EXIF chunk

--===============1048488808787152594==
Content-Type: multipart/alternative; boundary="000000000000bb9fe3061661a3a3"

--000000000000bb9fe3061661a3a3
Content-Type: text/plain; charset="UTF-8"

I CCed this to png-mng-implement in an attempt to catch all participants
but if you are reading it there please respond *ONLY *on png-mng.misc.

This was last discussed extensively in 2016/2017.  It's just become an
urgent issue.  The W3C PNG v3 spec has a specification for eXIf which is,
so far as I can tell, identical to my original one from 2017 and Glenn (I
suspect) did add support for the chunk as a public chunk; not a private
up-for-review chunk.

I believe, given Glenn's approach, that there therefore was a *successful*
vote on the black-box safe-to-copy proposal (as in my original proposal as
in the PNG v3 specifiction).

Unfortunately the extensions document on libpng.org (1.4.6) does not seem
to reflect this and ++unfortunately the W3C now seem to be changing the
spec to require a decoder which handles the chunk to respond to the
orientation tag and rotate the PNG.  libpng doesn't care about this but to
make it happen eXIf (still safe-to-copy) has to happen before the first
IDAT and that change has been made to the current spec:

https://www.w3.org/TR/png-3/#5ChunkOrdering

(Scroll down to eXIf.)

*So was it actually approved before?  If it was then can someone identify
the successful vote?*

The first vote (20170203:
https://sourceforge.net/p/png-mng/mailman/message/35644915/) failed but it
failed on unsafe vs safe to copy and, apparently, nit-picking.

SourceForge searches do not show a second CFV and the CFD seems to have
petered out here:

https://sourceforge.net/p/png-mng/mailman/message/35884822/

Yet Glenn, or someone DID implement it as a *public *chunk.  I don't
understand.

John Bowler <[email protected]>

--000000000000bb9fe3061661a3a3
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div>I CCed this to png-mng-implement in an attempt to cat=
ch all participants but if you are reading it there please respond <b>ONLY =
</b>on png-mng.misc.</div><div><br></div>This was last discussed extensivel=
y in 2016/2017.=C2=A0 It&#39;s just become an urgent issue.=C2=A0 The W3C P=
NG v3 spec has a specification for eXIf which is, so far as I can tell, ide=
ntical to my original one from 2017 and Glenn (I suspect) did add support f=
or the chunk as a public chunk; not a private up-for-review chunk.<div><br>=
</div><div>I believe, given Glenn&#39;s approach, that there therefore was =
a *successful* vote on the black-box safe-to-copy proposal (as in my origin=
al proposal as in the PNG v3 specifiction).</div><div><br></div><div>Unfort=
unately the extensions document on <a href=3D"http://libpng.org">libpng.org=
</a> (1.4.6) does not seem to reflect this and=C2=A0++unfortunately the W3C=
 now seem to be changing the spec to require a decoder which handles the ch=
unk to respond to the orientation tag and rotate the PNG.=C2=A0 libpng does=
n&#39;t care about this but to make it happen eXIf (still safe-to-copy) has=
 to happen before the first IDAT and that change has been made to the curre=
nt spec:</div><div><br></div><div><a href=3D"https://www.w3.org/TR/png-3/#5=
ChunkOrdering">https://www.w3.org/TR/png-3/#5ChunkOrdering</a></div><div><b=
r></div><div>(Scroll down to eXIf.)</div><div><br></div><div><b><i>So was i=
t actually approved before?=C2=A0 If it was then can someone identify the s=
uccessful vote?</i></b></div><div><br></div><div>The first vote (20170203: =
<a href=3D"https://sourceforge.net/p/png-mng/mailman/message/35644915/">htt=
ps://sourceforge.net/p/png-mng/mailman/message/35644915/</a>) failed but it=
 failed on unsafe vs safe to copy and, apparently, nit-picking.</div><div><=
br></div><div>SourceForge searches do not show a second CFV and the CFD see=
ms to have petered out here:</div><div><br></div><div><a href=3D"https://so=
urceforge.net/p/png-mng/mailman/message/35884822/">https://sourceforge.net/=
p/png-mng/mailman/message/35884822/</a></div><div><br></div><div>Yet Glenn,=
 or someone DID implement it as a <b>public </b>chunk.=C2=A0 I don&#39;t un=
derstand.<br><div><br clear=3D"all"><div><div dir=3D"ltr" class=3D"gmail_si=
gnature" data-smartmail=3D"gmail_signature"><div dir=3D"ltr"><div>John Bowl=
er &lt;<a href=3D"mailto:[email protected]" target=3D"_blank">[email protected]=
</a><font color=3D"#000120"></font><u></u>&gt;</div><div><br></div></div></=
div></div></div></div></div>

--000000000000bb9fe3061661a3a3--


--===============1048488808787152594==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline


--===============1048488808787152594==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

_______________________________________________
png-mng-implement mailing list
png-mng-implement-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
https://lists.sourceforge.net/lists/listinfo/png-mng-implement

--===============1048488808787152594==--