Re: [WOFF] Checksums, IDs, and Local Cacheing

Doug Schepers wrote:
> Obviously, the name of the font (even with consideration for the major
> and minor versions) is not a good identifier... there are many duplicate
> names for different font families (how many are named "LED", for
> example) 

>   @font-face {
>     font-family: "Obscure Serif Bold";
>     src: url("http://www.example.com/resources/ObscureSeBd.woff");
>     checksum: "8675309";
>   }

Doug,

in the @font-face mechanism, the name of the font is specified by the
web developer, NOT the font developer. I.e. I can have something like:

  @font-face {
    font-family: "Obscure Serif Bold";
    src: url("http://www.example.com/resources/ObscureSeBd.woff");
  }

in one page and

  @font-face {
    font-family: "Obscure Serif Bold";
    src: url("http://www.example.com/resources/Arial.ttf");
  }

in another. It's the web developer who writes the CSS who specifies the
name by which he then later identifies the font in the CSS. This is
important, otherwise it wouldn't work: if browsers pulled the names
directly from the font files, you'd end up in hell because different
operating systems and different text APIs pull different names out of
fonts, also depending on which format the fonts are in.

It's important to remember that in CSS, the family names for @font-face
are really just like author-specified class or id names. It seems that
many people forget about this.

I agree with John that the URL is the only conceivable unique identifier
that should be used as a basis for caching. After all, it is the
*unique* resource locator.

Adam

-- 

Adam Twardoch
| Language Typography Unicode Fonts OpenType
| twardoch.com | silesian.com | fontlab.net

Reporter: "So what will your trip to Ireland look like?"
Lech Wałęsa: "I get into a car, then onto a plane, and then the other
way around."

Received on Sunday, 25 October 2009 07:13:19 UTC