Re: CFD: eXIf 2017-05-31

Glenn Randers-Pehrson <[email protected]> Wed, 7 Jun 2017 17:07:05 -0400
Newsgroups gmane.comp.graphics.png.general
Message-ID <CA+PdXctv-g3eAB290grJCCHaJnjrw-0DC9VVFbRp=xX5h3A+YA@mail.gmail.com>
On Wed, Jun 7, 2017 at 4:58 PM, Pavel Zlatovratskii <[email protected]> wrote:
> 07.06.2017 16:06, Glenn Randers-Pehrson пишет:
>>
>> Editors that don't want to guess can discard the eXIf chunk deliberately,
>> even though it's "safe-to-copy"
>
> At first, this will break backward compatibility: current editors knows
> nothing about eXIf chunk, fact this could be preferred over their data and
> trick that it could be discarded to avoid that.
>
> At second, what benefit in keeping undefined priority between eXIf and other
> chunks that you agreed to destroy all other (more or less useful) historical
> data?

We did not agree to that.  The chunk is safe-to-copy so it should only
be destroyed by a deliberate decision.  Someone who is perplexed by
this whole thing may decide to discard the eXIf chunk even though it
is safe-to-copy.

Our choices are
1. Make the chunk safe-to-copy, and sometimes material will
be copied even though it should have been updated.  The
important parts (camera data, etc., are preserved)

2. Make the chunk unsafe-to-copy, so all existing editors
will discard the important parts along with the unsafe parts.

3. Define the chunk entirely differently, similar to the December 2000
proposal, in which encoders are required to understand the
internals of the EXIF profile.

Glenn

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
png-mng-misc mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/png-mng-misc