Posted by Jukka K. Korpela on 05/06/06 13:49

arbpen wrote:

> Albert Wiersch wote:
>
>>I'm implementing a check for web-safe typefaces.
- -
> I'm sure this has been discussed here many times.

It has, but Albert Wiersch does not want to participate in discussions.
As you may have noticed, he has, during the last six years or so, posted
_only_ to advertize his phoney validator every now and then. How could
he know what we have been discussing?

Maybe this time he genuinely wants to get free help to add a feechure to
his "validator". This is symptomatic, since it reveals the nature of
what he claims to be added value in his "validator" as compared with
validators: additional checks, which cause error messages and warnings
based on the taste and (mis)conceptions of Albert Wiersch alone.

The very idea of checking for "web-safe" typefaces is rather absurd.
There are _no_ typefaces that are safe in the sense that if you suggest
a typeface, it will always be used. The simple proof of this is that
there are modes of rendering that use no fonts at all. Moreover, no
"web-safe" typefaces are needed. Some typefaces are available more often
than others, but how relevant is that? In any case, I would not rely on
a phoney "validator" give me useful information about that.

Apparently Mr. Wiersch just needs more features into his product and
does not mind their being as useless (or worse) as many of the existing
features.

It _would_ be possible to make useful heuristic checks on typeface
names, though. Some typefaces, most notably Symbol but also Wingdings
fonts, are almost certainly used in an attempt to hack browsers into
rendering characters incorrectly, so it would be quite adequate to issue
a warning. Some typefaces, especially Code2000, look very poor without
font smoothing. More questionably, a warning could be given about
Verdana in copy text (assuming you can automagically distinguish copy
text from headings and special texts). You could also issue a warning
about the use fantasy fonts in copy text. Finally, a checker could check
the font names against a largish list of actually existing fonts and
warn at least about such names that are apparently misspellings, such as
Ariel for Arial...

> Best to offer a list
> of fonts with a generic font as fallback (font-family:arial, helvitca,
> sans-serif), or don't supply a font at all and let the user's brower
> decide.

.... or helvitca for Helvetica. :-) Of course, it would not matter much
here, since browsers that don't have Arial would here ignore helvitca as
unrecognize and fall back to sans-serif, i.e. the browser's default
sans-serif font, which is probably as good as (or better than) Helvetica.

[Back to original message]