[MPEG-OTSPEC] Scheduling Zoom meeting to discuss new proposals and AHG recommendations

[Dave Crossland]

Hi Skef

On Wed, Jul 12, 2023 at 5:38 PM Skef Iterum <skef at skef.org> wrote:

> Responding as Josh is currently on PTO and I'm privy to some internal
> discussion on this.
>

Thanks!!



> On 7/12/23 14:48, Dave Crossland wrote:
>
>
> Hi Josh
>
> On Fri, Jun 30, 2023 at 1:01 PM Dave Crossland <dcrossland at google.com>
> wrote:
>
>>
>> Any chance you or anyone from Adobe will be attending the TypeCon event
>> in August
>>
>
> It looks like Rod and I have travel approved by our VP to meet in Portland
> in August.
>
> I reached out to Neil Summerour at SoTA but haven't heard back, but since
> Typecon isn't in a hotel venue this year, I am guessing they may not be
> able to offering a large meeting room space anyway.
>
>
> I lived pretty close to this area until about 2 years ago. Another option,
> assuming some source of funding, would be an appropriate part of the
> Jupiter Hotel, which is one of the primary hotels for the event (assuming
> there is still availability, of course).
>
> https://www.jupiterhotel.com/events
>

Sure. Everyone expecting to join, please fill in the form, and I'll sort
something out!



> Will Adobe be able to host a meeting? I see from a quick websearch there
> is an Adobe office in Portland (1500 SW First Ave) :)
>
> Christopher and I chatted about this and (at least according to reported
> employee counts) the Adobe Portland office is if anything more modest than
> the Google office. In any case the largest room seats 14 in a pinch.
>

Got it. Google should be fine then :)

Another f2f opportunity I see before EOY is the nonlatin.org event at
Stanford, for which the Adobe HQ might be more amenable :)


> How does GPOS/GSUB work for CFF2 fonts' 70,000th glyph?
>
> I believe in the same way it would work for glyf/loca, i.e. any shaping
> issues above 64K are beyond the scope of the table. So it is prepared for
> the future without being entirely responsible for it.
>

Good stuff!


> It's a shame, though, that COLR v1 apparently didn't take CFF2 into
> account by allowing larger glyphIDs and exceeding the maxp count. Because
> basically all shaping is tied to the base glyph ID, that would have allowed
> exceeding 64K "shapes" for compositing in COLR. (Or come very close to it
> -- a few other tables might need related revisions.)
>

Indeed it is a shame no one (afaik) raised this before the v1 spec was
finalized.

Still, it remains a good goal for the b64k work in hand to address all
tables, including COLRv1 :)

Cheers
Dave
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.aau.at/pipermail/mpeg-otspec/attachments/20230712/c3c9e3bb/attachment-0001.html>