MIME type discussion (was RE: Telcon minutes - April 13, 2011)

Hello WG,

We discussed the issue of media types and 'font-woff' in particular during today's telcon. While we realize that there are many devices with different capabilities, the media type application/font-woff was registered specifically for use with WOFF resources, and the WOFF spec makes it clear that the primary purpose of this resource type is to be used for Web documents where fonts are linked by means of CSS @font-face rule, which have its own mechanism for format hints. Therefore, no additional optional parameters are needed for fonts-woff, even though we acknowledged the fact that other applications may benefit from having different media types and, possibly, additional optional parameters defined for fonts.

The participants on the call recommend that this should be the position of the WebFonts WG w.r.t. to the question raised by ISO ad hoc group.

Comments, objections?

Thank you,
Vlad


From: public-webfonts-wg-request@w3.org [mailto:public-webfonts-wg-request@w3.org] On Behalf Of Levantovsky, Vladimir
Sent: Wednesday, April 13, 2011 4:54 PM
To: public-webfonts-wg@w3.org Group
Subject: Telcon minutes - April 13, 2011

http://www.w3.org/2011/04/13-webfonts-minutes.html

and also as plain text below:

                               - DRAFT -

                 WebFonts Working Group Teleconference

13 Apr 2011

   See also: [2]IRC log

      [2] http://www.w3.org/2011/04/13-webfonts-irc

Attendees

   Present
   Regrets
   Chair
          SV_MEETING_CHAIR

   Scribe
          sylvaing

Contents

     * [3]Topics
     * [4]Summary of Action Items
     _________________________________________________________

   <trackbot> Date: 13 April 2011

   <John> zakim aaaa is John

   <John> Sylvain, are you joining the call?

   trackbot-ng, start telcon

   <trackbot> Meeting: WebFonts Working Group Teleconference

   <trackbot> Date: 13 April 2011

   <scribe> scribenick: sylvaing

   UNKNOWN_SPEAKER: discussing MIME type for web fonts

   vlad: once it was realized that woff already used a MIME type, it
   wasn't clear what the application was for
   ... for a container format, one also needs to know what is inside
   the container
   ... hence the question as to whether a parameter to define the
   contained type would be necessary

   john: how critical is this for CSS ?

   <John> CFF

   sylvaing: not critical for browsers which care about where the
   request comes from e.g. src descriptor in @font-face
   ... so which apps need a MIME type ?

   vlad: without explicit knowledge of what applications will use this,
   we wanted to be explicit

   sergeym: one can be explicit about this in CSS using the format hint
   ... scribe is catching up. We are discussing use-cases where WOFF is
   not loaded through CSS
   ... do we have any applications in mind where fonts will be
   downloaded that won't involve CSS ?

   vlad: yes, Java/J2ME apps on a mobile phone

   sylvaing: who's asking for a MIME type ?

   vlad: some font developers
   ... they want a set of MIME types that covers not just WOFF but
   other formats

   sylvaing: so this is outside WOFF's scope and our WG's ?

   vlad: yes, but we should at least offer our opinion
   ... and, for WOFF, should we allow for parameters to describe what
   is inside the WOFF container

   john: but where do you draw the line ? some people will want to know
   what kind of layout tables and other features are available

   vlad: in this case there are well-defined limits as to what can be a
   MIME type

   sylvaing: but if it's ok for CSS format hints to just say woff, why
   do mime types need extra parameters

   vlad: it's up to the implementation. since we're outside CSS, we
   don't really know what's needed

   sylvaing: I'm ok with it in principle but it's hard to argue without
   use-cases

   vlad: right. the discussion at ISO started as 'other types have this
   therefore fonts should have it' and this started the whole
   discussion

   john: I was wondering if this could be of use for a mobile phone
   e.g. windows phone to get a content-type header vs. waiting for the
   resource to sniff it ?

   sylvaing: nice in principle but experience shows that in practice
   this header can't be trusted by browsers whether mobile or desktop

   john: for WOFF, we've effectively presumed the use of CSS

   sylvaing: yes, and thus for use-cases that involve WOFF and other
   formats outside CSS, use-cases would be very helpful

   john: we acknowledge there are use-cases but they are outside web
   standards and thus beyond WOFF's scope. E.g. use in Blu-ray players
   does not involve web documents and is not addressed by WOFF or the
   WebFonts WG.

   <scribe> ACTION: Vlad to email the WebFonts WG with the proposed
   response to ISO [recorded in
   [5]http://www.w3.org/2011/04/13-webfonts-minutes.html#action01]

   <trackbot> Created ACTION-86 - Email the WebFonts WG with the
   proposed response to ISO [on Vladimir Levantovsky - due 2011-04-20].

   <John> Looks good

   <Vlad> trackbot, make minutes

   <trackbot> Sorry, Vlad, I don't understand 'trackbot, make minutes'.
   Please refer to [6]http://www.w3.org/2005/06/tracker/irc for help

      [6] http://www.w3.org/2005/06/tracker/irc

Summary of Action Items

   [NEW] ACTION: Vlad to email the WebFonts WG with the proposed
   response to ISO [recorded in
   [7]http://www.w3.org/2011/04/13-webfonts-minutes.html#action01]

   [End of minutes]