Recommendations on Comments to draft-ietf-fax-tiff-fx-11/12.txt

For the San Francisco WG meeting, I prepared draft-ietf-fax-tiff-fx-12.txt,
modified per the results of the TIFF-FX Interoperability Matrix presented in
Atlanta. However, one line in the spec that needed to be modified was
missed, so another draft is needed. 
 
From the minutes of the San Francisco WG meeting:
--------------- begin ------------------------
The slide fax-4 explains for one comment based on the previous
(tiff-fx-11) and an additional check about raised points in the document. It
implies the addition of RFC 3249 (Implementers Guide) as NON normative
reference. Also, ITU IPR statements will be added as a reference. Most of
the other points were examined, but the editors recommended no changes to
the text. Claudio Allocchio asked the WG, and it agreed with the editors
recommendations. The question will be taken again on the ML, before the new
version (tiff-fx-13) is published. In this new version, also, an editorial
fix in section 7.2.1 will be made.
--------------- end ------------------------
 
The content from the slides, including the recommendations from San
Francisco, follows. 
 
Rob
--------------------------------------------
 
 
(1) Addition of image/tiff-fx MIME type
 
Not a technical feature of the file format spec.
 
Additional note: In San Francisco, I recommended no change to the text.
However, removing the tiff-fx registration from Section 9 was also suggested
and that is what I would recommend now. 
 
Recommendation: Replace the text of Section 9 with a pointer to the
image/tiff and image/tiff-fx registration documents (RFCs 3302 and 3250).
 
(2) FillOrder=1 mandatory for all readers except profile S
 
Multiple independent interoperable implementations documented of FillOrder=1
for all profiles, except Profile S.
                                                        
Recommendation: No change to text.
 
(3) elimination of ColorMap for profile L
 
Profile L interoperability documented for implementations without the
ColorMap field.
Additional note: In the end, none of the Profile L readers/writers
implemented this field.
 
Recommendation: No change to text.  
 
(4) changes to profile M (from Annex C)
    Compression=1 for LayerBaseColorIFDs, adding StripByteCounts
 
Compression=1 for LayerBaseColorIFDs: Same capability as Proposed Standard;
implementation modified and verified during interop testing (see
ImageBaseColor in Interoperability report and matrix).
 
Recommendation: No change to text.
 
Adding StripByteCounts: This field was included by reference to Sec. 2.2.1
in Proposed Standard, clarifying text only was added.  
 
Recommendation: No change to text.
 
(5) addition of GlobalParamtersIFD, FaxProfile, CodingMethods as options to
profile F (in the table in section 4.7).
 
GlobalParamtersIFD and the other fields were optional in Profile F according
to Section 2.2.4 of RFC 2301, but omitted from Table 4.7; use of
GlobalParameters IFD included in Profile F interop testing.
                           
Recommendation: No change to text. 
 
(6) Orientation != 1   (optional for both readers and writers will result in
non-interoperability for mirror-images)
 
Orientation is an "information" field: facsimile doesn't typically generate
mirror images.
 
Recommendation: No change to text. 
 
(7) PhotometricInterpretation=1 for profile F (optional for readers and
writers, will result in poor interoperability)
 
Table 4.7 shows PhotometricInterpretation=1 required for Profile F.
 
Recommendation: No change to text. 
 
(8) Comments from section 5.2.2 of RFC 3249 should be folded into
description of profile C.
 
Recommendation: Add RFC 3249 as a non-normative reference. 
 
(9) Need an applicability statement to insure that this is not taken as a
general 'document image' standard.
 
The intended use of the standard is given in Section 1. 
 
Recommendation: No change to text.  
 
(10)Editorially change places where the word 'TIFF' is used incorrectly to
describe 'TIFF-FX' (especially in the phrase 'TIFF for Facsimile').
 
The phrase "TIFF for facsimile" has been in the text since before the
Proposed Standard in 1998 and has not caused any confusion. Should be OK as
long as its use is confined to this document.
 
Recommendation: No change to text. 
 
(11)The IPR statement in the document only points to www.ietf.org/ipr.html,
but that location does not contain any mention of JBIG patents. Should it
also point to the ITU IPR statements?
 
Recommendation: Add reference to ITU IPR statements to www.ietf.org/ipr.html