[OpenType] CFF2 and lsb

Wed Mar 1 16:41:04 CET 2017

Hi Dave, all,

How are you? Just wanted to remind you that the next WG11 (MPEG) meeting will take place April 3-7 and we have until March 28 to finalize and submit an input contribution with the proposed changes. Process-wise, it would be great if the input contribution is submitted as a result of the email discussion on the AHG list (mpeg-OTspec at yahoogroups.com<mailto:mpeg-OTspec at yahoogroups.com>) – when you think you are ready to propose an updated text to address the lsb issue I can kick-start the discussion by simply forwarding Werner’s original email to the list.

Also, I’d really appreciate if you could find a time to review the updated Working Draft text of the OFF (sent to the AHG list, see https://groups.yahoo.com/neo/groups/mpeg-OTspec/conversations/messages/1575) and submit your comments via email list as well.

Thank you,
Vlad

From: Dave Arnold [mailto:darnold at adobe.com]
Sent: Friday, February 10, 2017 6:30 PM
To: Levantovsky, Vladimir
Cc: Peter Constable (petercon at microsoft.com); Read Roberts
Subject: Re: [OpenType] CFF2 and lsb

Hi Vlad,

I've been able to discuss this only briefly with colleagues at Adobe. There are some aspects of LSB, both for variable fonts and for non-variable fonts, that we feel will require further discussion and it does not seem appropriate to delay your submission now. So, we have no changes to propose for this round.

Peter, on the variable font aspects, I already noted that we concluded LSBmapping was not useful for CFF2. But I was struggling to explain why the situation was different for TT. One issue that concerns us is the non-linearity of Min/Max and the difficulty of relying on interpolation to compute LSB for variable fonts. I'd like to discuss this with you before proposing wording for the spec.

Thanks.

-Dave

On 2/10/2017 9:24 AM, Levantovsky, Vladimir wrote:

Off-list

Hi Dave,

I wanted to follow up on the email exchange on OpenType list regarding spec clarifications. We are approaching the deadline for submission of the updated working draft OFF, and it would be an excellent opportunity to include some additional statements into the spec, before it is made available for public review.

I can drag my feet a little longer if you think you can suggest some additional language soon, please let me know if you think it is feasible to add some new CFF2-related requirements and/or recommendations at this time.

Thank you,

Vlad

-----Original Message-----

From: listmaster at indx.co.uk<mailto:listmaster at indx.co.uk> [mailto:listmaster at indx.co.uk] On Behalf Of Peter Constable

Sent: Wednesday, February 08, 2017 11:49 AM

To: listmaster at indx.co.uk<mailto:listmaster at indx.co.uk>

Subject: RE: [OpenType] CFF2 and lsb

Message from OpenType list:

Dave, can you suggest some statements for requirements or recommendations, and which chapter you think they belong.

Thanks

Peter

-----Original Message-----

From: listmaster at indx.co.uk<mailto:listmaster at indx.co.uk> [mailto:listmaster at indx.co.uk] On Behalf Of Dave Arnold

Sent: Tuesday, February 7, 2017 1:13 PM

To: listmaster at indx.co.uk<mailto:listmaster at indx.co.uk>

Subject: Re: [OpenType] CFF2 and lsb

Message from OpenType list:

****** Attachments to this email message have been removed ******

Hi Werner,

Adobe builds CFF fonts with head.flags bit 1 set and with lsb in the hmtx table set to xMin. As with the hmtx advance width, the lsb is informational only; its value is not used in rendering.

For CFF2, hmtx should be built the same way, using the default instance.

For CFF2, lsb in HVAR is not useful, so an LSBmapping table is not provided..

I agree the spec could be clarified on these points.

Thanks.

-Dave

________________________________

From: listmaster at indx.co.uk<mailto:listmaster at indx.co.uk> <listmaster at indx.co.uk><mailto:listmaster at indx.co.uk> on behalf of Werner LEMBERG <werner at lemberg.at><mailto:werner at lemberg.at>

Sent: Monday, February 6, 2017 10:08:28 PM

To: listmaster at indx.co.uk<mailto:listmaster at indx.co.uk>

Subject: Re: [OpenType] CFF2 and lsb

Message from OpenType list:

In non-variable fonts, bit 1 of head.flags is used to indicate whether

or not lsb = xMin (lsb point at x = 0).  That flag pertains to

TrueType outlines only.

Thanks.

Adobe folk — or someone who understands CFF(2) better than me — can

clarify the CFF question.

It's basically the question how CFF and CFF2 is integrated into the SFNT format.  My gut feeling is that bit 1 is always set for CFF and CFF2, but it would be nice have something in the specs that clarifies the issue.

    Werner

List archive: http://www.indx.co.uk/biglistarchive/

List settings: http://www.indx.co.uk/biglistarchive/?mode=usersettings

subscribe: opentype-subscribe at indx.co.uk<mailto:opentype-subscribe at indx.co.uk>

unsubscribe: opentype-unsubscribe at indx.co.uk<mailto:opentype-unsubscribe at indx.co.uk>

messages: opentype-list at indx.co.uk<mailto:opentype-list at indx.co.uk>

****** Attachments to this email message have been removed ******

List archive: http://www.indx.co.uk/biglistarchive/

List settings: http://www.indx.co.uk/biglistarchive/?mode=usersettings

subscribe: opentype-subscribe at indx.co.uk<mailto:opentype-subscribe at indx.co.uk>

unsubscribe: opentype-unsubscribe at indx.co.uk<mailto:opentype-unsubscribe at indx.co.uk>

messages: opentype-list at indx.co.uk<mailto:opentype-list at indx.co.uk>

List archive: http://www.indx.co.uk/biglistarchive/

List settings: http://www.indx.co.uk/biglistarchive/?mode=usersettings

subscribe: opentype-subscribe at indx.co.uk<mailto:opentype-subscribe at indx.co.uk>

unsubscribe: opentype-unsubscribe at indx.co.uk<mailto:opentype-unsubscribe at indx.co.uk>

messages: opentype-list at indx.co.uk<mailto:opentype-list at indx.co.uk>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.aau.at/pipermail/mpeg-otspec/attachments/20170301/dd78e19b/attachment.html>