Uncomprehensible css validation warnings

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi all,

I hope this is the right newsgroup for css questions, didn’t find
another one.

I am a bit puzzled by some warnings the css validator gives me.

The validation of my style sheet can be found here:
http://jigsaw.w3.org/css-validator/v...le%2Fstyle.css

I do not understand the warning: you will see that I did give a color
for background in div.contentsBox (taken over from Wikipedia, btw).

Anybody care to explain?

(all other comments welcome too)

H.
- --
Hendrik Maryns
http://tcl.sfs.uni-tuebingen.de/~hendrik/
==================
http://aouw.org
Ask smart questions, get good answers:
http://www.catb.org/~esr/faqs/smart-questions.html
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2 (GNU/Linux)

iD8DBQFFE6/2e+7xMGD3itQRAkSyAJ9OhcnGFhrHfapWzNmiuOaWfh6POgCaA ybc
PrjMrvA6QdqvUy/9N4YToO0=
=eg6K
-----END PGP SIGNATURE-----

Hendrik Maryns

Hendrik Maryns wrote:

> The validation of my style sheet can be found here:

http://jigsaw.w3.org/css-validator/v...le%2Fstyle.css

> I do not understand the warning: you will see that I did give a color
> for background in div.contentsBox (taken over from Wikipedia, btw).

* You have no color with your background-color : div.contentsBox

* div.contentsBox {
[...]
o background-color : #f9f9f9;
}

You nave no _color_ (foreground color) with your background color.
What would happen if the user already had a default color of #f9f9f9 ?
-- your new background would make the text disappear against it.

In general, always set both colours together, using readable
combinations.

The W3C validator is a bit obsessed with this warning, which can be
safely ignored so long as you are setting both colours somewhere
reliable, then just changing one for small "highlights" within this.

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Andy Dingley schreef:
> Hendrik Maryns wrote:
>
>> The validation of my style sheet can be found here:
>
> http://jigsaw.w3.org/css-validator/v...le%2Fstyle.css
>
>> I do not understand the warning: you will see that I did give a color
>> for background in div.contentsBox (taken over from Wikipedia, btw).
>
> * You have no color with your background-color : div.contentsBox
>
> * div.contentsBox {
> [...]
> o background-color : #f9f9f9;
> }
>
> You nave no _color_ (foreground color) with your background color.
> What would happen if the user already had a default color of #f9f9f9 ?
> -- your new background would make the text disappear against it.
>
> In general, always set both colours together, using readable
> combinations.
>
> The W3C validator is a bit obsessed with this warning, which can be
> safely ignored so long as you are setting both colours somewhere
> reliable, then just changing one for small "highlights" within this.

I see, thanks. IOW, I can safely ignore it here. Hm, or not: I didn’t
explicitly define link colors...

H.
- --
Hendrik Maryns
http://tcl.sfs.uni-tuebingen.de/~hendrik/
==================
http://aouw.org
Ask smart questions, get good answers:
http://www.catb.org/~esr/faqs/smart-questions.html
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2 (GNU/Linux)

iD8DBQFFE7XCe+7xMGD3itQRAiseAJ9MydioqYWDv6GjwRR734 3GTZaF7gCfcbyb
M9x76Eo/meXiuek/PlZy5yU=
=I12b
-----END PGP SIGNATURE-----

Andy Dingley wrote:

> The W3C validator is a bit obsessed with this warning,

Isn't it a bit early in the morning for anthropomorphism?

> which can be safely ignored so long as you are setting both colours somewhere
> reliable, then just changing one for small "highlights" within this.

"Somewhere reliable" meaning "in the same block", since you can't know
how your stylesheet is going to interact with browser and (more
especially) user stylesheets.

Hendrik Maryns wrote:

> -----BEGIN PGP SIGNED MESSAGE-----

Don't be ridiculous. This is Usenet.

> I see, thanks.

No, you don't. You indicated lack of comprehensive reading by your
comprehensive quotation of the article you are nominally commenting on.

> IOW, I can safely ignore it here.

If you are going to ignore messages that you don't understand even after
having them explained to you, why do you use the "validator" in the first
place?

Well, Andy's explanation wasn't quite correct, but it surely contained an
important "if", which you most probably ignored.

Start from the "CSS validator" FAQ. It takes a few clicks to find it, but
here's the direct URL: http://www.websitedev.de/css/validator-faq

If problems remain, try reading the archives of
comp.infosystems.www.authoring.stylesheets for longer explanations. Using
some key phrase from the error message might be a good starting point when
using Google Groups.

> iD8DBQFFE7XCe+7xMGD3itQRAiseAJ9MydioqYWDv6GjwRR734 3GTZaF7gCfcbyb

Indeed.

--
Jukka K. Korpela ("Yucca")
http://www.cs.tut.fi/~jkorpela/

Jukka K. Korpela wrote:
> Hendrik Maryns wrote:
>
>> -----BEGIN PGP SIGNED MESSAGE-----
>
> Don't be ridiculous. This is Usenet.

Usenet is a medium where it's often reasonably easy to forge somebody's
identify. A PGP signature can be used to verify your identity.

In this case, it's probably not of particular importance though, as he
would likely have been given the same CSS advice whether or not we were
able to firmly establish who he is.

--
Toby A Inkster BSc (Hons) ARCS
Contact Me ~ http://tobyinkster.co.uk/contact

Toby Inkster wrote:

> Usenet is a medium where it's often reasonably easy to forge
> somebody's identify.

So is the Real World (TM). Sending a signed paper with someone else's name
under it is extremely simple.

> A PGP signature can be used to verify your identity.

That _is_ just ridiculous on Usenet. Nobody ever checks the PGP signature,
they are write-only nonsense on Usenet. If you wanted to check someone's PGP
signature, would you _really_ rely on the result? Why? The signature alone
does not prove anyone's identity the least.

Indirectly, using a PGP signature on Usenet tells that the poster is a PGP
enthusiast who does not know Usenet or does not care about how Usenet works.
Therefore, it may act as a useful warning signal indeed, so we should expect
the poster to be FAQ challenged, too, among other things.

--
Jukka K. Korpela ("Yucca")
http://www.cs.tut.fi/~jkorpela/

Jukka K. Korpela schreef:
> Hendrik Maryns wrote:
>
>> -----BEGIN PGP SIGNED MESSAGE-----
>
> Don't be ridiculous. This is Usenet.

Hm, I read you other post with objections against PGP on Usenet, and am
still very in doubt whether I agree. But I am happy to click on the
little pen in the lower right border for you.

As of that I am ripe for the FAQ, indeed, for HTML and CSS I am, but
please do not tell me I do not know Usenet.

>> I see, thanks.
>
> No, you don't. You indicated lack of comprehensive reading by your
> comprehensive quotation of the article you are nominally commenting on.

Why oh why do people always have to start fighting about this Usenet
stuff like cutting posts, etc. I do not like top-posting myself, and
may be caught commenting on it, but I like to include long citations.

>> IOW, I can safely ignore it here.
>
> If you are going to ignore messages that you don't understand even after
> having them explained to you, why do you use the "validator" in the
> first place?
>
> Well, Andy's explanation wasn't quite correct, but it surely contained
> an important "if", which you most probably ignored.

No, I didn’t really understand it, and concluded from the body of the
explanation that I could ignore it. Others have pointed me out that
that is not a good idea.

> Start from the "CSS validator" FAQ. It takes a few clicks to find it,
> but here's the direct URL: http://www.websitedev.de/css/validator-faq

Thanks.

> If problems remain, try reading the archives of
> comp.infosystems.www.authoring.stylesheets

That was the newsgroup I was looking for! The naming is logical, but I
expected the letters css in it.

for longer explanations.
> Using some key phrase from the error message might be a good starting
> point when using Google Groups.

I’ll remember that.

H.
--
Hendrik Maryns
http://tcl.sfs.uni-tuebingen.de/~hendrik/
==================
http://aouw.org
Ask smart questions, get good answers:
http://www.catb.org/~esr/faqs/smart-questions.html