[urn] Re: Proposing a way forward re: GLUE and "third-pa rty" registration

--===============3159378737440926754==
Content-Type: multipart/alternative; boundary="000000000000f119ff0650b85415"

--000000000000f119ff0650b85415
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Hey Dale,

The identifiers you want to ask an agent about would be GS1 GLNs, GLEIF
LEIs, and DUNs numbers, open corporates numbers, etc.

Some of these numbers are paid for, others are assigned by processing
government registries.

It would be awesome if there were a way for new identifier issuing
authorities to be able to easily register their identifier system, in a
spec required registry, without needing to do more than comply with syntax
requirements and provide a reference to a spec.

For the use case, I think the shorter the identifier, the better. I think a
URI is better solution than a URN because of this.

What do you think?

Regards,

OS


On Thu, Apr 30, 2026, 8:58=E2=80=AFPM Dale R. Worley <[email protected]> w=
rote:

> I recently got a subscription to ChatGPT.  I've seen various reports
> that it's good for finding references to information when you don't have
> any idea where that information might be located.  So in regard to GLUE
> -- and as an attempt to make ChatGPT do something useful for me -- I
> tried to see if I could find any information about the GS1 (UPC/EAN)
> identifier system, particularly whether it has undertaken the
> commitments that would be needed to incorporate GS1 as a sub-registrant
> for URNs.
>
> [email protected] (Dale R. Worley) writes:
> > If I stretch, I can argue that there may be some instances where the
> > third-party organization has in some way committed to the public to
> > respect all the needed properties of URNs, and the registrant has
> > recorded a suitable citation of that commitment.  In such a case it
> > could be argued that no explicit consultation between the registrant an=
d
> > the third-party organization is needed.
>
> Of course, I don't *trust* what ChatGPT dug up because I didn't track it
> back to the real specification documents.  But between ChatGPT's
> excerpts and the Wikipedia pages about GS1, it seems to me that some
> serious work is needed to verify that GS1 identifiers are suitable.
>
> Specifically, identifiers are assigned to individual "products" (SKUs,
> really) by "manufacturers".  Manufacturers are assigned prefixes in the
> GS1 number-space.  So the GLUE registry should clarify that the
> identifier to be used in the URN is the manufacturer prefix of the GS1
> code.
>
> ChatGPT also said:
>
>     Are prefixes ever reused?
>
>     Yes, but with strict controls.
>
>     GS1 policy allows reuse only after a long period and under
>     conditions designed to avoid confusion.
>
> As I said, I don't *trust* this, but conversely, allowing prefixes to be
> used in URNs requires that we obtain positive verification that a
> prefix that is assigned to an organization can never be reassigned to a
> "different" organization.
>
> This effort I consider to be a start on verifying that GS1 can be
> incorporated into a URN.  Similar investigation is needed for the other
> four proposed incorporated identifiers.
>
> Dale
>

--000000000000f119ff0650b85415
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"auto">Hey Dale,<div dir=3D"auto"><br></div><div dir=3D"auto">Th=
e identifiers you want to ask an agent about would be GS1 GLNs, GLEIF LEIs,=
 and DUNs numbers, open corporates numbers, etc.</div><div dir=3D"auto"><br=
></div><div dir=3D"auto">Some of these numbers are paid for, others are ass=
igned by processing government registries.</div><div dir=3D"auto"><br></div=
><div dir=3D"auto">It would be awesome if there were a way for new identifi=
er issuing authorities to be able to easily register their identifier syste=
m, in a spec required registry, without needing to do more than comply with=
 syntax requirements and provide a reference to a spec.</div><div dir=3D"au=
to"><br></div><div dir=3D"auto">For the use case, I think the shorter the i=
dentifier, the better. I think a URI is better solution than a URN because =
of this.</div><div dir=3D"auto"><br></div><div dir=3D"auto">What do you thi=
nk?</div><div dir=3D"auto"><br></div><div dir=3D"auto">Regards,</div><div d=
ir=3D"auto"><br></div><div dir=3D"auto">OS</div><br><br><div class=3D"gmail=
_quote gmail_quote_container" dir=3D"auto"><div dir=3D"ltr" class=3D"gmail_=
attr">On Thu, Apr 30, 2026, 8:58=E2=80=AFPM Dale R. Worley &lt;<a href=3D"m=
ailto:[email protected]">[email protected]</a>&gt; wrote:<br></div><block=
quote class=3D"gmail_quote" style=3D"margin:0px 0px 0px 0.8ex;border-left:1=
px solid rgb(204,204,204);padding-left:1ex">I recently got a subscription t=
o ChatGPT.=C2=A0 I&#39;ve seen various reports<br>
that it&#39;s good for finding references to information when you don&#39;t=
 have<br>
any idea where that information might be located.=C2=A0 So in regard to GLU=
E<br>
-- and as an attempt to make ChatGPT do something useful for me -- I<br>
tried to see if I could find any information about the GS1 (UPC/EAN)<br>
identifier system, particularly whether it has undertaken the<br>
commitments that would be needed to incorporate GS1 as a sub-registrant<br>
for URNs.<br>
<br>
<a href=3D"mailto:[email protected]" target=3D"_blank" rel=3D"noreferrer">=
[email protected]</a> (Dale R. Worley) writes:<br>
&gt; If I stretch, I can argue that there may be some instances where the<b=
r>
&gt; third-party organization has in some way committed to the public to<br=
>
&gt; respect all the needed properties of URNs, and the registrant has<br>
&gt; recorded a suitable citation of that commitment.=C2=A0 In such a case =
it<br>
&gt; could be argued that no explicit consultation between the registrant a=
nd<br>
&gt; the third-party organization is needed.<br>
<br>
Of course, I don&#39;t *trust* what ChatGPT dug up because I didn&#39;t tra=
ck it<br>
back to the real specification documents.=C2=A0 But between ChatGPT&#39;s<b=
r>
excerpts and the Wikipedia pages about GS1, it seems to me that some<br>
serious work is needed to verify that GS1 identifiers are suitable.<br>
<br>
Specifically, identifiers are assigned to individual &quot;products&quot; (=
SKUs,<br>
really) by &quot;manufacturers&quot;.=C2=A0 Manufacturers are assigned pref=
ixes in the<br>
GS1 number-space.=C2=A0 So the GLUE registry should clarify that the<br>
identifier to be used in the URN is the manufacturer prefix of the GS1<br>
code.<br>
<br>
ChatGPT also said:<br>
<br>
=C2=A0 =C2=A0 Are prefixes ever reused?<br>
<br>
=C2=A0 =C2=A0 Yes, but with strict controls.<br>
<br>
=C2=A0 =C2=A0 GS1 policy allows reuse only after a long period and under<br=
>
=C2=A0 =C2=A0 conditions designed to avoid confusion.<br>
<br>
As I said, I don&#39;t *trust* this, but conversely, allowing prefixes to b=
e<br>
used in URNs requires that we obtain positive verification that a<br>
prefix that is assigned to an organization can never be reassigned to a<br>
&quot;different&quot; organization.<br>
<br>
This effort I consider to be a start on verifying that GS1 can be<br>
incorporated into a URN.=C2=A0 Similar investigation is needed for the othe=
r<br>
four proposed incorporated identifiers.<br>
<br>
Dale<br>
</blockquote></div></div>

--000000000000f119ff0650b85415--


--===============3159378737440926754==
Content-Type: text/plain; charset="utf-8"
MIME-Version: 1.0
Content-Transfer-Encoding: base64
Content-Disposition: inline

X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX18KdXJuIG1haWxp
bmcgbGlzdCAtLSB1cm5AaWV0Zi5vcmcKVG8gdW5zdWJzY3JpYmUgc2VuZCBhbiBlbWFpbCB0byB1
cm4tbGVhdmVAaWV0Zi5vcmcK

--===============3159378737440926754==--