
wrote:<br>
&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; Those are the cards we've been
dealt, yeah.<br>
&gt;&gt;<br>
&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; Sent from my phone.<br>
&gt;&gt;<br>
&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; On May 3, 2017, at 19:53, John P.
McCaskey<br>
&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; &lt;mailbox=40JOHNMCCASKEY.COM
&lt;<a href=3D=22mailto:mailbox=40JOHNMCCASKEY.COM=22>mailto:mailbox=40JO=
HNMCCASKEY.COM</a>&gt;&gt;
wrote:<br>
&gt;&gt;<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; So, bottom line: A standalone
fragment identifier refers to the<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; loaded document and all the
xml:base values above it in the<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; hierarchy are irrelevant.
Encoders cannot use xml:base to direct<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; a standalone =23fragment value
to a location outside the loaded<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; document.<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; Is that right=3F<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; --<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; On 5/3/2017 6:20 PM, Hugh
Cayless wrote:<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; No. This clarifies the
expected behavior when (e.g.) you have<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; =40xml:base=3D=22=23frag=22. It =
says
nothing whatever about &lt;p<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; rendition=3D=22=23foo=22&gt;. I
don't think these specifications interact<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; in the way you're
positing. Quite the opposite.<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp; On Wed, May 3, 2017 at
6:13 PM, John P. McCaskey<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;
&lt;mailbox=40johnmccaskey.com &lt;<a href=3D=22mailto:mailbox=40johnmcca=
skey.com=22>mailto:mailbox=40johnmccaskey.com</a>&gt;&gt;
wrote:<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
The special behavior of same-document references in R=46C<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
3986 is disavowed by W3C:<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;
4.4 Interpretation of same-document references<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;
R=46C 3986 defines certain relative URI references, in<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;
particular the empty string and those of the form<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;
=23fragment, as same-document references. Dereferencing<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;
of same-document references is handled specially.<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;
However, their use as the value of an *xml:base<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;
attribute does not* *involve *dereferencing, and XML<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;
Base processors should resolve them in the usual way.<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;
In particular, xml:base=3D=22=22 does not reset the base URI<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;
to that of the containing document.<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;
Note:<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;
Some existing processors do treat these xml:base values<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;
as resetting the base URI to that of the containing<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;
document, so the use of such values is strongly<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;
discouraged.<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
This says:<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;
R=46C 3986 defines special =E2=80=9Cdereferencing=E2=80=9D of empty<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;
strings and =23fragments. But over here in XML-land, we<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;
don=E2=80=99t do =E2=80=9Cdereferencing.=E2=80=9C That=E2=80=99s not a wo=
rd we use<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;
here. Ignore that stuff about same-document references.<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;
Just resolve empty strings and =23fragments as specified<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;
in the W3C Recommendation above. And those of you who<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;
did carry that stuff over from 3986 to XML-land, shame<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;
on you. You messed things up for the rest of us.<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; A
note has:<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;
5. The meanings of xml:base=3D=22=22 and xml:base=3D=22=23frag=22<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;
have been clarified;<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
This says:<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;
In this second version of this recommendation, we added<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;
paragraph 4.4 specifically to get you 3986 people to<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;
stop polluting our W3C with your special cases. Stop<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nb=
sp;&nbsp;&nbsp;
doing that.<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
No=3F<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; --
John<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; On
5/3/2017 5:09 PM, Hugh Cayless wrote:<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
That's exactly what it does. It's just that the behavior<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
of same-document references is prescribed in such a way<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
that they end up resolving to the current document<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
regardless of the value of the =40xml:base.<br>
&gt;&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
Put another way, =40xml:base has an influence on what the<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
client application will retrieve when it dereferences a<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
URI and retrieves the referenced document, but in the case<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
of same-document references no such retrieval is expected<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
to occur=E2=80=94it's assumed the client already has the document.<br>
&gt;&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
On Wed, May 3, 2017 at 4:29 PM, John P. McCaskey<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
&lt;mailbox=40johnmccaskey.com<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
&lt;<a href=3D=22mailto:mailbox=40johnmccaskey.com=22>mailto:mailbox=40jo=
hnmccaskey.com</a>&gt;&gt;
wrote:<br>
&gt;&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
Oh. I thought it obvious they all resolved the same.<br>
&gt;&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
I think the same about this one. You think otherwise=3F<br>
&gt;&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
&lt;div xml:base=3D=22<a href=3D=22http://www.myteiproject.com/=22>http:/=
/www.myteiproject.com/</a>=22<br>

&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
&lt;<a href=3D=22http://www.myteiproject.com/=22>http://www.myteiproject.=
com/</a>&gt;
&lt;div<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
xml:base=3D=22images/=22&gt; &lt;div&gt; &lt;graphic
url=3D=22logo.jpg=22&gt;<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
&lt;/div&gt; &lt;div xml:base=3D=22<a href=3D=22=22></a>http://www.dictio=
nary.com/words/<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
&lt;<a href=3D=22http://dictionary.com/words/=22>http://dictionary.com/wo=
rds/</a>&gt;=22&gt;
&lt;p xml:base=3D=22a.html=22&gt;<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
&lt;ref target=3D=22=23apple=22&gt;apple&lt;/ref&gt; &lt;/p&gt;
&lt;/div&gt; &lt;/div&gt;<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
&lt;/div&gt;<br>
&gt;&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
I didn=E2=80=99t think xml:base had any necessary relation to<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
the current document. I thought in XML (not HTML), it<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
just sets a path for URIs below it.<br>
&gt;&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
--<br>
&gt;&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;
On 5/3/2017 3:42 PM, Hugh Cayless wrote:<br>
&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;
Well, we don't know about the third one, except that<br>
&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;
it points to whatever element *in the same document*<br>
&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;
has the =40xml:id =22apple=22. When an application attempts<br>
&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;
to dereference it, it should make the URI in =40ref<br>
&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;
absolute and check it against the base<br>
&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;
(<a href=3D=22=22></a>http://www.dictionary.com/words/a.html<br>
&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;
&lt;<a href=3D=22http://www.dictionary.com/words/a.html=22>http://www.dic=
tionary.com/words/a.html</a>&gt;),
discover<br>
&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;
they are identical, decide it doesn't need to fetch<br>
&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;
anything, and go looking in the current document for<br>
&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;
the element with the id =22apple=22.<br>
&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;
On Wed, May 3, 2017 at 3:27 PM, John P. McCaskey<br>
&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;
&lt;mailbox=40johnmccaskey.com<br>
&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;
&lt;<a href=3D=22mailto:mailbox=40johnmccaskey.com=22>mailto:mailbox=40jo=
hnmccaskey.com</a>&gt;&gt;
wrote:<br>
&gt;&gt;&gt;&gt;&gt;&gt;<br>
&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
Are people proposing that these targets do not<br>
&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
all resolve to the same<br>
&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
<a href=3D=22http://www.dictionary.com/words/a.html=23apple=22>http://www=
.dictionary.com/words/a.html=23apple</a><br>

&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
&lt;<a href=3D=22http://www.dictionary.com/words/a.html=23apple=22>http:/=
/www.dictionary.com/words/a.html=23apple</a>&gt;=3F<br>

&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
&lt;div xml:base=3D=22<a href=3D=22http://www.dictionary.com/=22>http://w=
ww.dictionary.com/</a>=22<br>
&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
&lt;<a href=3D=22http://www.dictionary.com/=22>http://www.dictionary.com/=
</a>&gt;&gt;
&lt;p<br>
&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
xml:base=3D=22words/=22&gt; &lt;ref<br>
&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
target=3D=22a.html=23apple=22&gt;apple&lt;/ref&gt; &lt;/p&gt;
&lt;/div&gt;<br>
&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
&lt;div xml:base=3D=22<a href=3D=22http://www.dictionary.com/words/=22>ht=
tp://www.dictionary.com/words/</a>=22<br>

&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
&lt;<a href=3D=22http://www.dictionary.com/words/=22>http://www.dictionar=
y.com/words/</a>&gt;&gt;
&lt;p<br>
&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
xml:base=3D=22a.html=22&gt; &lt;ref<br>
&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
target=3D=22=23apple=22&gt;apple&lt;/ref&gt; &lt;/p&gt; &lt;/div&gt;
&lt;div<br>
&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
xml:base=3D=22<a href=3D=22http://www.dictionary.com/words/a.html=22>http=
://www.dictionary.com/words/a.html</a>=22<br>

&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
&lt;<a href=3D=22http://www.dictionary.com/words/a.html=22>http://www.dic=
tionary.com/words/a.html</a>&gt;&gt;
&lt;p&gt;<br>
&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
&lt;ref target=3D=22=23apple=22&gt;apple&lt;/ref&gt; &lt;/p&gt;
&lt;/div&gt; &lt;div<br>
&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
xml:base=3D=22<a href=3D=22http://www.dictionary.com/=22>http://www.dicti=
onary.com/</a>=22<br>
&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
&lt;<a href=3D=22http://www.dictionary.com/=22>http://www.dictionary.com/=
</a>&gt;&gt;
&lt;p&gt; &lt;ref<br>
&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
target=3D=22words/a.html=23apple=22&gt;apple&lt;/ref&gt; &lt;/p&gt;<br>
&gt;&gt;&gt;&gt;&gt;&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&=
nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;
&lt;/div&gt;--<br>
&gt;&gt;&gt;&gt;&gt;&gt;<br>
<br>
------------------------------<br>
<br>
Date:&nbsp;&nbsp;&nbsp; Thu, 4 May 2017 14:32:37 -0600<br>
=46rom:&nbsp;&nbsp;&nbsp; =22C. M. Sperberg-McQueen=22
&lt;cmsmcq=40BLACKMESATECH.COM&gt;<br>
Subject: Re: =40xml:base with =40rendition (and maybe other
pointers)<br>
<br>
&gt; On May 4, 2017, at 12:28 PM, Hugh Cayless
&lt;philomousos=40GMAIL.COM&gt; wrote:<br>
&gt;<br>
&gt; That's a rather favorable interpretation on your part. One
person agrees with you without elucidating,<br>
<br>
I=E2=80=99m not sure this is true.&nbsp; What Eliot Kimber said is that i=
n
the context given, =E2=80=98=23apple=E2=80=99 identifies the same thing a=
s
=E2=80=98http://www.dictionary.com/a.html=23apple'.&nbsp; Does that
distinguish between the two (=3F are there just two=3F) =E2=80=94 does th=
at
distinguish among the various interpretations of the situation
offered so far=3F<br>
<br>
I don=E2=80=99t think so.&nbsp; I think the disagreement we have is not
over the statement&nbsp; affirmed by Eliot Kimber, but over the
following two claims:<br>
<br>
&nbsp; C1 In the context described, =E2=80=98=23apple=E2=80=99 is a same-=
document
reference and<br>
&nbsp; can therefore by definition be dereferenced without a new
retrieval action.<br>
<br>
&nbsp; C2 In the context described, =E2=80=98=23apple=E2=80=99 does not r=
efer to the
element in the<br>
&nbsp; current document with xml:id=3D=E2=80=9Capple=E2=80=9D (if any); i=
t&nbsp;
cannot be dereferenced<br>
&nbsp; without a new retrieval action.<br>
<br>
I intend C1 as a representation of the interpretation of 3986 I=E2=80=99v=
e
been offering, and C2 as a representation of the interpretation
offered by John McCaskey. (SRCBS and XVAI, in my note of earlier
today.<br>
<br>
&gt; one says this discussion has jumped the shark (which is fair),
and the third (Michael Kay) gives a fuller answer which adds up to
=22it depends=22. Michael Kay is quite correct that in the context
where a document retrieval is *expected* to occur, the URI would
indeed be computed with reference to its base and fetched.<br>
<br>
The URI is *always* computed with reference to its
base.&nbsp;<br>
<br>
Optimizations which produce the same result are, of course,
allowed.&nbsp; The preceding paragraph is a claim about the meaning
of certain language constructs, not a claim about what the CPU and
network controller do during evaluation of an expression by a
conforming processor.<br>
<br>
If it is then determined to be a same-document URI, the resource
identified by that URI is then =E2=80=9Cdefined to be within=E2=80=9D the=
 current
document; in consequence no new retrieval is necessary and a new
retrieval should be avoided.&nbsp; The 'should' here means that
3986 recommends that new retrievals be avoided, but does not forbid
new retrievals; if conforming processors or specs have good reason
for launching new retrievals, that's not a violation of the rules
of 3986.&nbsp; The XSLT spec defines the document() function as
always launching a new retrieval.&nbsp; (Note that this does not
amount to any claim by the XSLT spec that the relevant resource is
not within the current document.)<br>
<br>
&gt;<br>
&gt; The thing is, I'm not aware of any TEI attributes or
element/attribute combinations which are defined as *forcing* a
retrieval action. I'd be happy to be corrected if I'm missing any,
of course.<br>
&gt;<br>
&gt; It's fair to ask not just how one might expect them to behave,
but what same-document references *mean* in the context of TEI
documents with =40xml:base. I agree this is something we ought to
make clear. I think there is some possibility of wiggle room, given
that TEI has its own media type. But I also think that we'd be
better off adhering to the letter of R=46C 3986. The use of
same-document references in TEI documents is ubiquitous, and I'm
firmly against anything that might break them.<br>
<br>
Is it clear what counts as breakage, here=3F<br>
<br>
If a given interpretation of the URI specs causes some URI
references to break (by which I mean: to have an interpretation
different from what the encoders intended), can we be confident
that a contrary interpretation will not break any=3F&nbsp; Or is it
the case that one interpretation will break some URI references,
and a different interpretation will break others=3F<br>
<br>
When there are two possible interpreations of a given rule in a
spec, it=E2=80=99s seldom the case that everyone interprets it the same
way.&nbsp; There is some risk that your choice does not lie between
breaking things in TEI documents and not breaking them, but between
breaking those belonging to one project and breaking those
belonging to another project.<br>
<br>
&gt; =46or what it's worth, modern web browsers seem to agree with
your interpretation (mutatis mutandis=E2=80=94HTML base is not =40xml:bas=
e).
As far as I can tell, probably because of a desire on the part of
the Mozilla developers back in the day to maintain compatibility
with IE 4(=21).=5B1=5D<br>
<br>
Can you expound=3F&nbsp;<br>
<br>
Do you mean that in an HTML 4.01 or XHTML 1.0 document at <a href=3D=22ht=
tp://example.com/doc.html=22>http://example.com/doc.html</a>
containing &lt;base href=3D=E2=80=9Chttp://example.org/base.html=E2=80=9D=
&gt;, a link
of the form &lt;a href=3D=E2=80=9C=23quovadis=E2=80=9D&gt;where does this=
 connect
to=3F&lt;/a&gt; will go to =E2=80=A6 where=3F&nbsp;<br>
<br>
I think the two obvious things one might expect to see in the
address bar after traversing that link are<br>
<br>
&nbsp;&nbsp;&nbsp; <a href=3D=22http://example.com/doc.html=23quovadis=22=
>http://example.com/doc.html=23quovadis</a><br>

&nbsp;&nbsp;&nbsp; <a href=3D=22http://example.org/base.html=23quovaids=22=
>http://example.org/base.html=23quovaids</a><br>

<br>
In the first case, this will happen either after a document refresh
or without a refresh (just scrolling to the place).<br>
<br>
If a browser shows one of these addresses, does that constitute a
claim that the other identifies a different resource=3F&nbsp; Or does
it only constitute a decision on the part of the browser about
which of the two possible paths to a given resource it should
take=3F<br>
<br>
<br>
&gt; To further complicate matters, the author of R=46C 3986, Roy
=46ielding, has said that using =40xml:base in the way you propose,
i.e. to enable shorthand references rather than to set a canonical
URI for the current document, is abusive.=5B2=5D<br>
<br>
Thank you for that reference.<br>
<br>
I think the rules in 3986 make perfect sense if one assumes Roy
=46ielding=E2=80=99s principle that the base URI within a document should=
 be
the base URI of the document, and that using xml:base or html:base
to shorten references which would otherwise be long is not a
scenario worth bending over backwards for.&nbsp; (That is, it seems
to me to simplify matters, rather than complicating them.)<br>
<br>
If one doesn=E2=80=99t want to make that assumption, some relatively simp=
le
rules like the following might go some distance towards reducing
the likelihood of unpleasant surprises:<br>
<br>
&nbsp; - Use the form =E2=80=9C=23fragment=E2=80=9D only for references t=
o locations
in the current document.&nbsp; These will always be same-document
references within the meaning of R=46C 3986.<br>
&nbsp; - Use xml:base to set all but the last bit of the URI, but
not to set a full document URI:&nbsp;
xml:base=3D=E2=80=9Chttp://dictionary.example.org/entries/=E2=80=9C with =
relative
references to =E2=80=9Ca.html=23apple=E2=80=9D and =E2=80=9Ca.html=23anod=
yne=E2=80=9D will be better
than xml:base=3D=E2=80=9Chttp://dictionary.example.org/entries/a.html=E2=80=
=9D with
relative referencs to =E2=80=9C=23apple=E2=80=9D and =E2=80=9C=23anodyne=E2=
=80=9D.&nbsp; These will
never be taken to be same-document references.<br>
<br>
Note, however, that while =E2=80=9C=23fragment=E2=80=9D will always be a
same-document reference, it will ALSO always be a reference to the
given fragment in the resource identified by the base URI.&nbsp; If
that=E2=80=99s not logically the same as the resource within which the
reference occurs, you=E2=80=99re playing with fire:&nbsp; R=46C 3986 says=

=E2=80=9C=23fragment=E2=80=9D =E2=80=9Cshould=E2=80=9D be dereferenced wi=
thout a new retrieval, not
that it must be.&nbsp; Any software will be perfectly within its
rights to retrieve the base URI and look for the fragment
there.<br>
&nbsp;<br>
&gt;<br>
&gt; References:<br>
&gt; 1. <a href=3D=22http://w3future.com/weblog/2005/01/13.xml=23stillBug=
sInTheImplementationOfHtmlHyperlinks=22>
http://w3future.com/weblog/2005/01/13.xml=23stillBugsInTheImplementationO=
fHtmlHyperlinks</a><br>

&gt; 2. <a href=3D=22http://w3future.com/weblog/2005/08/14.xml=23howToUse=
BaseUris=22>http://w3future.com/weblog/2005/08/14.xml=23howToUseBaseUris<=
/a><br>

<br>
<br>
********************************************<br>
C. M. Sperberg-McQueen<br>
Black Mesa Technologies LLC<br>
cmsmcq=40blackmesatech.com<br>
<a href=3D=22http://www.blackmesatech.com=22>http://www.blackmesatech.com=
</a><br>

********************************************<br>
<br>
------------------------------<br>
<br>
Date:&nbsp;&nbsp;&nbsp; Thu, 4 May 2017 14:42:11 -0600<br>
=46rom:&nbsp;&nbsp;&nbsp; =22C. M. Sperberg-McQueen=22
&lt;cmsmcq=40BLACKMESATECH.COM&gt;<br>
Subject: Re: =40xml:base with =40rendition (and maybe other
pointers)<br>
<br>
&gt; On May 4, 2017, at 1:31 PM, John P. McCaskey
&lt;mailbox=40JOHNMCCASKEY.COM&gt; wrote:<br>
&gt;<br>
&gt; What is the short bottom-line guidance for someone trying to
encode a document=3F Is it this=3F<br>
&gt;<br>
&gt; To point inside a TEI document, as for =40rendition, use
pointers of the form =23destination. Any later changes to xml:base
values will not mess up your internal =23 pointers.<br>
&gt;<br>
&gt; Don=E2=80=99t try to point outside your TEI document by using xml:ba=
se
plus a =23destination in the pointer attribute. No xml:base attribute
will be prepended to a pointer that begins with a =23.<br>
<br>
No.&nbsp;&nbsp; Sorry.&nbsp; That=E2=80=99s not the way it works, and I
think this discussion has already demonstrated that it=E2=80=99s a
dangerous way to describe the behavior.&nbsp;<br>
<br>
If you want to keep things simple for the encoders, I=E2=80=99d rephrase
this as something like:&nbsp; Don=E2=80=99t try to point outside your TEI=

document by using xml:base plus a =23destination in the pointer
attribute.&nbsp; It does not have the desired meaning.&nbsp; Point
outside your TEI document either using an absolute URI or an
xml:base attribute plus the final part of the path (the =E2=80=98file nam=
e=E2=80=99
part) and the fragment identifier.&nbsp; So NOT &lt;ptr
xml:base=3D=E2=80=9Clib/foo.xml=E2=80=9D target=3D=E2=80=9C=23bar=E2=80=9D=
/&gt; but &lt;ptr
xml:base=3D=E2=80=9Clib/=E2=80=9D target=3D=E2=80=9Cfoo.xml=23bar=E2=80=9D=
/&gt;.<br>
<br>
<br>
********************************************<br>
C. M. Sperberg-McQueen<br>
Black Mesa Technologies LLC<br>
cmsmcq=40blackmesatech.com<br>
<a href=3D=22http://www.blackmesatech.com=22>http://www.blackmesatech.com=
</a><br>

********************************************<br>
<br>
------------------------------<br>
<br>
Date:&nbsp;&nbsp;&nbsp; Thu, 4 May 2017 20:44:29 +0000<br>
=46rom:&nbsp;&nbsp;&nbsp; Martin Mueller
&lt;martinmueller=40NORTHWESTERN.EDU&gt;<br>
Subject: Re: =40xml:base with =40rendition (and maybe other
pointers)<br>
<br>
I=E2=80=99m following this thread from a distance (much of it is way over=

my head), but it reminds me of something I=E2=80=99ve said before. Every
few months or so there is an extended discussion on the TEI list
that raises a tough issue.&nbsp; It would be helpful to the members
at large if the Council assumed the responsibility for digesting
such a discussion into a one or two-page summary.&nbsp; There might
a special place on the site for such papers. Sometimes these
discussion reach agreement, sometimes they just help clarify
positions.<br>
<br>
The obvious objection to my suggestion is that the members of the
Council have other and more pressing things to do. On the other
hand, documentation is clearly a responsibility of the Council, and
position papers of this type are a kind of documentation. Or one
could say =E2=80=9CWhy don=E2=80=99t the readers of the list take or make=
 the time
to read the extended discussion=3F=E2=80=9D A good question, but there is=
 a
very high time cost involved in tracking a discussion that has half
a dozen participants and dozens of entries.<br>
<br>
I can=E2=80=99t be the only follower of this list who wouldn=E2=80=99t be=
 grateful
for a succinct account of =E2=80=9Cwhat was all this about=3F=E2=80=9D<br=
>
<br>
On 5/4/17, 3:32 PM, =22TEI (Text Encoding Initiative) public
discussion list on behalf of C. M. Sperberg-McQueen=22
&lt;TEI-L=40LISTSERV.BROWN.EDU on behalf of
cmsmcq=40BLACKMESATECH.COM&gt; wrote:<br>
<br>
&nbsp;&nbsp;&nbsp; &gt; On May 4, 2017, at 12:28 PM, Hugh Cayless
&lt;philomousos=40GMAIL.COM&gt; wrote:<br>
&nbsp;&nbsp;&nbsp; &gt;<br>
&nbsp;&nbsp;&nbsp; &gt; That's a rather favorable interpretation on
your part. One person agrees with you without elucidating,<br>
&nbsp;&nbsp;&nbsp;<br>
&nbsp;&nbsp;&nbsp; I=E2=80=99m not sure this is true.&nbsp; What Eliot
Kimber said is that in the context given, =E2=80=98=23apple=E2=80=99 iden=
tifies the
same thing as
=E2=80=98https://urldefense.proofpoint.com/v2/url=3Fu=3Dhttp-3A=5F=5Fwww.=
dictionary.com=5Fa.html-23apple&amp;d=3DDwI=46aQ&amp;c=3DyHlS04HhBraes5BQ=
9ueu5zKhE7rtNXt=5Fd012z2PA6ws&amp;r=3DrG8zxOdssqSzDRz4x1GLlmLOW60xyVXydxw=
nJZpkxbk&amp;m=3DbLi7T2xrOg4vV=46Rr7hE1SVQa7jmldPgLzCrevoRJWI4&amp;s=3DOl=
hWZ8ch2cUtq8qgCtAvoznqSamkaD9OBrjp2Js-eJs&amp;e=3D
'.&nbsp; Does that distinguish between the two (=3F are there just
two=3F) =E2=80=94 does that distinguish among the various interpretations=
 of
the situation offered so far=3F<br>
&nbsp;&nbsp;&nbsp;<br>
&nbsp;&nbsp;&nbsp; I don=E2=80=99t think so.&nbsp; I think the disagreeme=
nt
we have is not over the statement&nbsp; affirmed by Eliot Kimber,
but over the following two claims:<br>
&nbsp;&nbsp;&nbsp;<br>
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; C1 In the context described,
=E2=80=98=23apple=E2=80=99 is a same-document reference and<br>
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; can therefore by definition be
dereferenced without a new retrieval action.<br>
&nbsp;&nbsp;&nbsp;<br>
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; C2 In the context described,
=E2=80=98=23apple=E2=80=99 does not refer to the element in the<br>
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; current document with xml:id=3D=E2=80=9Cap=
ple=E2=80=9D
(if any); it&nbsp; cannot be dereferenced<br>
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; without a new retrieval
action.<br>
&nbsp;&nbsp;&nbsp;<br>
&nbsp;&nbsp;&nbsp; I intend C1 as a representation of the
interpretation of 3986 I=E2=80=99ve been offering, and C2 as a
representation of the interpretation offered by John McCaskey.
(SRCBS and XVAI, in my note of earlier today.<br>
&nbsp;&nbsp;&nbsp;<br>
&nbsp;&nbsp;&nbsp; &gt; one says this discussion has jumped the
shark (which is fair), and the third (Michael Kay) gives a fuller
answer which adds up to =22it depends=22. Michael Kay is quite correct
that in the context where a document retrieval is *expected* to
occur, the URI would indeed be computed with reference to its base
and fetched.<br>
&nbsp;&nbsp;&nbsp;<br>
&nbsp;&nbsp;&nbsp; The URI is *always* computed with reference to
its base.&nbsp;<br>
&nbsp;&nbsp;&nbsp;<br>
&nbsp;&nbsp;&nbsp; Optimizations which produce the same result are,
of course, allowed.&nbsp; The preceding paragraph is a claim about
the meaning of certain language constructs, not a claim about what
the CPU and network controller do during evaluation of an
expression by a conforming processor.<br>
&nbsp;&nbsp;&nbsp;<br>
&nbsp;&nbsp;&nbsp; If it is then determined to be a same-document
URI, the resource identified by that URI is then =E2=80=9Cdefined to be
within=E2=80=9D the current document; in consequence no new retrieval is
necessary and a new retrieval should be avoided.&nbsp; The 'should'
here means that 3986 recommends that new retrievals be avoided, but
does not forbid new retrievals; if conforming processors or specs
have good reason for launching new retrievals, that's not a
violation of the rules of 3986.&nbsp; The XSLT spec defines the
document() function as always launching a new retrieval.&nbsp;
(Note that this does not amount to any claim by the XSLT spec that
the relevant resource is not within the current document.)<br>
&nbsp;&nbsp;&nbsp;<br>
&nbsp;&nbsp;&nbsp; &gt;<br>
&nbsp;&nbsp;&nbsp; &gt; The thing is, I'm not aware of any TEI
attributes or element/attribute combinations which are defined as
*forcing* a retrieval action. I'd be happy to be corrected if I'm
missing any, of course.<br>
&nbsp;&nbsp;&nbsp; &gt;<br>
&nbsp;&nbsp;&nbsp; &gt; It's fair to ask not just how one might
expect them to behave, but what same-document references *mean* in
the context of TEI documents with =40xml:base. I agree this is
something we ought to make clear. I think there is some possibility
of wiggle room, given that TEI has its own media type. But I also
think that we'd be better off adhering to the letter of R=46C 3986.
The use of same-document references in TEI documents is ubiquitous,
and I'm firmly against anything that might break them.<br>
&nbsp;&nbsp;&nbsp;<br>
&nbsp;&nbsp;&nbsp; Is it clear what counts as breakage, here=3F<br>
&nbsp;&nbsp;&nbsp;<br>
&nbsp;&nbsp;&nbsp; If a given interpretation of the URI specs
causes some URI references to break (by which I mean: to have an
interpretation different from what the encoders intended), can we
be confident that a contrary interpretation will not break
any=3F&nbsp; Or is it the case that one interpretation will break
some URI references, and a different interpretation will break
others=3F<br>
&nbsp;&nbsp;&nbsp;<br>
&nbsp;&nbsp;&nbsp; When there are two possible interpreations of a
given rule in a spec, it=E2=80=99s seldom the case that everyone interpre=
ts
it the same way.&nbsp; There is some risk that your choice does not
lie between breaking things in TEI documents and not breaking them,
but between breaking those belonging to one project and breaking
those belonging to another project.<br>
&nbsp;&nbsp;&nbsp;<br>
&nbsp;&nbsp;&nbsp; &gt; =46or what it's worth, modern web browsers
seem to agree with your interpretation (mutatis mutandis=E2=80=94HTML bas=
e
is not =40xml:base). As far as I can tell, probably because of a
desire on the part of the Mozilla developers back in the day to
maintain compatibility with IE 4(=21).=5B1=5D<br>
&nbsp;&nbsp;&nbsp;<br>
&nbsp;&nbsp;&nbsp; Can you expound=3F&nbsp;<br>
&nbsp;&nbsp;&nbsp;<br>
&nbsp;&nbsp;&nbsp; Do you mean that in an HTML 4.01 or XHTML 1.0
document at <a href=3D=22https://urldefense.proofpoint.com/v2/url=3Fu=3Dh=
ttp-3A=5F=5Fexample.com=5Fdoc.html&amp;d=3DDwI=46aQ&amp;c=3DyHlS04HhBraes=
5BQ9ueu5zKhE7rtNXt=5Fd012z2PA6ws&amp;r=3DrG8zxOdssqSzDRz4x1GLlmLOW60xyVXy=
dxwnJZpkxbk&amp;m=3DbLi7T2xrOg4vV=46Rr7hE1SVQa7jmldPgLzCrevoRJWI4&amp;s=3D=
29y8v-TK44tIIjTyqctxB4Lf9WPymQrVdzrLIvEW6qI&amp;e=3D=22>
https://urldefense.proofpoint.com/v2/url=3Fu=3Dhttp-3A=5F=5Fexample.com=5F=
doc.html&amp;d=3DDwI=46aQ&amp;c=3DyHlS04HhBraes5BQ9ueu5zKhE7rtNXt=5Fd012z=
2PA6ws&amp;r=3DrG8zxOdssqSzDRz4x1GLlmLOW60xyVXydxwnJZpkxbk&amp;m=3DbLi7T2=
xrOg4vV=46Rr7hE1SVQa7jmldPgLzCrevoRJWI4&amp;s=3D29y8v-TK44tIIjTyqctxB4Lf9=
WPymQrVdzrLIvEW6qI&amp;e=3D</a>&nbsp;
containing &lt;base
href=3D=E2=80=9Chttps://urldefense.proofpoint.com/v2/url=3Fu=3Dhttp-3A=5F=
=5Fexample.org=5Fbase.html-25E=46-25B=46-25BD&amp;d=3DDwI=46aQ&amp;c=3DyH=
lS04HhBraes5BQ9ueu5zKhE7rtNXt=5Fd012z2PA6ws&amp;r=3DrG8zxOdssqSzDRz4x1GLl=
mLOW60xyVXydxwnJZpkxbk&amp;m=3DbLi7T2xrOg4vV=46Rr7hE1SVQa7jmldPgLzCrevoRJ=
WI4&amp;s=3DlknBDjsTtVRI870bNMIn9dPdWUvd3whJSBhji92091Q&amp;e=3D
=EF=BF=BD=EF=BF=BD&gt;, a link of the form &lt;a href=3D=E2=80=9C=23quova=
dis=E2=80=9D&gt;where does
this connect to=3F&lt;/a&gt; will go to =E2=80=A6 where=3F&nbsp;<br>
&nbsp;&nbsp;&nbsp;<br>
&nbsp;&nbsp;&nbsp; I think the two obvious things one might expect
to see in the address bar after traversing that link are<br>
&nbsp;&nbsp;&nbsp;<br>
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; <a href=3D=22https://urldefens=
e.proofpoint.com/v2/url=3Fu=3Dhttp-3A=5F=5Fexample.com=5Fdoc.html-23quova=
dis&amp;d=3DDwI=46aQ&amp;c=3DyHlS04HhBraes5BQ9ueu5zKhE7rtNXt=5Fd012z2PA6w=
s&amp;r=3DrG8zxOdssqSzDRz4x1GLlmLOW60xyVXydxwnJZpkxbk&amp;m=3DbLi7T2xrOg4=
vV=46Rr7hE1SVQa7jmldPgLzCrevoRJWI4&amp;s=3D1hCsliHZ=46QuB2OctN=5F76mP=46b=
m9rUqdBn1A2SVJccT=46c&amp;e=3D=22>
https://urldefense.proofpoint.com/v2/url=3Fu=3Dhttp-3A=5F=5Fexample.com=5F=
doc.html-23quovadis&amp;d=3DDwI=46aQ&amp;c=3DyHlS04HhBraes5BQ9ueu5zKhE7rt=
NXt=5Fd012z2PA6ws&amp;r=3DrG8zxOdssqSzDRz4x1GLlmLOW60xyVXydxwnJZpkxbk&amp=
;m=3DbLi7T2xrOg4vV=46Rr7hE1SVQa7jmldPgLzCrevoRJWI4&amp;s=3D1hCsliHZ=46QuB=
2OctN=5F76mP=46bm9rUqdBn1A2SVJccT=46c&amp;e=3D</a><br>

&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; <a href=3D=22https://urldefens=
e.proofpoint.com/v2/url=3Fu=3Dhttp-3A=5F=5Fexample.org=5Fbase.html-23quov=
aids&amp;d=3DDwI=46aQ&amp;c=3DyHlS04HhBraes5BQ9ueu5zKhE7rtNXt=5Fd012z2PA6=
ws&amp;r=3DrG8zxOdssqSzDRz4x1GLlmLOW60xyVXydxwnJZpkxbk&amp;m=3DbLi7T2xrOg=
4vV=46Rr7hE1SVQa7jmldPgLzCrevoRJWI4&amp;s=3DCOMDCJ41VanHpbGX9WMOSGuIRx38S=
zt4ldxlhdtB8vs&amp;e=3D=22>
https://urldefense.proofpoint.com/v2/url=3Fu=3Dhttp-3A=5F=5Fexample.org=5F=
base.html-23quovaids&amp;d=3DDwI=46aQ&amp;c=3DyHlS04HhBraes5BQ9ueu5zKhE7r=
tNXt=5Fd012z2PA6ws&amp;r=3DrG8zxOdssqSzDRz4x1GLlmLOW60xyVXydxwnJZpkxbk&am=
p;m=3DbLi7T2xrOg4vV=46Rr7hE1SVQa7jmldPgLzCrevoRJWI4&amp;s=3DCOMDCJ41VanHp=
bGX9WMOSGuIRx38Szt4ldxlhdtB8vs&amp;e=3D</a><br>

&nbsp;&nbsp;&nbsp;<br>
&nbsp;&nbsp;&nbsp; In the first case, this will happen either after
a document refresh or without a refresh (just scrolling to the
place).<br>
&nbsp;&nbsp;&nbsp;<br>
&nbsp;&nbsp;&nbsp; If a browser shows one of these addresses, does
that constitute a claim that the other identifies a different
resource=3F&nbsp; Or does it only constitute a decision on the part
of the browser about which of the two possible paths to a given
resource it should take=3F<br>
&nbsp;&nbsp;&nbsp;<br>
&nbsp;&nbsp;&nbsp;<br>
&nbsp;&nbsp;&nbsp; &gt; To further complicate matters, the author
of R=46C 3986, Roy =46ielding, has said that using =40xml:base in the way=

you propose, i.e. to enable shorthand references rather than to set
a canonical URI for the current document, is abusive.=5B2=5D<br>
&nbsp;&nbsp;&nbsp;<br>
&nbsp;&nbsp;&nbsp; Thank you for that reference.<br>
&nbsp;&nbsp;&nbsp;<br>
&nbsp;&nbsp;&nbsp; I think the rules in 3986 make perfect sense if
one assumes Roy =46ielding=E2=80=99s principle that the base URI within a=

document should be the base URI of the document, and that using
xml:base or html:base to shorten references which would otherwise
be long is not a scenario worth bending over backwards for.&nbsp;
(That is, it seems to me to simplify matters, rather than
complicating them.)<br>
&nbsp;&nbsp;&nbsp;<br>
&nbsp;&nbsp;&nbsp; If one doesn=E2=80=99t want to make that assumption,
some relatively simple rules like the following might go some
distance towards reducing the likelihood of unpleasant
surprises:<br>
&nbsp;&nbsp;&nbsp;<br>
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; - Use the form =E2=80=9C=23fragment=E2=80=9D=
 only for
references to locations in the current document.&nbsp; These will
always be same-document references within the meaning of R=46C
3986.<br>
&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; - Use xml:base to set all but the
last bit of the URI, but not to set a full document URI:&nbsp;
xml:base=3D=E2=80=9Chttps://urldefense.proofpoint.com/v2/url=3Fu=3Dhttp-3=
A=5F=5Fdictionary.example.org=5Fentries=5F&amp;d=3DDwI=46aQ&amp;c=3DyHlS0=
4HhBraes5BQ9ueu5zKhE7rtNXt=5Fd012z2PA6ws&amp;r=3DrG8zxOdssqSzDRz4x1GLlmLO=
W60xyVXydxwnJZpkxbk&amp;m=3DbLi7T2xrOg4vV=46Rr7hE1SVQa7jmldPgLzCrevoRJWI4=
&amp;s=3Dui=46Zhf2Cm=46nQIHzkLrsBKwvBnPTf2AtYINKYtwKyWMI&amp;e=3D
=E2=80=9C with relative references to =E2=80=9Ca.html=23apple=E2=80=9D an=
d =E2=80=9Ca.html=23anodyne=E2=80=9D
will be better than
xml:base=3D=E2=80=9Chttps://urldefense.proofpoint.com/v2/url=3Fu=3Dhttp-3=
A=5F=5Fdictionary.example.org=5Fentries=5Fa.html&amp;d=3DDwI=46aQ&amp;c=3D=
yHlS04HhBraes5BQ9ueu5zKhE7rtNXt=5Fd012z2PA6ws&amp;r=3DrG8zxOdssqSzDRz4x1G=
LlmLOW60xyVXydxwnJZpkxbk&amp;m=3DbLi7T2xrOg4vV=46Rr7hE1SVQa7jmldPgLzCrevo=
RJWI4&amp;s=3DpNUfEVwly=46PDAjVMWdE4ChwxmVlqytmyd6fi54Odw5A&amp;e=3D
=E2=80=9D with relative referencs to =E2=80=9C=23apple=E2=80=9D and =E2=80=
=9C=23anodyne=E2=80=9D.&nbsp; These
will never be taken to be same-document references.<br>
&nbsp;&nbsp;&nbsp;<br>
&nbsp;&nbsp;&nbsp; Note, however, that while =E2=80=9C=23fragment=E2=80=9D=
 will
always be a same-document reference, it will ALSO always be a
reference to the given fragment in the resource identified by the
base URI.&nbsp; If that=E2=80=99s not logically the same as the resource
within which the reference occurs, you=E2=80=99re playing with fire:&nbsp=
;
R=46C 3986 says =E2=80=9C=23fragment=E2=80=9D =E2=80=9Cshould=E2=80=9D be=
 dereferenced without a new
retrieval, not that it must be.&nbsp; Any software will be
perfectly within its rights to retrieve the base URI and look for
the fragment there.<br>
&nbsp;&nbsp;&nbsp;&nbsp;<br>
&nbsp;&nbsp;&nbsp; &gt;<br>
&nbsp;&nbsp;&nbsp; &gt; References:<br>
&nbsp;&nbsp;&nbsp; &gt; 1. <a href=3D=22https://urldefense.proofpoint.com=
/v2/url=3Fu=3Dhttp-3A=5F=5Fw3future.com=5Fweblog=5F2005=5F01=5F13.xml-23s=
tillBugsInTheImplementationOfHtmlHyperlinks&amp;d=3DDwI=46aQ&amp;c=3DyHlS=
04HhBraes5BQ9ueu5zKhE7rtNXt=5Fd012z2PA6ws&amp;r=3DrG8zxOdssqSzDRz4x1GLlmL=
OW60xyVXydxwnJZpkxbk&amp;m=3DbLi7T2xrOg4vV=46Rr7hE1SVQa7jmldPgLzCrevoRJWI=
4&amp;s=3DEOYAXmorJnWNneuoby45ZricHwP8Ll9=5Fub7gKR2Ck-4&amp;e=3D=22>
https://urldefense.proofpoint.com/v2/url=3Fu=3Dhttp-3A=5F=5Fw3future.com=5F=
weblog=5F2005=5F01=5F13.xml-23stillBugsInTheImplementationOfHtmlHyperlink=
s&amp;d=3DDwI=46aQ&amp;c=3DyHlS04HhBraes5BQ9ueu5zKhE7rtNXt=5Fd012z2PA6ws&=
amp;r=3DrG8zxOdssqSzDRz4x1GLlmLOW60xyVXydxwnJZpkxbk&amp;m=3DbLi7T2xrOg4vV=
=46Rr7hE1SVQa7jmldPgLzCrevoRJWI4&amp;s=3DEOYAXmorJnWNneuoby45ZricHwP8Ll9=5F=
ub7gKR2Ck-4&amp;e=3D</a><br>

&nbsp;&nbsp;&nbsp; &gt; 2. <a href=3D=22https://urldefense.proofpoint.com=
/v2/url=3Fu=3Dhttp-3A=5F=5Fw3future.com=5Fweblog=5F2005=5F08=5F14.xml-23h=
owToUseBaseUris&amp;d=3DDwI=46aQ&amp;c=3DyHlS04HhBraes5BQ9ueu5zKhE7rtNXt=5F=
d012z2PA6ws&amp;r=3DrG8zxOdssqSzDRz4x1GLlmLOW60xyVXydxwnJZpkxbk&amp;m=3Db=
Li7T2xrOg4vV=46Rr7hE1SVQa7jmldPgLzCrevoRJWI4&amp;s=3DpUcTcRc=5FmAHAy8r8nP=
Cz8cD8nuQBDOqDvJ-37BCwJJQ&amp;e=3D=22>
https://urldefense.proofpoint.com/v2/url=3Fu=3Dhttp-3A=5F=5Fw3future.com=5F=
weblog=5F2005=5F08=5F14.xml-23howToUseBaseUris&amp;d=3DDwI=46aQ&amp;c=3Dy=
HlS04HhBraes5BQ9ueu5zKhE7rtNXt=5Fd012z2PA6ws&amp;r=3DrG8zxOdssqSzDRz4x1GL=
lmLOW60xyVXydxwnJZpkxbk&amp;m=3DbLi7T2xrOg4vV=46Rr7hE1SVQa7jmldPgLzCrevoR=
JWI4&amp;s=3DpUcTcRc=5FmAHAy8r8nPCz8cD8nuQBDOqDvJ-37BCwJJQ&amp;e=3D</a>&n=
bsp;<br>

&nbsp;&nbsp;&nbsp;<br>
&nbsp;&nbsp;&nbsp;<br>
&nbsp;&nbsp;&nbsp;
********************************************<br>
&nbsp;&nbsp;&nbsp; C. M. Sperberg-McQueen<br>
&nbsp;&nbsp;&nbsp; Black Mesa Technologies LLC<br>
&nbsp;&nbsp;&nbsp; cmsmcq=40blackmesatech.com<br>
&nbsp;&nbsp;&nbsp; <a href=3D=22https://urldefense.proofpoint.com/v2/url=3F=
u=3Dhttp-3A=5F=5Fwww.blackmesatech.com&amp;d=3DDwI=46aQ&amp;c=3DyHlS04HhB=
raes5BQ9ueu5zKhE7rtNXt=5Fd012z2PA6ws&amp;r=3DrG8zxOdssqSzDRz4x1GLlmLOW60x=
yVXydxwnJZpkxbk&amp;m=3DbLi7T2xrOg4vV=46Rr7hE1SVQa7jmldPgLzCrevoRJWI4&amp=
;s=3DceMDgmbvrnWoyJVZ8VidJIp4udgcdq0tzNL8qVw=5FitM&amp;e=3D=22>
https://urldefense.proofpoint.com/v2/url=3Fu=3Dhttp-3A=5F=5Fwww.blackmesa=
tech.com&amp;d=3DDwI=46aQ&amp;c=3DyHlS04HhBraes5BQ9ueu5zKhE7rtNXt=5Fd012z=
2PA6ws&amp;r=3DrG8zxOdssqSzDRz4x1GLlmLOW60xyVXydxwnJZpkxbk&amp;m=3DbLi7T2=
xrOg4vV=46Rr7hE1SVQa7jmldPgLzCrevoRJWI4&amp;s=3DceMDgmbvrnWoyJVZ8VidJIp4u=
dgcdq0tzNL8qVw=5FitM&amp;e=3D</a><br>

&nbsp;&nbsp;&nbsp;
********************************************<br>
&nbsp;&nbsp;&nbsp;<br>
<br>
------------------------------<br>
<br>
Date:&nbsp;&nbsp;&nbsp; Thu, 4 May 2017 17:07:00 -0400<br>
=46rom:&nbsp;&nbsp;&nbsp; Hugh Cayless
&lt;philomousos=40GMAIL.COM&gt;<br>
Subject: Re: =40xml:base with =40rendition (and maybe other
pointers)<br>
<br>
On Thu, May 4, 2017 at 4:32 PM, C. M. Sperberg-McQueen &lt;<br>
cmsmcq=40blackmesatech.com&gt; wrote:<br>
<br>
&gt;<br>
&gt;<br>
&gt; &gt; one says this discussion has jumped the shark (which is
fair), and the<br>
&gt; third (Michael Kay) gives a fuller answer which adds up to =22it
depends=22.<br>
&gt; Michael Kay is quite correct that in the context where a
document retrieval<br>
&gt; is *expected* to occur, the URI would indeed be computed with
reference to<br>
&gt; its base and fetched.<br>
&gt;<br>
&gt; The URI is *always* computed with reference to its base.<br>
&gt;<br>
<br>
Right. It's the =22and fetched=22 part that's crucial. In a context
where<br>
retrieval is forced, if the base is different from the document URI
you'd<br>
expect to end up somewhere else.<br>
<br>
&gt;<br>
&gt; Optimizations which produce the same result are, of course,
allowed.&nbsp; The<br>
&gt; preceding paragraph is a claim about the meaning of certain
language<br>
&gt; constructs, not a claim about what the CPU and network
controller do during<br>
&gt; evaluation of an expression by a conforming processor.<br>
&gt;<br>
&gt; If it is then determined to be a same-document URI, the
resource<br>
&gt; identified by that URI is then =E2=80=9Cdefined to be within=E2=80=9D=
 the
current document;<br>
&gt; in consequence no new retrieval is necessary and a new
retrieval should be<br>
&gt; avoided.&nbsp; The 'should' here means that 3986 recommends
that new retrievals<br>
&gt; be avoided, but does not forbid new retrievals; if conforming
processors or<br>
&gt; specs have good reason for launching new retrievals, that's
not a violation<br>
&gt; of the rules of 3986.&nbsp; The XSLT spec defines the
document() function as<br>
&gt; always launching a new retrieval.&nbsp; (Note that this does
not amount to any<br>
&gt; claim by the XSLT spec that the relevant resource is not
within the current<br>
&gt; document.)<br>
&gt;<br>
&gt; &gt;<br>
&gt; &gt; The thing is, I'm not aware of any TEI attributes or
element/attribute<br>
&gt; combinations which are defined as *forcing* a retrieval
action. I'd be<br>
&gt; happy to be corrected if I'm missing any, of course.<br>
&gt; &gt;<br>
&gt; &gt; It's fair to ask not just how one might expect them to
behave, but what<br>
&gt; same-document references *mean* in the context of TEI
documents with<br>
&gt; =40xml:base. I agree this is something we ought to make clear. I
think there<br>
&gt; is some possibility of wiggle room, given that TEI has its own
media type.<br>
&gt; But I also think that we'd be better off adhering to the
letter of R=46C<br>
&gt; 3986. The use of same-document references in TEI documents is
ubiquitous,<br>
&gt; and I'm firmly against anything that might break them.<br>
&gt;<br>
&gt; Is it clear what counts as breakage, here=3F<br>
&gt;<br>
<br>
Well, from my selfish perspective, =22breakage=22 mainly means we've
got to go<br>
all over the Guidelines and add notes to the effect that
'=23fragment'<br>
pointers may behave differently if =40xml:base is set. To an extent,
wanting<br>
to avoid this is sheer laziness on my part. But I'm also convinced
that<br>
&lt;tei:ref target=3D=22=23foo=22&gt; in P5 is intended to mean the same
thing that &lt;ref<br>
target=3D=22foo=22&gt; did in P4, when the value of =40target was IDRE=46=
S
rather than<br>
teidata.pointer+ (I believe you mentioned this earlier), and that
the<br>
introduction of =40xml:base was not intended to affect that meaning.
I think<br>
we're better off assuming that same-document references are
referring to<br>
the document that contains them.<br>
<br>
<br>
&gt;<br>
&gt; If a given interpretation of the URI specs causes some URI
references to<br>
&gt; break (by which I mean: to have an interpretation different
from what the<br>
&gt; encoders intended), can we be confident that a contrary
interpretation will<br>
&gt; not break any=3F&nbsp; Or is it the case that one interpretation
will break some<br>
&gt; URI references, and a different interpretation will break
others=3F<br>
&gt;<br>
&gt; When there are two possible interpreations of a given rule in
a spec, it=E2=80=99s<br>
&gt; seldom the case that everyone interprets it the same
way.&nbsp; There is some<br>
&gt; risk that your choice does not lie between breaking things in
TEI documents<br>
&gt; and not breaking them, but between breaking those belonging to
one project<br>
&gt; and breaking those belonging to another project.<br>
&gt;<br>
&gt; &gt; =46or what it's worth, modern web browsers seem to agree
with your<br>
&gt; interpretation (mutatis mutandis=E2=80=94HTML base is not =40xml:bas=
e).
As far as I<br>
&gt; can tell, probably because of a desire on the part of the
Mozilla<br>
&gt; developers back in the day to maintain compatibility with IE
4(=21).=5B1=5D<br>
&gt;<br>
&gt; Can you expound=3F<br>
&gt;<br>
&gt; Do you mean that in an HTML 4.01 or XHTML 1.0 document
at<br>
&gt; <a href=3D=22http://example.com/doc.html=22>http://example.com/doc.h=
tml</a>
containing &lt;base href=3D=E2=80=9C<br>
&gt; <a href=3D=22http://example.org/base.html%E2%80%9D=22>http://example=
.org/base.html=E2=80=9D</a>&gt;,
a link of the form &lt;a<br>
&gt; href=3D=E2=80=9C=23quovadis=E2=80=9D&gt;where does this connect to=3F=
&lt;/a&gt; will
go to =E2=80=A6 where=3F<br>
&gt;<br>
<br>
In my brief experimentation, Chrome, =46irefox, and Safari will all
load<br>
<a href=3D=22http://example.org/base.html=23quovadis=22>http://example.or=
g/base.html=23quovadis</a>
in this case, rather than scroll to<br>
the element with id=3D=22quovadis=22 in the current page.<br>
<br>
&gt;<br>
&gt; I think the two obvious things one might expect to see in the
address bar<br>
&gt; after traversing that link are<br>
&gt;<br>
&gt;&nbsp;&nbsp;&nbsp;&nbsp; <a href=3D=22http://example.com/doc.html=23q=
uovadis=22>http://example.com/doc.html=23quovadis</a><br>

&gt;&nbsp;&nbsp;&nbsp;&nbsp; <a href=3D=22http://example.org/base.html=23=
quovaids=22>http://example.org/base.html=23quovaids</a><br>

&gt;<br>
&gt; In the first case, this will happen either after a document
refresh or<br>
&gt; without a refresh (just scrolling to the place).<br>
&gt;<br>
&gt; If a browser shows one of these addresses, does that
constitute a claim<br>
&gt; that the other identifies a different resource=3F&nbsp; Or does
it only constitute<br>
&gt; a decision on the part of the browser about which of the two
possible paths<br>
&gt; to a given resource it should take=3F<br>
&gt;<br>
&gt; I think it means the browser implementers decided to favor
stability over<br>
technical correctness.<br>
<br>
<br>
&gt;<br>
&gt; &gt; To further complicate matters, the author of R=46C 3986,
Roy =46ielding, has<br>
&gt; said that using =40xml:base in the way you propose, i.e. to
enable shorthand<br>
&gt; references rather than to set a canonical URI for the current
document, is<br>
&gt; abusive.=5B2=5D<br>
&gt;<br>
&gt; Thank you for that reference.<br>
&gt;<br>
&gt; I think the rules in 3986 make perfect sense if one assumes
Roy =46ielding=E2=80=99s<br>
&gt; principle that the base URI within a document should be the
base URI of the<br>
&gt; document, and that using xml:base or html:base to shorten
references which<br>
&gt; would otherwise be long is not a scenario worth bending over
backwards<br>
&gt; for.&nbsp; (That is, it seems to me to simplify matters,
rather than<br>
&gt; complicating them.)<br>
&gt;<br>
&gt; If one doesn=E2=80=99t want to make that assumption, some relatively=

simple rules<br>
&gt; like the following might go some distance towards reducing the
likelihood<br>
&gt; of unpleasant surprises:<br>
&gt;<br>
&gt;&nbsp;&nbsp; - Use the form =E2=80=9C=23fragment=E2=80=9D only for re=
ferences to
locations in the<br>
&gt; current document.&nbsp; These will always be same-document
references within the<br>
&gt; meaning of R=46C 3986.<br>
&gt;&nbsp;&nbsp; - Use xml:base to set all but the last bit of the
URI, but not to set a<br>
&gt; full document URI:&nbsp;
xml:base=3D=E2=80=9Chttp://dictionary.example.org/entries/=E2=80=9C<br>
&gt; with relative references to =E2=80=9Ca.html=23apple=E2=80=9D and
=E2=80=9Ca.html=23anodyne=E2=80=9D will be<br>
&gt; better than
xml:base=3D=E2=80=9Chttp://dictionary.example.org/entries/a.html=E2=80=9D=
 with<br>
&gt; relative referencs to =E2=80=9C=23apple=E2=80=9D and =E2=80=9C=23ano=
dyne=E2=80=9D.&nbsp; These
will never be taken<br>
&gt; to be same-document references.<br>
&gt;<br>
&gt; Note, however, that while =E2=80=9C=23fragment=E2=80=9D will always =
be a
same-document<br>
&gt; reference, it will ALSO always be a reference to the given
fragment in the<br>
&gt; resource identified by the base URI.&nbsp; If that=E2=80=99s not
logically the same as<br>
&gt; the resource within which the reference occurs, you=E2=80=99re playi=
ng
with fire:<br>
&gt; R=46C 3986 says =E2=80=9C=23fragment=E2=80=9D =E2=80=9Cshould=E2=80=9D=
 be dereferenced without a
new retrieval,<br>
&gt; not that it must be.&nbsp; Any software will be perfectly
within its rights to<br>
&gt; retrieve the base URI and look for the fragment there.<br>
&gt;<br>
<br>
I think this is perfectly reasonable. I believe we're justified in
saying<br>
what the TEI expects =22=23fragment=22 to mean in the context of a TEI
document,<br>
but we can't guarantee that some piece of software that processes
your<br>
document won't make different decisions than we expect.<br>
<br>
&gt;<br>
&gt; &gt;<br>
&gt; &gt; References:<br>
&gt; &gt; 1. <a href=3D=22http://w3future.com/weblog/2005/01/13.xml=23=22=
>http://w3future.com/weblog/2005/01/13.xml=23</a><br>

&gt; stillBugsInTheImplementationOfHtmlHyperlinks<br>
&gt; &gt; 2. <a href=3D=22http://w3future.com/weblog/2005/08/14.xml=23how=
ToUseBaseUris=22>http://w3future.com/weblog/2005/08/14.xml=23howToUseBase=
Uris</a><br>

&gt;<br>
&gt;<br>
&gt; ********************************************<br>
&gt; C. M. Sperberg-McQueen<br>
&gt; Black Mesa Technologies LLC<br>
&gt; cmsmcq=40blackmesatech.com<br>
&gt; <a href=3D=22http://www.blackmesatech.com=22>http://www.blackmesatec=
h.com</a><br>

&gt; ********************************************<br>
&gt;<br>
&gt;<br>
<br>
------------------------------<br>
<br>
End of TEI-L Digest - 3 May 2017 to 4 May 2017 (=232017-101)<br>
**********************************************************<br></span></fo=
nt></div>
</div>
</div>


</div></div></span></blockquote></div></body></html>
--590c713e_4691fe5f_d8c--