DR 09-0043 _ WML, Fonts: notTrueType attribute and bitmap fonts
mpsuzuki at hiroshima-u.ac.jp
Thu Nov 11 02:00:50 CET 2010
I think this is OK, many thanks to keep updated.
Murata-san, I think notTrueType in the final version would be:
a) the most conventional TrueType -> false (yes, TrueType)
b) TrueType Collection -> false
c) CFF OpenType (PostScript OpenType) -> true
d) others -> true
Maybe the result of b) would look odd, but I think it should
be like this, because:
* it is consistent with existing implementation by Microsoft
* the embedding of TrueType Collection into OOXML is not
supported by OOXML spec, nor by Microsoft Office implementation
When SC34/WG4 extends OOXML (or Microsoft extends Microsoft
Office implemenation) to support TTC embedding in future,
the handling of this element should be discussed, because
"embedded TTC" is new marginal case in the day.
suzuki toshiya, Hiroshima University, Japan
Chris Rae wrote (2010/11/11 7:53):
> Final version attached.
> If nobody objects, I think we can close this without a call, although happy to wait for a call too.
> -----Original Message-----
> From: suzuki toshiya [mailto:mpsuzuki at hiroshima-u.ac.jp]
> Sent: 09 November 2010 17:52
> To: Chris Rae
> Cc: e-SC34-WG4 at ecma-international.org; eb2m-mrt at asahi-net.or.jp
> Subject: Re: DR 09-0043 _ WML, Fonts: notTrueType attribute and bitmap fonts
> Dear Chris,
> Thank you for correction! I apologize to say a few more.
> In the last part, there is different wording: "a font including TrueType outlining". I recommend to use the same word in above, "outline" is better than "outlining".
> suzuki toshiya, Hiroshima University, Japan
> Chris Rae wrote:
>> Hello Suzuki-san and WG4. Suzuki-san - my apologies for my failure to comprehend this very successfully. I now remember you explained this distinction to me at the WG4 meeting in Tokyo and you've now very patiently explained it again.
>> Does the attached modified wording look acceptable as a final version?
>> -----Original Message-----
>> From: mpsuzuki at hiroshima-u.ac.jp [mailto:mpsuzuki at hiroshima-u.ac.jp]
>> Sent: 28 October 2010 03:32
>> To: Chris Rae
>> Cc: e-SC34-WG4 at ecma-international.org; eb2m-mrt at asahi-net.or.jp;
>> mpsuzuki at hiroshima-u.ac.jp
>> Subject: Re: DR 09-0043 _ WML, Fonts: notTrueType attribute and bitmap
>> Dear Chris,
>> I've checked Office for Mac v.X and 2008. They cannot embed any fonts into the document, so I think existing Microsoft implementations for OOXML had never embedded a TrueType font without outline.
>> Also I have to correct my previous post "Mac OS X accepts a TrueType font without outline glyph data".
>> Until Mac OS X 10.4 (so-called Tiger), it is true.
>> But after Mac OS X 10.5 (so-called Leopard), it is false. A TrueType font without outlne glyph data is refused as "this font has serious problem in sfnt structure, it cannot be used" (a font with minor problem is warned but accepted - a font with serious problem is not accepted).
>> # NOTE:
>> # When I write as "a TrueType font without outline # glyph data", it means a TrueType font without # loca/glyf table (tables for TrueType outline) # nor CFF table (table for PostScript outline).
>> # A TrueType font including loca/glyf table that # content is only single glyph for white square # (to be used as a fallback for missing glyph) # is NOT "a TrueType font without outline glyph data".
>> On Wed, 27 Oct 2010 22:17:45 +0900
>> mpsuzuki at hiroshima-u.ac.jp wrote:
>>> One of my concern was that "OpenType (or ISO/IEC 14496-22) permits a
>>> font including bitmap glyph only, and without any outline glyph (not
>>> TrueType, not PostScript), it should not be marked as notTrueType?".
>>> It seems that Mac OS X accepts such bitmap-only OpenType.
>>> I will check Office on Macintosh platforms, within 24 hours.
>>> Unfortunately, I have no access for the latest Office for Mac 2011.
>>> On Mon, 25 Oct 2010 20:44:41 +0000
>>> Chris Rae <Chris.Rae at microsoft.com> wrote:
>>>> This DR we discussed in Tokyo and my notes say I had some minor writing up to do and then we could close it. Unfortunately my notes don't say any more than that, so I am relying on my memory to work out what that writing up was. I *think* that the issue was the use of "TypeType" when we really meant "TrueType or OpenType", so I've edited the text to include that. I also spotted a typo which I've fixed.
>>>> Suzuki-san, is there a chance you could take a look at this one and confirm that these are the right changes? If they are, we should be able to close it easily on the next call.
More information about the sc34wg4