francky wrote:
>[EMAIL PROTECTED] wrote:
>
>
>>I found this as an example on one of the font specification pages
>>referenced via the font thread email. This would be useful, but appears
>>to be only implemented in CSS2. As a designer or user how do you
>>access/force CSS2? Is it even viable at this time?
>>
>>h2 { font-stretch: expanded; }
>>
>>
>The w3c site says:
>[...]
>I am afraid you cannot force a browser to support a non-supported
>property.
>[...]
>
>
Then [EMAIL PROTECTED] wrote off list:
What was the expression you used a couple months back? Addicting?
Yes i think so. Yes it is also all very confusing at times. I also
found the font-stretch description on the w3c website - as well as
the ericmeyer page. Maybe I'm asking the wrong questions. These
questions may sound rather elementary, but...
1) Is CSS a function of the browser or OS, meaning that the viewer
is limited by what browser and version is installed on the PC?
2) If so, is there probably to be found a cross-reference chart to
compare?
3) If not, is it a function of the OS, say Linux, Mac, Windows 98-XP?
4) How do you check the version of CSS?
5) Depending on the above, what is the accepted standard to code for?
If you think the answers should go to the group, that is fine.
-Bob
= = = = =
Hi Bob,
I'm not a specialist in semantic definitions, but I'll try to say what I
understand. I send it to the list as well, for additions and to check if
I'm right... ;-)
Ad 1)
I think everything you see on screen, is in some way a function of the
basic instructions of the computer: the OS.
Next "layer" is the browser: kind of function of the OS. The Windows-OS
can be the basic of IE, FF, Opera, and so on. But other OS's need
sometimes adapted browsers or other types of browsers. The Mac OS has
some "Mac-only" browsers, see
http://darrel.knutson.com/mac/www/browsers.html.
The OS is a condition for which (adapted) version of a browser you can
download/install. For instance the W3C browser/editor Amaya has also
different downloads for different versions of an OS (Win98 or WinXP,
..): http://www.w3.org/Amaya/User/BinDist.html
Html and css are "languages": instruction-sets how a browser has to
display a web page.
* If the source code has a <p>, then every browser has to know that
it means a new paragraph is starting. How a browser is making the
input <p> to the output "new paragraph on screen" is something the
browser-program has to do: in case of IE the machine Iexplorer.exe
and all the dll's, client-settings in the registry and so on,
which are connected with the browser program. The browser has to
interpret the html!
* If the css has a style #div2 { width: 76.5%; }, nested in a #div1
{ width: 80.3%; }, the css-rules say what a browser has to do in
order to compute the result for #div2 at a certain available
window-width. It depends on the built-in instructions of the
browser, what is the result. First take the amount of px of the
screen, multiply with 80.3%, rounding off to 1px (or 2, 3, ..
decimals), then multiply with 76.5%, and rounding off to 1px? - Or
first take the amount of px of the screen, multiply with 80.3%,
not rounding off, multiply with 76.5%, and then rounding off to
1px? Or take a non-inheritance as starting point, and just compute
76.5% of the screen width? The result will be different! - And how
rounding off? 840.6px => 840px or 841px?
The browser has to interpret the css too!
* But the css-rules are recommendations, no laws! (Wish they were).
And sometimes indeed the css-rules let it open how to be
interpreted. Anyway, it is not forbidden if a browser doesn't
follow the guidelines; no punishment, no sanctions...
* Especially A Certain Browser is thinking it can have it's own
rules. I call it "systematic errors". But a legion of web makers
have to trick and hack to get workarounds...
* Complicating factor: some web developing programs (no names,
MS-Frontpage! The MS-Word "one click conversion to web page" is
another nightmare) are building beautiful web pages just full of
proprietary codes, working not standards compliant / cross
browser, but only for The Certain One (without mentioning that
they are making IE-only pages)... feeding the upside down
exclamations "Hé, in IE my page is good, FF doesn't display it in
the right way!".
* And sometimes there are "just" browser-bugs: sometimes
incidentally, triggered by a complex "surrounding css" (and
sometimes by nobody-does-know). See for instance the "known bugs
by browser" in PIE: http://www.positioniseverything.net/explorer.html
So: yes, the visitor is dependent on the browser(-interpretations) on
his computer. If the browser is limited (not supporting a certain
css-rule - or not supporting a certain proprietary code made for an
other browser), the visitor is limited too: he/she cannot see on his/her
screen what was the "input" in the source code.
Aside: text readers and other non-visual browsers have their own
conditions to be accessible.
2)
There are quite a lot of browsers and versions, and all have their
peculiarities... So a cross-reference chart, AFAIK, no there isn't (if
there was, I guess it would be a full printed paper of at least
5x20meter in font-size 8pt).
* In a dutch on line css tutorial, for some browsers the support of
each css-property is described (has to be updated though), for
instance:
http://www.handleidinghtml.nl/css/eigenschappen/background/beschrijving.html
(second part of the page: you see much txt, and this is only for 1
property!)
I don't know if there are EN equivalents to this.
But if you want a certain layout, you can find on the web lots of proved
cross browser solutions, tips and tricks.
As in our css-WIKI: http://css-discuss.incutio.com/.
3)
No, css is not a direct "function" of the OS - the outcome of the css
language is dependent of the translating system: the (OS/)browser
(combination).
And of course of the hardware and software chains after that to get the
browser results on screen: a broken video card shows black, a monochrome
monitor doesn't show { color: #FFF0FF; }. ;-)
4)
How do you check the version of CSS?
* If you mean: check if a page you made is compliant to a
css-version, then you can put the page in the css-validator. If
wanted, you can ask the validator to test in a certain "profile"
(css-version): http://jigsaw.w3.org/css-validator/validator-uri.html
* If you mean: check if a certain browser(version) is supporting a
certain css-version, I'm afraid there is no general yes or no. Css
is a rather big set of rules, sometimes a browser is supporting
one property but not the other.
* Sometimes browser manufacturers are mentioning which css version
is supported.
o Opera for instance: "Standards support - Opera prides itself
in supporting all major Web standards currently in use,
including CSS 2.1, XHTML 1.1, HTML 4.01, WML 2.0,
ECMAScript, DOM 2 and SVG 1.1 tiny."
http://www.opera.com/features/
o Or IE6: "Internet Explorer 6 SP1 Features - Provides full
support for Cascading Style Sheets, Level 1 (CSS1)" [quick
boys!!!; francky]
http://www.microsoft.com/windows/ie/ie6/evaluation/features/default.mspx
and "CSS Enhancements in Internet Explorer 6" (the more
words, the less clear...)
http://msdn.microsoft.com/library/default.asp?url=/library/en-us/dnie60/html/cssenhancements.asp
* If all PR-talk is true, can be tested in several test batteries,
for instance the Eric Meyer's test suite you mentioned:
http://www.meyerweb.com/eric/css/tests/css2/
Or the w3c test suite:
http://www.w3.org/Style/CSS/Test/
Or the WaSP "Acid2 Browser Test":
http://webstandards.org/action/acid2/
5)
Depending on the above, what is the accepted standard to code for?
As a rule of thumb, I think css2 is (for a great part) supported by the
most modern browsers. But if you want visitors with older browser
versions too ...
My conclusion:
It is not always easy to develop css-webpages which are friends with all
browsers (unless they are very simple). Browsers (versions) are so
different, that testing a web page in several browsers (/versions) is
needed to see if they are doing what you want.
* Some more over browser types:
http://www.howtocreate.co.uk/why.html
* To see results in browsers you don't have on your system, browser
shots can help:
http://browsershots.org/
And if some browsers don't do what you want: the tips, tricks and hacks
... and this css-list!
If all browsers would be standards compliant, web developing would be
cheaper: less work to do! And I guess the css-d list would have only 3%
of it's volume. :-)
Now the corrections on what I said can come! ;-)
Greetings,
francky
______________________________________________________________________
css-discuss [EMAIL PROTECTED]
http://www.css-discuss.org/mailman/listinfo/css-d
IE7b2 testing hub -- http://css-discuss.incutio.com/?page=IE7
List wiki/FAQ -- http://css-discuss.incutio.com/
Supported by evolt.org -- http://www.evolt.org/help_support_evolt/
