Announcement: www.kde.org/testing

[Sebastian Faubel]

> * I guess you already know that the html source doesn't validate ;-)

Of course. It doesn't validate because the RDF's not working, and that's
because i used absolute instead of relational links. The reason why i
used absolute links is because i do not have a RDF file on my local
machine...but that's unimportant. Absolute links are not allowed (or
don't work) in the CVS either....was my mistake.

> * Like Dre said, do not require JavaScript. On a related note, *never* use 
> 'javascript:' or '#' links. (Unless '#' == top of the page, which is 
> allowed).

I never require JavaScript for Links. I was just playing around with it
a bit and didn't deactivate it clean enough, sorry.

> * For people who like to visit the site using telnet and read the source... 
> ;-)
> e.g. <div id="r5c2e7_ann"> isn't clear. What's that? Row 5, column 2, 
> E-blahblah, Ann who???
> Simply use id="announcements" so every potential web designer understands it. 
> Makes life easier for us :-)

Please blame me if you don't think so.

The way how to write the CSS IDs depends on the developer. In my opinion
the E-blablah is better because it gives you a straight order in the CSS
file and much more overview. You can see the relation or position of an
element without needing to see the code. Also it enables a lot more
structured and defined programming. 

That's what makes life much more easier for the maintainer and
I'm not writing the code for others than the maintainers.
For them the naming scheme is pretty much explained in 4 sentences. 

By the way you almost got it! The "e" stands for "Element", the rest is
a short description - perhaps this one was too short ;-)

> * When you can, try to avoid the B and the BR elements. (Sometimes called Bed 
> and BReakfast ;-) They are used for presentational purposes, which is what 
> xhtml & css are trying to avoid. See 
> http://tantek.com/log/2002/10.html#L20021022t1432
> (Is the same reason why TABLE should be avoided too...)
> 
> * Use the CITE element for citations
> http://www.w3.org/TR/1999/REC-html401-19991224/struct/text.html#edef-CITE
> 
> * Also the ACRONYM element would useful
> e.g. explaining what WYSIWYG, CVS, IRC, PIM are...
> 
> * There is a H1 element missing.
> <cite>Headings should reflect the logical structure of the document; they 
> should not be used simply to add emphasis, or to change the font size.</cite>
> Meaning: You should use <h1> tag when your page contains <h2> and <h3> tags.
> When those headers are used in the correct way, you should see a correct 
> "outline" of your page. See:
> http://validator.w3.org/check?uri=http%3A%2F%2Fkde.org%2Ftesting%2F&outline=1#outline

I will consider this in future, thanx!

> * Remark about the scripts:
> Why use
> $rdf_pieces = $news->openRDF("http://www.kde.org/dot/rdfplus");
> in www/testing/media/includes/content.php ?
> Since every mirror contains /dot/rdfplus wouldn't it be better to use that 
> file directly instead of making a connection? Would be much quicker... and 
> makes my local (testing) kde.org mirror work again. (Doesn't work behind my 
> ISP's proxy :-p )

I already mentioned it....

> Well... that's about it, i guess :-)
> I can honestly say this proposal looks *much* better than our current site.

Thanks a lot..:-)

root66