W3C home > Mailing lists > Public > www-html@w3.org > January 2003

Re: different uses of xhtml

From: (wrong string) äper <christoph.paeper@tu-clausthal.de>
Date: Thu, 16 Jan 2003 07:49:06 +0100
Message-ID: <032901c2bd2b$5dbc4c60$3ef4ae8b@heim4.tuclausthal.de>
To: <www-html@w3.org>

Lorenzo De Tomasi:
>
> Now if I write an xhtml strict doc and I view it with the browser default
> settings the result is horrible: we need better default setting that
people
> like and doesn't need to change.

To build a good browser stylesheet is the job of browser manufacturers.
They're free to follow the informative appendix of HTML4, which has a sample
stylesheet, and mostly do, but are not forced to. This is good, because in
theory it alllows browsers with better default rendering.
In practice web designers would complain about a browser that e.g. used more
than black and white or a different font-family than serif (TNR) for text,
because they would have to declare more CSS than they do now. You can see
how ignorant they are if you set your browser background color to anything
different than white; now disable <body bgcolor>...

> If I try to open an xhtml strict doc with Acrobat Reader it doesn't open:

Of course not. AAR and browsers don't have much in common. Neither do the
formats.

> At least I may want to change the default layout style of the browser

A user stylesheet. Or a selectable user agent stylesheet.

> I think that xhtml2 is walking on the right path introducing elements as
> section, h, p, l, blockquote, quote, cite, code, samp, dfn, address, abbr,
> span, strong, em, var,

Most of these are already in HTML, but rarely used.

> - hire some experts in information architecture, typography and graphic
> design to create a functional default layout and presentational style for
> each xhtml element and default class;

You've seen <http://www.w3.org/StyleSheets/Core/>?

> this default style will be adopted identical by every browser

That would be bad. They're too equal already.

> must have every useful character like euro, ...

That's an issue of the DTD, if there's one, available fonts and (in very
rare cases nowadays) Unicode, but definitely not of default stylesheets. All
characters of ISO-8859-1 and some more currently have a named entity in
HTML.

> - hire some experts of legibility, readability and typography to adopt or
> create very legible and public domain serif and sans serif fonts

That's not that easy or cheap, although I'd welcome the promotion of a well
designed font package covering all of Unicode.

> and distribute with and use it in each browser as default.

You're again wanting to take away the freedom of choice from browser
developers. Who says that there isn't a better font?

> A graphic designer usually knows he can obtain what he wants writing some
> complex (for him) CSS rule but still prefers the easier solution of using
an
> element or an attribute directly in xhtml.

Only very few graphic designers know what HTML is at all. They know
Photoshop instead.
It was a good thing if they were taught CSS in design school, though.

> More the settings are the same identical for each browser and very
> functional, the less I have to personalize to obtain a good site.

The power of CSS is to cascade beautifullly, if all style contributors do
their job right.

> The right solution for the content editor (1) may be that of creating a
list
> of default attributes:
>     * <note class="sideright">This creates a note in a column on the right

This is considered harmful. ;) The class attribute is taken, authors are
free to choose values for it (among the limits CSS sets for class names).
You may introduce a new attribute to tell whether it's a side-, end- or
foot-note, though, but don't call it "class".
Left and right is purely presentational and should be left to CSS.

> 3. Programmers way
> Usually programmers prefer to write complex and precise code because it's
> their job and they understand it.

That's the reason why they (not "programmers" but "coders"/"hackers") should
write the actual code. The content editor provides the [structured] text
[plus informative graphics] in a format he understands (aka: his word
processor produces), the graphics designer provides a [fluid] colored layout
and decorative images either electronically or on paper and the coder throws
this together for publishing. They need to talk a lot of course, else
they'll soon need a manager (who's paid as much as all of them together and
knows something about everything, but nothing about something) to coordinate
them.

>       content: url("note.gif") "Note " counter(note-counter) ":";

That should read

   content: "Note:";
   content: url("note.gif") "Note:";
   content: url("note.gif") "Note " counter(note-counter) ":";

Christoph Päper
Received on Thursday, 16 January 2003 01:48:50 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 March 2012 18:15:54 GMT