www.gnupg.org - new design

[Michael Nahrath]

Lorenzo Cappelletti <l.cappelletti@mail.com> schrieb am 2002-12-04 17:02
Uhr:

>> While taking a look at the code ("who did they do this ..." ;-) my edito=
r
>> printed out an error report. Maybe you are interested.

> :..-(  But compatibility is one of my primary concern.

I guess none of my proposal would rise any compatibility problems, they'd
rather avoid some.

> I'll keep your
> very detailed report for future reference, when I fix all those bugs.
> I hope you'll be available for consults.. ;-)

I couldn't resist (again ...) and just subscribed to this list.
=20
>> <meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3Dutf-8" />
>> is only usefull (if ever ...) if it is used as early in the code as
>> possible. It should be the very first child of 'body'.
>=20
> I didn't know this...

Precisely spoken it should occure before the first non-US-ASCII character i=
n
the code.

If you use=20
    <meta name=3D"keywords" content=3D"=DCberwachung, W=E4chter, ...">
or
    <title>GPG-Schl=FCssel</title>
it has to stand before those elements.

At the moment it is overwritten by the server sending another Charset value
anyway.

You can see this for yourself if you use Mozoilla's "View Page Info".
=20
>> Ever preferable was to correctly set up the server by putting
>>     AddCharset  utf-8  .html
>> into the .htaccess file for the site.
>=20
> We don't have Apache, rather Caudium (and, to be honest, I've never use
> it).

There must be something similar.

Maybe <http://ppewww.ph.gla.ac.uk/~flavell/charset/checklist.html> and
<http://www.cs.tut.fi/~jkorpela/chars/index.html> will be of some general
usefullness on these topics.
=20
>> Werner uses entities for the German umlauts (=C4 =3D =C4) anyway.
>=20
> It's not Werner who uses entities, but WML (http://www.thewml.org/) that
> translates them (see http://www.gnupg.org/index.wml).

I see. Werner's pieces are written in ISO-8859-1 Encoding and your processo=
r
does the encoding to entities.
=20
>> The <span class=3D"gnu">Gnu</span><span class=3D"pg">PG</span> at the top of
>> each page might be marked up as 'h1' ...
>=20
> I dont' use <h1> for that porpouse but for page title, like "The GNU
> Privacy Guard" in http://www.gnupg.org/.

I'd mark up "GnuPG" as 'h1' and "The GNU Privacy Guard" as 'h2' (CSS would
make the rendering the same it is now), but this is a matter of personal
taste ...
=20

Two more things:
=20
1. I took a glimpse to the list archive and saw the question about the
'lang' attribute in XHTML.

It is _completely_ independent to all character coding issues.

Today graphical browsers hardly make any use of it (the only real world
application I am aware of is the rendering of the 'q' element according to
<http://www.w3.org/TR/html401/struct/text.html#edef-Q> by some not so wide
spread browsers like iCab).
Aural browsers (reading websites) might use it to find the correct
pronounciation for a site, but I am not shure if there are any that do.

<http://www.htmlhelp.org/reference/html40/attrs.html#i18n> is the
recommended documentation.


2. Using unordered lists for the navigation is a very good thing to do.
But currently their rendering is completely different in Internet Explorer
5.2.2 and Mozilla 1.3a (both on MacOS X).
I dealt with this issue some time ago and wrote a short demonstration page:
<http://www.subotnik.net/style/list-box-test.html>

Greeting, Michi