W3C home > Mailing lists > Public > www-style@w3.org > April 2004

Re: Fallbacks

From: Anne van Kesteren (fora) <fora@annevankesteren.nl>
Date: Tue, 13 Apr 2004 13:04:59 +0200
Message-ID: <407BC95B.1030505@annevankesteren.nl>
To: Jens Meiert <jens.meiert@erde3.com>, www-style@w3.org

> If we're about to exchange our preferences: I /don't/ 'like' it.

This and Tantek's proposal?

>>There are quite some designers who "hate" 
>>'Verdana' for being to big, however, it is also a "nice" font (for 
>>Windows).
> 
> Is 'hate' and 'nice' in any relationship to a CSS specification? I don't
> even see any parallel to any other development related issue, since
> subjective connotations and valuations ain't our job, either.

There is no relationship, I never said there was one. I was explaining a 
use for it and perhaps it was not a very good example, but I'm not a 
designer. I was inspired by a weblog entry from which I can't find the 
URI, which stated that different fonts need different font related 
properties.

>>   font:80% Verdana,100% Arial,sans-serif;
> 
> AFAIR, this case breaks the current implementation:

I bet this is not the only property that is changed in CSS3. (It is 
backwards compatible though. Browsers who read pages described in CSS 
2.1 will still give the same result regarding the 'font' property, which 
is most important imo.)

>>Designers could actually use the font without creating 
>>accessibility/usability problems.
> 
> Designers don't create accessibility/usability problems with fonts. You
> maybe mix it up with faulty implementations (e.g. the 'IE and pixel' tale)
> and/or refer to Wingdings and friends.

Again, you misread the point I was trying to make and respond to 
something that is not really important.

>>When a browser comes across that would happen to support 'Verdana' it 
>>will load that font, if a browser doesn't found that font in its font 
>>database it will go the next "serie" and tries to load the font 
>>specified there [...].
> 
> And what does an UA do encountering an
> 
>     font-family: 'trebuchet ms', arial, helvetica, sans-serif;
> 
> declaration...?

The same as now described in CSS 2.1 and as described by me. If a 
browser doesn't support 'trebuchet ms' it will go to the next "serie" 
and will try to load it, in this case 'arial'. Et cetera.

>>   color:change-red-amount(255),red;
>>
>>CSS3 parsers would ignore the first value and give the element a red 
>>color. Future parser may support the first value and will change only 
>>the amount of red in the color and ignore the last property.
> 
> And that story is everything but compatible. Remember 'A user agent must
> ignore a declaration with an invalid property name or an invalid value' [2]?
> And current CSS implementations ain't that bad that they'd ignore this rule.

Well, you could write down:

   color:red;
   color:change-red-amount(255),red;

If you want to be compatible with exesting parsers. ('color' is probably 
  not a very good example, since it can only contain a single value.)

> [1] http://www.w3.org/TR/CSS21/fonts.html#font-family-prop
> [2] http://www.w3.org/TR/CSS21/syndata.html#declaration


-- 
  Anne van Kesteren
  <http://annevankesteren.nl/>
Received on Tuesday, 13 April 2004 07:05:26 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 27 April 2009 13:54:29 GMT