W3C home > Mailing lists > Public > www-style@w3.org > June 2009

RE: New work on fonts at W3C

From: Sylvain Galineau <sylvaing@microsoft.com>
Date: Tue, 23 Jun 2009 13:13:11 -0700
To: Aryeh Gregor <Simetrical+w3c@gmail.com>
CC: CSS 3 W3C Group <www-style@w3.org>
Message-ID: <DBCD4020258D924A9887739500CF8E599CDA1126C1@NA-EXMSG-C125.redmond.corp.microsoft.com>

>-----Original Message-----
>From: simetrical@gmail.com [mailto:simetrical@gmail.com] On Behalf Of
>Aryeh Gregor

>Services like Typekit attempt to make OTF/TTF "safe" enough for
>foundries to accept without browser-side changes, by means of
>client-side scripting and possibly other things (sneaky hacks in the
>font files? clever server-side scripts?).  Likewise, techniques such
>as John Daggett outlined could provide most of the "safety" that font
>foundries want in the near future without explicit browser support.
>Do you feel confident that these efforts will fail, or that they'd be
>significantly technically inferior to what a new font format could do?

I have not been able to use Typekit to evaluate their solution, nor do I know which foundries participate, how, what the Typekit EULA will require etc. All insights welcome. But I can note that the solution does require extra code to run within your application - i.e. the browser bits are unaffected but your own application acquires a runtime dependency - code that will likely be specific to Typekit and the constraints of its EULA. Should such hosted services be successful, I can imagine a large site licensing fonts from two or more of them, each requiring its own licensing enforcement code. Would that really better than one single, simple, standard encoding a plurality of font vendors are comfortable with, enabling users to decide whether 'hosted' or 'direct' licensing best fit their needs ? In other words, are you comfortable with the success of such services - or that of a single one - as a long-term solution to this issue ? Is the ability to reference raw font files so technically important that you would be willing to restrict commercial font use on the web to proprietary hosted solutions ? 

Regardless of the answer I'm not sure I understand why (or how) I'd have to be confident in font hosting's failure in order to support a standard solution that would be orthogonal to this specific business model; Typekit may very well use such a font encoding if it were available and still grow its business. As you note, it may even be part of their solution ! That Typekit can exist today should not preclude options that can expand user choice both in terms of fonts available for web use and licensing options.
 
As for the suitability of John's solution, I must take a much closer look at it; but I certainly welcome any constructive proposal.
 
Lastly, fwiw, I don't think Ascender's proposal constitutes a 'new font format'. Stating, even implicitly, that font vendors demand a new font format vastly exaggerates the scope of the proposed solutions to date and is, imo, misleading. It would be more accurate to state that font vendors prefer licensed TTFs and OTFs to be encoded for web use and that this encoding shall be as simple as possible; John's proposal would clearly be the simplest such encoding option to date.



Received on Tuesday, 23 June 2009 20:13:56 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:19 GMT