My messy stylesheets just ain't got no style....

On Sat, 04 Oct 2003 13:02:08 GMT, Brian
<usenet1 (AT) mangymutt (DOT) com.invalid-remove-this-part> wrote:

Herbert wrote:
Well Brian, it did occur to me to add "please don't bother commenting
on whether specifying 80% or any particular font is a good idea,
etc.".

Which would have exactly 0 effect. What do you think this is, your
personal help desk?

What on Earth are you beefing about? I'm 100% open to advice. A
help-desk might actually answer the query, not critise the context.

However, I'd hoped follow-ups might be from people who
realised these were examples and might go on to actually provide some
useful information, but I guess it's just too difficult to resist, eh?

I did provide useful information. You're just too thick to realize it.

Actually, I'm startlingly intelligent. You didn't provide info, you
told me what to do. And 100% ignored the actual query.

Unless I'd specified some deviation from 100%, there'd have been no
relative size-difference to describe, now, would there?

problem solved.

Do you actually have a website, Brian?

And I am well aware of the kind of advice given as to
allowing the 'user decide' (which in my experience is almost
invariably provided by people who couldn't construct an engaging
website if they tried).

Yeah, insult the group. That'll get you more help.

It's not an insult, it's e x p e r i e n c e, Brian.

Sites with 100% anything
tend to look like infant's text books.

On your browser, perhaps. (Don't know how to set a font-size?) Not
on mine.

yawn. Yes, I do know how. Just use the one browser do you? Are the
settings at default.

Most real people (as distinct
from the fabled crowd conjured by coding-geeks)

More insults! Don't stop now, you're on a roll.

No, 'coding-geek' isn't an insult (unless you are a coding geek under
the impression you're representative of the world at large and
resistant to acknowledging how anal-retentive one can appear). A
'coding geek' would be someone who endeavours to comprehend and
implement website design within the paramters of - say - the W3C, and
whose efforts signifcantly constrain the more flamboyant tweaking of
the browser-creators. However, the essence of the web is very much
based in freedom, and when a coding geek forgets this, and that people
are in fact messy, they become tedious.

"you're too thick" would be an insult, Brian.

don't give it a
thought provided it's within a - fairly wide - legibility range.

And you determine this how?

E x p e r i e n c e, Brian.

I know you are capable of providing helpful guidance. But come on,
this follow-up was 90% lame. I know this is presented as 'won't
help', but it could equally well be masking 'can't help'.

rotf,l
Was this meant to goad me into helping you break another web site?

Brian, my interest in 'goading' you is 0. Why bother, when you
obviously goad yourself to your heart's content.

Now who wants to be first to say 'don't top post'?

As Jukka said, continue to top-post. Makes it easier to ignore you.

Ooo! Ouch. You've ruined my day, now, Captain Brian, you hero you..

On Sat, 4 Oct 2003 17:29:50 -0400, Stan Brown
<the_stan_brown (AT) fastmail (DOT) fm> wrote:

In article <v3lsnvogcdgb44lnpapjm358p7u81keuj5 (AT) 4ax (DOT) com> in
comp.infosystems.www.authoring.stylesheets, Herbert
doughnut (AT) email (DOT) me.ok> wrote:
Well Brian, it did occur to me to add "please don't bother commenting
on whether specifying 80% or any particular font is a good idea,
etc.". However, I'd hoped follow-ups might be from people who
realised these were examples

Examples of what? If you wrote "80%" rather than "100%" then
presumably you had some reason for doing so. Brian is helping you by
pointing out that it's a bad idea. If you already knew it was a bad
idea, why did you include it? If you didn't know it was a bad idea
then you should thank Brian for pointing it out.

Because, you hasty berk, I was asking about how specified font sizes
relate to one another.

I was trying to find out more about it, ok?

And really, honestly, truly, if you asked your doctor for advice about
bad breath, and he just said "don't wear green socks" without
explanation, what would you think?

I am pessimistically curious as to the verdana verboten.

Then google for it: it's been discussed here numerous times and
there's no reason for us to rehash it again -- particularly with
your attitude.

Oooh, snooty! Look, I don't need or want your 'help'. And you can
stick your exemplary attitude right back up where you produced it.

On Sat, 04 Oct 2003 07:11:55 -0500, kchayka <kcha-un-yka (AT) sihope (DOT) com>
wrote:

Herbert wrote:

Obviously it's very convenient to be able to specify such a lot over
and over just by using <p>Text</p>, and subheadings with <p
class="subHead">Heading</p>.

It would be even more convenient, not to mention symantically more
meaningful, if you used the right markup for the job. Headings deserve
hx> markup, using the appropriate levels.

h1
h2
h3
h3
h2

etc. and style each heading level as you see fit.

You really should learn some HTML before trying CSS.

I know HTML very well, thank you. Enough to know that if you only
want a line break under a heading, rather than a paragraph break, you
don't use <h#></h#>, but have to codge it with font size and maybe
<bold> tags. Which was the salient point of my original post.

You might try being less impertinent.

On Sat, 04 Oct 2003 19:25:05 +0300, Mikko Rantalainen <mira (AT) st (DOT) jyu.fi>
wrote:

Herbert wrote:
I'm still relativey new to stylesheets, so I'm hoping that the way I'm
going about things can be seriously improved upon, i.e . I just
haven't undersood something obvious about the 'cascading' nature of

Umm.. I don't want to sound rude, but you should really focus on
perfecting your HTML markup before rushing to prefecting the rendering
with CSS. In long run that's the best approach even though the learning
curve is rather steep.

Not rude, just somewhat patronising! I can handle it...

I think I can illustrate the nature of the beast with this example,
using just two text styles:


p {font-family: "arial", "verdana","helvetica", sans-serif; font-size:
80%; color=#ff0000; font-weight: normal; text-align: justify;
margin-left: 2px; margin-right: 2px;}

Check out <URL:http://validator.w3.org/>. Even for an example CSS code,
stuff like "color=#ff0000" is unacceptable. Also, if you don't have a
good reason to add a specific declaration you should drop it. For
example, I cannot think of any valid reason for the need of
"font-weight: normal" in the above example. Especially if you target to
fully utilize the "nesting", you should always use bare minimum CSS
styles for each selector.

The = was a typo, brought about by too many years of html hand coding.
Sorry for the hiccup.

Also consider using 'em' unit for all stuff that has anything to do with
text.

Ok... I'll do more reading and think about it...

Obviously it's very convenient to be able to specify such a lot over
and over just by using <p>Text</p>, and subheadings with <p
class="subHead">Heading</p>.

Always, I repeat, always markup headers as headers, not as paragraphs,
divs or spans with some "special" class. CSS is supposed to only *hint*
default presentation and all the content should be fully readable
without any support for CSS.

Yes, I really really really do appreciate that, but in doing so, one
apparently forfeits the ability to follow a heading with text on the
'next line', and this was the nub of my query, to illustrate a
particular aspect of inheritance of text styles which evades me in its
implementation.

Where I find things get awkward is if I occasionally want to have:
#1 the text follow the subheading after only a <br> line-break (i.e.
within the <p></p>),
#2 or the subHead style used within the text paragraph using <span
class="subHead">include</span

If you have markup like:

h1>Heading</h1
p>text text text</p

and you want rendering like

HEADING
text text text

you just apply style
h1 { font-size: 100%; text-transform: uppercase; margin-bottom: 0;
padding-bottom: 0; }
p { margin-top: 0; padding-top: 0; }

Ah!!! At last, something which looks like it could be the information
I was looking for. Thank you. I don't want the uppercase, but the
margin and padding aspects look promising.

If you want rendering like

HEADING text text text

Not really, but I'm reading on...

you shouldn't modify the markup. I repeat, the content is still exactly
the same and there's no reason not to modify anything else but CSS
which controls the presentation. So you just add next rule to above
stylesheet:

h1 { display: run-in; }

Ok... do go on...

The fact that you add this rule with is own selector, after the above
style declarations means that all the old rules are first applied, then
the nesting takes place and this rule appends and overrides (if
required) the properties that I define here. If I don't want all headers
to run-in with the following paragraph, I just use a class for those
headers I want to run in, like this:

h1 class="notice">Run-in header</h1
p>and the text that follows it</p

and you replace the above style with

h1.notice { display: run-in; }

Now, only the 1st level headers with class notice are rendered run-in,
and also those 1st level headers use all the declarations defined for
all 1st level headers (h1) defined above. Again, nesting in the work.

Unfortunately, MSIE or Mozilla do not support display: run-in, so you
need to use some hacks instead if the actual presentation is so
important enough for the extra work; usually float property is abused to
achieve similar effect.

Happy hacking!

Sounds Brilliant! Thank you. Didn't understand it (at first reading)
but will study it until I do.

If only I hadn't had to wade through so much bile from other posters
on the way here. But one decent follow-up is all anyone could want.

Thanks again.

On Sat, 4 Oct 2003 17:26:48 -0400, Stan Brown
<the_stan_brown (AT) fastmail (DOT) fm> wrote:

In article <cmvrnv8asn188uogv08h8satoadb5p0p2b (AT) 4ax (DOT) com> in
comp.infosystems.www.authoring.stylesheets, Herbert
doughnut (AT) email (DOT) me.ok> wrote:
Obviously it's very convenient to be able to specify such a lot over
and over just by using <p>Text</p>, and subheadings with <p
class="subHead">Heading</p>.

Convenient, but IMHO wrong. Subheadings should be specified using
h2> or <h3> or ... or <h6> whatever level is appropriate. Style
_that_; don't create a separate class and apply it to paragraphs.

I appreciate the theory, but was trying to illustrate an area of
ignorance on my part over inherited styles, not to recommend 'faking'
headings!

Remember, you should always write your HTML so that as far as
possible it degrades gracefully if CSS is not supported. That means
headings should be headings, not paragraphs.

Where I find things get awkward is if I occasionally want to have:
#1 the text follow the subheading after only a <br> line-break (i.e.
within the <p></p>),

"Doctor, it hurts when I do this."
"Then don't."

Quoting Tommy Cooper at me, eh?

Seriously, A paragraph of text can't logically be "within" a
heading. Code the heading with its proper h number, and then code
the paragraph with a class that has margin-top:0.

Well, I wouldn't do that, html or css. The 'subheading' aspect really
was more misleading (and inciting) than I anticipated. I could
equally well have said 'penguin'.

Other folks will probably comment on your CSS; in particular it
usually doesn't sense to specify the text margin in pixels.

Well (at risk of incurring the ire of any anti-tables denizens) I
might use it when trying to keep text at a specific distance from
other content in certain cells a borderless table, and it seems to do
what I want just fine. I'll review the options again, but I'm open to
alternatives recommendations.

Thank you for taking the time to respond, Stan. A measured,
informative approach is always appreciated (as much as snappish
"don't do that" arrogance is to be deplored).

On Sun, 05 Oct 2003 13:56:04 +0200, Stephen Poley
<sbpoleySpicedHamTrap (AT) xs4all (DOT) nl> wrote:

On Sat, 04 Oct 2003 06:43:04 +0100, Herbert <doughnut (AT) email (DOT) me.ok
wrote:

Sites with 100% anything
tend to look like infant's text books. Most real people (as distinct
from the fabled crowd conjured by coding-geeks) don't give it a
thought provided it's within a - fairly wide - legibility range.

You appear to be contradicting yourself. Do you think that most people
find their default settings acceptable or not?

Suppose you tell me? On the whole, I do, but I think it's arrogant to
assume that what suits me suits the majority. When I do anything for
public consumption, I can only do what initially appears to me to be
the most attractive presentation, and attend to and try to accomodate
feedback. I certainly wouldn't do something just because the rules
dictate, or insist that my clients accept the results because that's
"the way things should be" (or because "Brian says so" :-) ).

Whatever, like several others, you're missing the point of my original
post, when I made it quite clear I was proposing examples to
illustrate a particular aspect of CSS inheritance I find awkward

FWIW I have used at least ten different versions of at least five
different browsers, and I have yet to encounter one where the default
text looked like "infant's text books". If your browser, on your
monitor, does, then perhaps it's time you learned how to adjust the
browser settings.

And maybe you learned to be less patronising.

I don't think it's appropriate to expect that users should adjust
anything to view any website. But since you come to hark back to
this point, yes, I do think that an information (i.e. text) rich web
page looks pretty ghastly when the font size is 100%. Dismount from
your elevated perspective and imagine printing a novel at the same
relative font-size.

I'm using Agent here, at its default font settings. I reckon the text
is about half the height of 100% arial in IE. And I've been persuaded
against using Times New Roman, though that, I accept, looks reasonably
ok at 100%. However, it seems commonly accepted that the sans-serif
fonts are just a little easier on the eye on screen, and convey a less
archaic impression (I gather that such an aesthetic consideration is a
moot point for people who put coding above content).

I know you are capable of providing helpful guidance. But come on,
this follow-up was 90% lame. I know this is presented as 'won't
help', but it could equally well be masking 'can't help'.

I think others have addressed the problems. In some cases one simply
can't avoid having a font or colour used in several different styles.
Updates then need to be done via a 'global replace' on the CSS file.
Alternatively one could have a look at preprocessing.

I am pessimistically curious as to the verdana verboten.

See http://www.xs4all.nl/~sbpoley/webmatters/verdana.html

Thank you. If Captain Brian had posted this URL instead of being
uninformatively bossy, I'd have understood - and accepted the point -
immediately.

I appreciate the time you've given in writing, Stephen (or at least
most of it ;-) ).

Herbert wrote:

I know HTML very well, thank you. Enough to know that if you only
want a line break under a heading, rather than a paragraph break

A what? Never heard of it. Or when you say "paragraph break" do you mean
margin?

, you don't use <h#></h#>, but have to codge it with font size and maybe
bold> tags.

Or you could just alter the margin-bottom of the heading, and the margin-top
of whatever element follows it.

--
David Dorward http://dorward.me.uk/

Hmmm, can this be the same Stan Brown who actually responded in a
civil - if not particularly revelant - fashion but three minutes
earlier?

On Sat, 4 Oct 2003 17:29:50 -0400, Stan Brown
<the_stan_brown (AT) fastmail (DOT) fm> wrote:

In article <v3lsnvogcdgb44lnpapjm358p7u81keuj5 (AT) 4ax (DOT) com> in
comp.infosystems.www.authoring.stylesheets, Herbert
doughnut (AT) email (DOT) me.ok> wrote:
Well Brian, it did occur to me to add "please don't bother commenting
on whether specifying 80% or any particular font is a good idea,
etc.". However, I'd hoped follow-ups might be from people who
realised these were examples

Examples of what? If you wrote "80%" rather than "100%" then
presumably you had some reason for doing so. Brian is helping you by
pointing out that it's a bad idea. If you already knew it was a bad
idea, why did you include it? If you didn't know it was a bad idea
then you should thank Brian for pointing it out.

I am pessimistically curious as to the verdana verboten.

Then google for it: it's been discussed here numerous times and
there's no reason for us to rehash it again -- particularly with
your attitude.

Top marks to Mikko, who not actually realised that when I gave my post
the subject line "My messy stylesheets just ain't got no style...." I
was acknowledging my shortcomings, but who ALSO appreciated and
responded to my specific query.

But so many examples of self-righteous squawking from others with too
much time on their hands... Will they forgive Mikko, I wonder?

And would it matter if they didn't?


On Sat, 04 Oct 2003 04:11:16 +0100, Herbert <doughnut (AT) email (DOT) me.ok>
wrote:

I'm still relativey new to stylesheets, so I'm hoping that the way I'm
going about things can be seriously improved upon, i.e . I just
haven't undersood something obvious about the 'cascading' nature of
the coding, which I believe concerns the way attributes relate to one
another when 'nested'...

I think I can illustrate the nature of the beast with this example,
using just two text styles:


p {font-family: "arial", "verdana","helvetica", sans-serif; font-size:
80%; color=#ff0000; font-weight: normal; text-align: justify;
margin-left: 2px; margin-right: 2px;}

.subHead {font-family: "arial", "verdana","helvetica", sans-serif;
font-size: 80%; color=#0000ff; font-weight: bold; text-align: left;
margin-left: 2px; margin-right: 2px;}


Obviously it's very convenient to be able to specify such a lot over
and over just by using <p>Text</p>, and subheadings with <p
class="subHead">Heading</p>.

Where I find things get awkward is if I occasionally want to have:
#1 the text follow the subheading after only a <br> line-break (i.e.
within the <p></p>),
#2 or the subHead style used within the text paragraph using <span
class="subHead">include</span

Then, of course, the subHead text will inherit the <p></p> paragraph's
attributes: i.e. it will be smaller and, in the case of the
line-break, will be further inset because the margin will be
incremental, and if it meets the right-hand edge of the allocated area
(say a table cell) will also be justified.


The only ways I can see to get round some of this are:
#1 to have 'relative' and 'absolute' versions of the two specified
styles,
#2 or to use more inline <span style="lots of style specs";></span
specifying.

Of course, both methods would increase the amount of coding
tremendously. The first would require creating many alternative
'relative' styles to match every combination found within the
documents, and the second approach would negate any advantage
conferred by css over html, with html's need to specify fully each and
every occurrence of 'non-standard' text.

Sorry this is long-winded, and if it's not very clear! It's taken me
quite a while just to identify where my inability lies in this regard.

Thanks for reading this far. Comments welcome...

On Sun, 05 Oct 2003 17:44:49 +0100, David Dorward <dorward (AT) yahoo (DOT) com>
wrote:

Herbert wrote:

I know HTML very well, thank you. Enough to know that if you only
want a line break under a heading, rather than a paragraph break

A what? Never heard of it. Or when you say "paragraph break" do you mean
margin?

Maybe it's the wrong term, but I meant the sort of break which follows
a paragraph as in <p></p>, a linebreak being <br>.

, you don't use <h#></h#>, but have to codge it with font size and maybe
bold> tags.

Or you could just alter the margin-bottom of the heading, and the margin-top
of whatever element follows it.

Yes, Mikko just pointed that out, and it is exactly the sort of info I
was looking for, thank you!