- From: Glenn Adams <glenn@skynav.com>
- Date: Sun, 27 Jan 2019 12:37:14 -0700
- To: Pierre-Anthony Lemieux <pal@sandflow.com>
- Cc: TTWG <public-tt@w3.org>
- Message-ID: <CACQ=j+cDs4y5ovvpk6x4TH8+qp_k+0-mugfN9pMO_L2LXsA7LA@mail.gmail.com>
+1 On Sun, Jan 27, 2019 at 9:26 AM Pierre-Anthony Lemieux <pal@sandflow.com> wrote: > Hi all, > > Unless someone can identify, by the end of the week, a use case not > covered by (a) and requiring (b), I suggest clarifying that (b) is not > supported. > > Best, > > -- Pierre > > On Fri, Jan 25, 2019 at 10:37 AM Glenn Adams <glenn@skynav.com> wrote: > >> Correct. >> >> On Fri, Jan 25, 2019 at 3:23 AM Pierre-Anthony Lemieux <pal@sandflow.com> >> wrote: >> >>> > But I'm willing to consider use cases for allowing it. I just can >>> think of a good one. If you are going to provide a font resource you can >>> give >>> > it any family name, so what is the need to override a generic family >>> name? >>> >>> Say the author wishes to both specify a generic font family >>> (proportionalSansSerif) and include the actual font file used at authoring >>> (Arial). >>> >>> (a) the author can specify tts:fontFamily="Arial,default" and embed Arial >>> (b) the author can specify tts:fontFamily="default" and embed Arial such >>> that it aliases to 'default' >>> >>> (a) is supported today, and (b) is ambiguous today and the the proposal >>> is to forbid it. Did I get this right? >>> >>> Best, >>> >>> -- Pierre >>> >>> On Thu, Jan 24, 2019 at 6:51 AM Glenn Adams <glenn@skynav.com> wrote: >>> >>>> Correct. That is my assumption. But I'm willing to consider use cases >>>> for allowing it. I just can think of a good one. If you are going to >>>> provide a font resource you can give it any family name, so what is the >>>> need to override a generic family name? >>>> >>>> On Thu, Jan 24, 2019 at 7:31 AM Pierre-Anthony Lemieux < >>>> pal@sandflow.com> wrote: >>>> >>>>> Hi Glenn, >>>>> >>>>> Your conclusion is that tts:fontFamily="proportionalSerif" is always >>>>> implementation-dependent and cannot be overridden by the author? >>>>> >>>>> Best, >>>>> >>>>> -- Pierre >>>>> >>>>> On Thu, Jan 24, 2019 at 2:27 PM Glenn Adams <glenn@skynav.com> wrote: >>>>> >>>>>> I have (since around October) had the following note on my TODO list >>>>>> for filing issues against TTML2: >>>>>> >>>>>> + [SUBSTANTIVE] do not allow @family on font element to be >>>>>> >>>>>> a <generic-family-name> (either unquoted or quoted); >>>>>> >>>>>> On Thu, Jan 24, 2019 at 2:08 AM Pierre-Anthony Lemieux < >>>>>> pal@sandflow.com> wrote: >>>>>> >>>>>>> Good morning/evening, >>>>>>> >>>>>>> Say the following is present in a TTML2 document: >>>>>>> >>>>>>> <resources> >>>>>>> <font family="default" range="u+20-7f,u+90-9f"> >>>>>>> <source src="http://example.com/fonts/default.otf" >>>>>>> type="font/otf"/> >>>>>>> </font> >>>>>>> </resources> >>>>>>> >>>>>>> In the following snippet, is a presentation processor expected to >>>>>>> use the embedded font, or is the font selection sill left to the >>>>>>> implementation? >>>>>>> >>>>>>> <p tts:fontFamily="default">font above or application-selected?</p> >>>>>>> >>>>>>> Best, >>>>>>> >>>>>>> -- Pierre >>>>>>> >>>>>>
Received on Sunday, 27 January 2019 19:37:49 UTC