>
> Hello Shillito
>
> On 08-Jan-01, you wrote:
>
> 8<--- snip ---
> >
> > The hardest aspect of implementing CSS is of course the treatment of
> > fonts, which is a complete mess on all computer platforms, and
> > everywhere else. CSS makes (in my opinion) the mistake of supporting
> > in parallel quite a few different font systems, no doubt in response
> > to the commercial interests represented in the committee which
> > designed it. It would have been better and cleaner to support just one
> > font system, and thus standardise things (which is what things like
> > HTML and CSS are supposed to do). At least, there is a good "generic"
> > sub-set of CSS's requirements which is reasonably practical to
> > implement, and which can up to a point fudge the many other apects
> > of it.
>
> 8<--- snip ---
>
> I don't follow what you mean when you say that CSS supports a "few
> different font systems". Can you elaborate?
>
1. First, it uses a system with styles font-family, font-weight,
font-stretch &c. as well as generic styles (serif, sans-serif,
cursive, fantasy and monospace), the font-family can be any
font at all, in any of the styles such as bitmapped, and all the
other types of fonts, e.g. the CSS2 standard mentions
.pfr,.eot,.pfb,.pfa,.ttf (portable font resource, embedded
open type, postscript, open type, true type, intellifont, speedo)
along with others.
2. Then, there are complex options for downloading fonts, which
also may be in many methods of packing, not just 1, along with
further complex (and not fully spelled out) methods of downloading
but not making cache-able subsets of copyright fonts for
displaying a web page but not useable elsewhere.
3. Along with the above, provisions for specifying one-by-one the
physical dimensions of each glyph of a font (with no real
specification how this fits in with the font-stretch &c styles)
4. The "panose" system, a set of numbers which Microsoft uses
to describe font attributes, also with no real specification
of how this interacts with the other methods.
It is quite possible (and probable) that styles would cascade down
in overlapping attributes from all these sorts of things at once.
The problem with the above, is that unlike various types of
"embedded objects", like flash video, mp3, applets, &c, each
of which can have "plug-ins" to implement them, fonts are
instrinsic to the display of web pages, rather than just a tack-on.
The descrption of how "user agents" are to attempt to meet the
requirements of an "@typeface rule" in 15.3.1 of the CSS2
specification is highly arcane, and the chapter on fonts
is very rambling and confusing.
Incidentally, the www.unicode.com has the entire CSS2
specification online.
--
Ken Shillito
Home Page: http://fast.to/shillito
Jesus is Lord!
_____________________________________________________________________
Voyager Mailing List - http://v3.vapor.com/
Voyager FAQ....: http://faq.vapor.com/voyager/
Listserver Help: mailto:[EMAIL PROTECTED]?Subject=HELP
Unsubscribe....: mailto:[EMAIL PROTECTED]?Subject=UNSUBSCRIBE
