Re: Updated draft coming up

Thanks for these notes.

I like the idea of keeping the format of the 'SVG ' table in one
stable spec and discussing profiles elsewhere.

This way, tooling and implementations for the 'SVG ' table can
proceed, referencing a spec that's stable on that front.

The "unified" draft of SVG-in-OT already says:

  Security

  It is required that all rendering of SVG glyphs be done in the
  ³secure animated mode² or ³secure static mode² specified in the W3C
  document SVG Integration <https://svgwg.org/specs/integration/>.
  These modes permit no script execution, external references,
  interactivity, or link traversal.

This "security" aspect feels somewhat different from the "by design"
aspect which I associate with prohibiting foreignObject (we don't want
anything not in SVG!). And	the fact that some SVG impls will not
support certain parts of SVG feels like yet another flavor of
profiles. There	may be other aspects of	SVG we prohibit	on the "by
design" front as well.

How about we keep the above wording as is, and skip mention of
foreignObject or other restrictions in the draft I'm about to send
out?

This way, we can continue to discuss the different flavors of profiles
and not have it impact the Note. A vaguely worded reference to
profiles is fine as well.

So there will be at least 3 parts to this proposal:

- SVG table

- CPAL table (version 0 stable, have started working on version 1
  which will include the name IDs, but that may take time to be
  reviewed). And yes, when CPAL	changes, the above SVG table spec
  won't.

- A reference to the SVG Integration doc and perhaps other profiling
  docs

Sairus



-----Original Message-----
From: Chris Lilley <chris@w3.org>
Organization: W3C
Date: Friday, September 27, 2013 10:07 AM
To: Sairus Patel <sppatel@adobe.com>
Cc: "public-svgopentype@w3.org" <public-svgopentype@w3.org>
Subject: Re: Updated draft coming up

>Hello Sairus,
>
>Thursday, September 26, 2013, 5:31:32 PM, you wrote:
>
>> This would be a good draft for the Chair to publish as this group's
>>first
>> "Report."
>
>
>> Changes so far from the "unified draft":
>
>> - Use CPAL. Thread "FW: [mpeg-OTspec] COLR, SVG tables: sharing color
>> palettes".
>>   I'm trying to obtain a standalone CPAL proposal to reference.
>
>So if CPAL changes, the svg-in-OpenType spec would not change, right?
>
>> - Remove wording around font bounding box (thanks, Nikos). Discussed in
>> "Unified draft of SVG-in-OT" thread.
>
>> - Not permitting foreignObject.
>
>> Others?
>
>Not permitting foreignObject is an obvious first step, but that
>implies that all of the rest of SVG is permitted.
>
>I could reasonably foresee one or two profiles of SVG
>- static (non animated) SVG profile for OpenType
>- animated SVG profile for OpenType
>
>However, those would also be separate documents. And these might be
>recommended subsets, or might be minimal supported subsets (ie you
>have to do this and you can do more). So perhaps some vague
>language such as
>
>"Specifications which reference SVG in OpenType may choose to
>establish minimal conformant subsets of SVG. Such profiling is outside
>the scope of this specification."
>
>or something like that. In other words, it may happen, but not in this
>spec, which is stable.
>
>[chair hat]
>Please let me know when the draft report is available, and I will
>issue a call for consensus to publish the report. I would expect a
>one-week period for objections, given that the diff from the previous
>spec is small. After that, it will be published as a Community Group
>Note.
>[/chair hat]
>
>
>-- 
>Best regards,
> Chris                            mailto:chris@w3.org
>

Received on Friday, 27 September 2013 18:08:26 UTC