First draft of the ballot comments on the new amendment

[Hin-Tak Leung]

Dear Vlad,

It would indeed be nice if someone from Microsoft can confirm the issue soon. Meanwhile, we have already Cosimo from Dalton Maag speaking up; maybe we can get Adam Twardoch to support this change from his experience with the implementation in Fontlab Studio, and Martin Hosken can revive his old work, come up with an inter-operable implementation, with the suggested change below? Whether it is for immediate ballot comments or a later revision, it would be nice to have multiple inter-operable implementations too.

Adam/Martin: can either or both of you corroborate the change?

Regards,
Hin-Tak

--------------------------------------------
On Fri, 22/4/16, Levantovsky, Vladimir <Vladimir.Levantovsky at monotype.com> wrote:


 Dear Hin-Tak,
 
 Thank you again for reviewing
 the draft ballot comments and for your proposal. A change
 you proposed aims to clarify/correct the existing standard
 and can be made via ballot comments. However, I'd like
 to hear from someone at Microsoft confirming the issue you
 uncovered since a similar correction would have to be made
 in the OT spec hosted online. If we receive a confirmation
 of the proposed change before Apr. 27 I will add it as yet
 another ballot comment, otherwise we'd have to postpone
 it until next major revision. 
 
 Thank you,
 Vlad
 
 > On Apr 21, 2016, at 9:22 PM, Hin-Tak Leung
 <htl10 at users.sourceforge.net>
 wrote:
 > 
 > Dear
 Vlad,
 > 
 > Thanks for
 putting it all together. Whether it is appropriate as a
 ballot comment now or a more involved revision later, to put
 in a format closer to your draft, I propose that
 > 
 > under clause 
 > "5.7.1 DSIG ­ Digital signature
 table" , 
 > 
 >
 paragraph
 > "Format 1: For whole
 fonts, with either TrueType outlines and/or CFF data",
 
 > 
 > item 
 > "1. If there is an existing DSIG
 table in the font,",
 > 
 > in the list of 5 steps under item 1,
 deleting step 
 > "4. Zero out the
 file checksum in the head table.", 
 > 
 > and renumbering 
 > "5. Add the usFlag (reserved, set at
 1 for now) to the stream of bytes"
 >
 as 
 > "4. ....".
 > 
 > 
 > The reason of change is to align the
 algorithm description in the spec with the most widely used
 implementation of signing and checking from Microsoft. For
 signing, as noted, there are a few 3rd party implementations
 e.g. from Fontlab and Dalton Maag; for checking, perhaps the
 only implementation is Microsoft's (chktrust.exe and the
 older Font Validator are regarded as the same, as they have
 a same wintrust mssipotf backend).
 > 
 > Hin-Tak