TAG review required for upcoming COLRv1 font standard I2S?

[Dominik Röttsches]

Hi API owners,

TL;DR - What are your thoughts regarding requiring or not requiring a TAG review for COLR v1 font support in Blink?

In a few weeks to months I am planning to publish an intent to ship for the COLRv1 font standard. 

The COLRv1 standard is a collaboration between Chrome and Google Fonts, as well as external contributors Behdad Esfahbod and others (Ben Wagner, Laurence Penney, Chris Lilley).

In short, COLRv1 fonts are a highly efficient storage mechanism for color vector drawings used as glyph shapes, encoded in an updated version of the OpenType COLR table. The way COLRv1 is specified, it integrates well with previous font concepts such as the TrueType and CFF2 contour formats, as well as OpenType variations. In addition it lends itself well to future web font concepts such as incremental transfer. 

On the specification level, the standard is out for open review for integration into OpenType 1.9. 

It is also submitted to ISO via INCITS for integration into the ISO/IEC 14496-22 "Coding of audio-visual objects — Part 22: Open Font Format" and about to go into the ballot there.

The working github repository, where we do tip of tree development, which then is pushed downstream to the OpenType version and the ISO Open Font Format version can be found here.

COLRv1 brings an efficient, future-proof color vector font standard to the web.

What I would like to hear your opinion on: Do you consider it a requirement for this addition to Blink to go through a TAG review or not? 

• Arguably, the specification of this font format is developed outside the scope of W3C in the font standardization processes mentioned above.
• When variable fonts were introduced, I filed a TAG review only for the CSS fonts spec related parts, TAG review #183. For COLRv1 no additional changes to the CSS fonts spec are required. Palette support is already in the CSS fonts spec, and optional feature detection of font technologies was discussed in TAG review 666.

Thank you for your input,

Dominik

[Yoav Weiss]

In the past, we did not require a TAG review for support for new file formats.

At the same time, I'm wondering regarding the specification state of this new format, and its implications on cross-browser adoption. IIUC, you're aiming to integrate it with the broader OpenType specification, that's being developed in ISO. Is that correct?

Does ISO have some similar form of review body that can verify that what you're planning to ship fits well with the rest of OpenType?

I think it may also be interesting to ask other vendors what they think, and if they'd be interested to implement.

--
You received this message because you are subscribed to the Google Groups "blink-api-owners-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to blink-api-owners-d...@chromium.org.
To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/blink-api-owners-discuss/CAN6muBsnMV5851D_7mgx1Wm0v1JsG%2BUMQ7yH1VJzXXvkNZVJ6w%40mail.gmail.com.

[Yoav Weiss]

On Thu, Sep 16, 2021 at 10:16 AM Yoav Weiss <yoav...@google.com> wrote:

In the past, we did not require a TAG review for support for new file formats.

At the same time, I'm wondering regarding the specification state of this new format, and its implications on cross-browser adoption. IIUC, you're aiming to integrate it with the broader OpenType specification, that's being developed in ISO. Is that correct?

Does ISO have some similar form of review body that can verify that what you're planning to ship fits well with the rest of OpenType?

I think it may also be interesting to ask other vendors what they think, and if they'd be interested to implement.

On that front, I was pointed to Apple's position, which is... interesting. I appreciate the thoughtfulness in which you replied to it, as well as the support and thoughtful response from the Microsoft Edge team.

Any ideas where Mozilla is positioned?

[Dominik Röttsches]

Hi Yoav,

On Thu, Sep 16, 2021 at 11:17 AM Yoav Weiss <yoav...@google.com> wrote:

In the past, we did not require a TAG review for support for new file formats.

Good to know, thanks. In a discussion with Ian, that was our recollection as well. 

 

At the same time, I'm wondering regarding the specification state of this new format, and its implications on cross-browser adoption. IIUC, you're aiming to integrate it with the broader OpenType specification, that's being developed in ISO. Is that correct?

Yes, font specs are complicated. We're adding this format to two types of specifications:

1) OpenType: OpenType is a specification maintained by Microsoft https://docs.microsoft.com/en-us/typography/opentype/spec/ - Peter Constable is the maintainer of that specification. As you saw from his response on behalf of MS on webkit-dev, he's not only on board with this specification, but also collaborated with us on it. Microsoft developed COLRv0, the predecessor. Peter contributed to developing COLRv1. Then, for OpenType, the preview of the next version of it is out here: https://docs.microsoft.com/en-us/typography/opentype/otspec190alpha/ot190alpha and published for Alpha review where interested parties can submit feedback. 

2) ISO / OFF (=Open Font Format): In some sense, OFF can be considered an ISO mirror/copy of OpenType. There's a lot of history around this process. We coordinate upstreaming the spec we worked on to ISO for specification into the Open Font Format, which is part of the MPEG set of standards in ISO. The latest version of our specification has been posted there in a rather final stage, which I understand to be the "ballot comments" stage, in which ISO members can comment and request changes before the specification finally goes to a voting / ballot. We do not expect any objections and the voting to succeed. Vladimir Levantovsky is the chair of the ad-hoc group in ISO that oversees the OFF.

Does ISO have some similar form of review body that can verify that what you're planning to ship fits well with the rest of OpenType?

ISO maintains the OFF spec and has the ballot comment period above, OpenType is maintained by MS and has the Alpha review - but in the end MS decides what they want to publish as OpenType 1.9. In both standards tracks I expect the specification to be published without any major changes.

I think it may also be interesting to ask other vendors what they think, and if they'd be interested to implement.

 

On that front, I was pointed to Apple's position, which is... interesting. I appreciate the thoughtfulness in which you replied to it, as well as the support and thoughtful response from the Microsoft Edge team.

Thank you!  

 

Any ideas where Mozilla is positioned? 

The request to Mozilla is posted here: https://github.com/mozilla/standards-positions/issues/497

Jonathan Kew (jfkthame@) is the font expert at Mozilla. He has personally reviewed the specification, and contributed feedback to it, and his personal review comes out positively. I've repeatedly asked him to come forward with an official position on behalf of Mozilla, and he's recently started more actively trying to get something official out, but they have not publicly commented yet.

I hope this answers your questions, 

Dominik

[Yoav Weiss]

On Wed, Sep 22, 2021 at 10:57 AM Dominik Röttsches <dr...@chromium.org> wrote:

Hi Yoav,

On Thu, Sep 16, 2021 at 11:17 AM Yoav Weiss <yoav...@google.com> wrote:

In the past, we did not require a TAG review for support for new file formats.

Good to know, thanks. In a discussion with Ian, that was our recollection as well. 

 

At the same time, I'm wondering regarding the specification state of this new format, and its implications on cross-browser adoption. IIUC, you're aiming to integrate it with the broader OpenType specification, that's being developed in ISO. Is that correct?

Yes, font specs are complicated. We're adding this format to two types of specifications:

1) OpenType: OpenType is a specification maintained by Microsoft https://docs.microsoft.com/en-us/typography/opentype/spec/ - Peter Constable is the maintainer of that specification. As you saw from his response on behalf of MS on webkit-dev, he's not only on board with this specification, but also collaborated with us on it. Microsoft developed COLRv0, the predecessor. Peter contributed to developing COLRv1. Then, for OpenType, the preview of the next version of it is out here: https://docs.microsoft.com/en-us/typography/opentype/otspec190alpha/ot190alpha and published for Alpha review where interested parties can submit feedback. 

2) ISO / OFF (=Open Font Format): In some sense, OFF can be considered an ISO mirror/copy of OpenType. There's a lot of history around this process. We coordinate upstreaming the spec we worked on to ISO for specification into the Open Font Format, which is part of the MPEG set of standards in ISO. The latest version of our specification has been posted there in a rather final stage, which I understand to be the "ballot comments" stage, in which ISO members can comment and request changes before the specification finally goes to a voting / ballot. We do not expect any objections and the voting to succeed. Vladimir Levantovsky is the chair of the ad-hoc group in ISO that oversees the OFF.

Does ISO have some similar form of review body that can verify that what you're planning to ship fits well with the rest of OpenType?

ISO maintains the OFF spec and has the ballot comment period above, OpenType is maintained by MS and has the Alpha review - but in the end MS decides what they want to publish as OpenType 1.9. In both standards tracks I expect the specification to be published without any major changes.

OK. In that case, I think we can consider those reviews as a TAG equivalent for this case.

[Chris Harrelson]

On Wed, Sep 22, 2021 at 6:19 AM 'Yoav Weiss' via blink-api-owners-discuss <blink-api-ow...@chromium.org> wrote:

On Wed, Sep 22, 2021 at 10:57 AM Dominik Röttsches <dr...@chromium.org> wrote:

Hi Yoav,

On Thu, Sep 16, 2021 at 11:17 AM Yoav Weiss <yoav...@google.com> wrote:

In the past, we did not require a TAG review for support for new file formats.

Good to know, thanks. In a discussion with Ian, that was our recollection as well. 

 

At the same time, I'm wondering regarding the specification state of this new format, and its implications on cross-browser adoption. IIUC, you're aiming to integrate it with the broader OpenType specification, that's being developed in ISO. Is that correct?

Yes, font specs are complicated. We're adding this format to two types of specifications:

1) OpenType: OpenType is a specification maintained by Microsoft https://docs.microsoft.com/en-us/typography/opentype/spec/ - Peter Constable is the maintainer of that specification. As you saw from his response on behalf of MS on webkit-dev, he's not only on board with this specification, but also collaborated with us on it. Microsoft developed COLRv0, the predecessor. Peter contributed to developing COLRv1. Then, for OpenType, the preview of the next version of it is out here: https://docs.microsoft.com/en-us/typography/opentype/otspec190alpha/ot190alpha and published for Alpha review where interested parties can submit feedback. 

2) ISO / OFF (=Open Font Format): In some sense, OFF can be considered an ISO mirror/copy of OpenType. There's a lot of history around this process. We coordinate upstreaming the spec we worked on to ISO for specification into the Open Font Format, which is part of the MPEG set of standards in ISO. The latest version of our specification has been posted there in a rather final stage, which I understand to be the "ballot comments" stage, in which ISO members can comment and request changes before the specification finally goes to a voting / ballot. We do not expect any objections and the voting to succeed. Vladimir Levantovsky is the chair of the ad-hoc group in ISO that oversees the OFF.

Does ISO have some similar form of review body that can verify that what you're planning to ship fits well with the rest of OpenType?

ISO maintains the OFF spec and has the ballot comment period above, OpenType is maintained by MS and has the Alpha review - but in the end MS decides what they want to publish as OpenType 1.9. In both standards tracks I expect the specification to be published without any major changes.

OK. In that case, I think we can consider those reviews as a TAG equivalent for this case.

I agree. Therefore you would not need a TAG review. Unless others object on this thread, let's consider that the answer.

When sending an intent-to-ship for this feature, just link to this thread as a justification for no TAG review.

 

To view this discussion on the web visit https://groups.google.com/a/chromium.org/d/msgid/blink-api-owners-discuss/CAL5BFfX-S2_mxapS%2B%3DdkaZuqnRrzhFe_yR8NZqifN3DDt%3DivzA%40mail.gmail.com.