Re: CFD: eXIf 2017-05-31

On Wed, Jun 7, 2017 at 3:51 AM, Pavel Zlatovratskii <[email protected]> wrote:
>
> 03.06.2017 14:07, Glenn Randers-Pehrson пишет:
>>
>> We have the following already in the PNG specification version 1.2.  I
>> suppose
>> we could copy it into the png extensions document, where the eXIf chunk
>> specification will reside if we approve it:
>>
>> <p>The words "must", "required", "should", "recommended", "may", and
>> "optional" in this document are to be interpreted as described in
>> RFC-2119, which is consistent with their plain English meanings.  The word
>> "can" carries the same force as "may".
>
> Well, I think this would be nice, especially as the most final (W3C/ISO)
> version of main PNG specification miss these words.

OK

>
> But why not to add this reference to proposal too? In practical terms that
> doesn't produce very much difference, but on formal side single addition
> will looks like word "recommended" is interpreted by readers will in
> proposal and magically transforms to rfc2119 in extension specification.

I"d rather not issue a new CFD for that.  Please just take this message
as basis for understanding the words "must", etc.

>
> It might be influence of my (bureaucratic) work, but I wish to prevent any
> possible misinterpretation and inconsistency. When I read EXIF specification
> I had strong feeling that authors supposed that thumbnail would be updated
> with image every time. But there is no explicit declarations of it and so we
> experience issues we discuss before.
>
> And there is at least one inconsistency in this paragraph for now:
> If recommendation about "data should be considered to be of historical value
> only" is strong condition, not weak one, then it is pretty clear set
> priority of PNG chunks over EXIF data.

It is intended to set the value (usefulness) of the EXIF data, without
esablishing priority among chunks and EXIF.

Glenn

> So sentence "It is beyond the scope
> of this specification to resolve potential conflicts between data in the
> eXIf chunk and in other PNG chunks." may be valid only in case of
> independent knowledge of the validity of both EXIF data and PNG chunks.
> I think addition of reference to this exact case (and change order of
> sentences) will make specification more clear. Or it is possible to just
> strike out this sentence.

The paragraph looks OK to me as it appears in the 05-31 CFD.  It's
pretty clear that the authors of  PNG decoders need to decide for
themselves how to handle redundant or conflicting information. We
can't do that for them.

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