Re: DRAFT: eXIF 2017-03-05

On 6 March 2017 at 20:48, Phil Harvey wrote:
> I have 2 problems with this.
>
> 1. I still object to saying that IFD0 may be omitted, because it may not.  You MUST have at least include the entry count for IFD0  (which will be 0 if the IFD is empty), so technically I maintain that IFD0 must exist if IFD1 exists.

I concur.

> 2. There is absolutely no need to say that any gap is filled.  It is only required that the IFD pointers point to the start of the IFD.  It doesn't matter what comes before it.  Some EXIF/TIFF writers will put the values first, and others will put the IFD first.   I have also see unreferenced data come first.  Any of these is fine.

I don't see a disagreement between Glenn and Phil here. My
interpretation of "padding bytes of unspecified content" is that Glenn
is saying the same thing as Phil: that there may be unreferenced data
of unspecified content before any IFD.

Maybe if Glenn avoids using "filled" and "padding", would that make it
less confusing? How about:

"There may be unreferenced data segments of unspecified content,
before or after any IFD."

Sincerely,
Cosmin

------------------------------------------------------------------------------
Announcing the Oxford Dictionaries API! The API offers world-renowned
dictionary content that is easy and intuitive to access. Sign up for an
account today to start using our lexical data to power your apps and
projects. Get started today and enter our developer competition.
http://sdm.link/oxford