Re: [WSG] Page-Break-X

2006-03-17 Thread Rene Saarsoo

Jack Pivac wrote:


on 17/03/06 15:04 R Walker (RMW Web Publishing) said the following:


  Try page-break-inside: avoid; on your divs.


Yeah am using that at the moment but it doesn't seem to do anything...


As much as I have tested, this only seems to work in Opera.

Maybe you should just try to force page-break after every third
office-div (it seems that three of those fit nicely on one page...
especially when you remove the header). Probably adding some
class=third office on the server side and something like:

.third {
page-break-after: always;
}

or this might work even better:

.third + div {
page-break-before: always;
}

--

Rene Saarsoo
**
The discussion list for  http://webstandardsgroup.org/

See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**



[WSG] Some studies about web standards usage (sites of Estonia and W3C members)

2006-03-12 Thread Rene Saarsoo

Hi,

I have made some further studies on Estonian web sites.

Compared to the survey done in August, the number of
valid sites has grown almost 50%.

http://triin.net/2006/03/11/Web_Standards_in_Estonia_vol_3

Also I tested the sites of W3C member organisations, and
found, that compared to the previous investigations (done
four years ago by Marko Karppinen) when only 2% of the
sites were valid, now 17% is.

http://triin.net/2006/03/05/Validating_sites_of_W3C_members

Hope, this information brings smile on your faces :)

--

Rene Saarsoo
**
The discussion list for  http://webstandardsgroup.org/

See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**



Re: [WSG] Some studies about web standards usage (sites of Estonia and W3C members)

2006-03-12 Thread Rene Saarsoo

Rimantas Liubertas wrote:



Quite interesting, but one thing confuses me: no HTML 4.01 Strict?



Only 50 pages used HTML 4.01 Strict (of those 7 were valid).
That's probably because when most people make a move to Strict,
they also move to XHTML.


Lachlan Hunt wrote:


This article states:
 | Character encodings
|
| 111 pages had no encoding specified in HTML. Of course not a
| requirement, but belongs to a good style.
 Wrong.  It is a requirement to specify the character encoding correctly.
It would be useful if your statistics indicated not only what encodings
were used, but whether they were specified in HTTP headers, the meta
element, the XML declaration or a combination of them, incl. how many did
so for each MIME type and how many had conflicting values.


What I meant was - it's not a requirement to specify character encoding
in HTML by using the meta element.

My methodology/script had some shortcomings (which I'll hope to improve
in future) and one of them was, that encoding-information was only
searched from meta-elements.


Lachlan Hunt also wrote:


Regarding the list of DOCTYPEs, it would be useful if you indicated which
HTML ones included the System Identifier so that it is possible to
determine which ones trigger standards mode, almost standards mode or
quirks mode.  For example, in Mozilla, HTML 4.01 Transitional will  
trigger

quirks mode without the SI and almost standards mode with the SI.


And that's another shortcoming - quite little information was gathered  
about

the doctypes, but thanks for pointing this out, I'll hope to improve...

And thanks for the feedback.

--

Rene Saarsoo
**
The discussion list for  http://webstandardsgroup.org/

See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**



Re: [WSG] The dilemma: tabular data with sublevels

2006-01-29 Thread Rene Saarsoo
On Sat, 28 Jan 2006 22:54:07 -, Andreas Boehmer [Addictive Media]  
[EMAIL PROTECTED] wrote:



trthFruits/th...
trthspanFruits  /spanApple/th...
trthspanFruits  /spanOrange/th...
trthspanFruits  /spanLemon/th...


The problem with this solution is that it allows for only two levels of
hierarchy (parents:children = fruit:apples). But the application we are
building is more like a file/folder structure in Windows Explorer - it  
can have an endless number of levels. Every folder can have a

subfolder.


So you need more levels... What if you just use some simple numbering:

trthspan1 /spanFruits/th...
trthspan1.1 /spanApple/th...
trthspan1.1.1 /spanRed apple/th...
trthspan1.1.2 /spanGreen apple/th...
trthspan1.1.2.1 /spanExtra green apple/th...
trthspan1.1.2.2 /spanMedium green apple/th...
trthspan1.2 /spanOrange/th...
trthspan1.3 /spanLemon/th...

This should make the hierarchy pretty clean to both screenreader and
text-browser users.

There is one usability problem still: the contents of th elements
are centered by default in most browsers, making the table look like
this:

1 Fruits  Add Edit Delete
   1.1 Apple  Add Edit Delete
1.1.1 Red apple   Add Edit Delete
   1.1.2 Green apple  Add Edit Delete
   1.1.2.1 Extra green apple  Add Edit Delete
   1.1.2.2 Medium green apple Add Edit Delete
   1.2 Orange Add Edit Delete
   1.3 Lemon  Add Edit Delete

Not very comprehensible... The solution would be to use td instead:

   1 Fruits   Add Edit Delete
   1.1 Apple  Add Edit Delete
   1.1.1 Red appleAdd Edit Delete
   1.1.2 Green apple  Add Edit Delete
   1.1.2.1 Extra green apple  Add Edit Delete
   1.1.2.2 Medium green apple Add Edit Delete
   1.2 Orange Add Edit Delete
   1.3 Lemon  Add Edit Delete

Maybe it makes the table a bit harder to understand for screenreader
users, but it surely adds clarity to all the other users.

Hopefully it doesn't make me look like someone who is very into
table-based solutions :)


Rene Saarsoo
**
The discussion list for  http://webstandardsgroup.org/

See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**



Re: [WSG] The dilemma: tabular data with sublevels

2006-01-29 Thread Rene Saarsoo
On Sun, 29 Jan 2006 11:04:07 -, Joshua Street [EMAIL PROTECTED]  
wrote:



Not very comprehensible... The solution would be to use td instead:


That's what we have stylesheets for. It's trivial to change the text
alignment of TH elements :-) TH strikes me as being semantically more
sound, particularly if you give them ID's and set the headers
attribute in your corresponding TD cells accordingly.


The whole point was to make it more usable in browsers with no CSS support.

But you can also give ID-s to TD elements and reference those with headers
attribute:

tr
  td id=header-1span1 /spanFruits/td
  td headers=header-1Add/td
  td headers=header-1Delete/td
  td headers=header-1Edit/td
/tr

Actually, according to HTML 4.01 secification, using TD for headers is
quite OK (http://www.w3.org/TR/html4/struct/tables.html#h-11.2.6):

  TH is for headers, TD for data, but for cells acting as both use TD


Rene
**
The discussion list for  http://webstandardsgroup.org/

See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**



Re: [WSG] The dilemma: tabular data with sublevels

2006-01-28 Thread Rene Saarsoo

Well... I agree, that the proposed markup-structure would be
semantically most correct:



ol
  liItem 1
ul
  liAdd/li
  liEdit/li
  liDelete/li
/ul
ol
  liItem 1.1
ul
  liAdd/li
  liEdit/li
  liDelete/li
/ul
  /li
/ol
  /li
/ol



But can you imagine working with that sort of list in a browser
where stylesheets aren't available? For example in Lynx it would
look something like the following:

1. Fruits
  + Add
  + Edit
  + Delete
 1. Apple
   o Add
   o Edit
   o Delete
 2. Orange
   o Add
   o Edit
   o Delete
 3. Lemon
   o Add
   o Edit
   o Delete
2. Vegetables
  + Add
  + Edit
  + Delete
 1. Carrot
   o Add
   o Edit
   o Delete

If this web page is meant to be used mainly for adding, editing
and deleting, then it's one quite uncomfortable user experience
at least. (There are 6 items in the list above - imagine 60!!!)

Whats so wrong with using good-old table... (skipped summary attribute
and possibly more, that should be added):

table
trthFruits/th tdAdd/td tdEdit/td tdDelete/td/tr
trthApple/th  tdAdd/td tdEdit/td tdDelete/td/tr
trthOrange/th tdAdd/td tdEdit/td tdDelete/td/tr
trthLemon/th  tdAdd/td tdEdit/td tdDelete/td/tr
trthVegetables/th tdAdd/td tdEdit/td tdDelete/td/tr
trthCarrot/th tdAdd/td tdEdit/td tdDelete/td/tr
/table


Well you need to maintain the hierarchy... so... what if you just add
some extra hierarchical information:

trthFruits/th...
trthspanFruits  /spanApple/th...
trthspanFruits  /spanOrange/th...
trthspanFruits  /spanLemon/th...

The result in Lynx would:

 FruitsAdd Edit Delete
 Fruits  AppleAdd Edit Delete
 Fruits  Orange   Add Edit Delete
 Fruits  LemonAdd Edit Delete
   Vegetables  Add Edit Delete
   Vegetables  Carrot Add Edit Delete


Well.. at least what I think.


Rene Saarsoo



**
The discussion list for  http://webstandardsgroup.org/

See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**



Re: [WSG] quot; or in copy?

2005-02-16 Thread Rene Saarsoo
On Thu, 17 Feb 2005 09:16:09 +1100, Dmitry Baranovskiy  
[EMAIL PROTECTED] wrote:

Actually  is an inch symbol. For quotes we should use #147; and
#148; in normal text.
And when you have already this far, I would say, *free your mind from  
eight bit*
and use unicode :) (Takes less space and is better to read then  
#147;entity
references#148;, not to mention all the other benefits...)

-
Rene Saarsoo
**
The discussion list for  http://webstandardsgroup.org/
See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**


Re: [WSG] Standard this

2005-02-16 Thread Rene Saarsoo
On Wed, 16 Feb 2005 23:13:46 +0100, Andy Pieters [EMAIL PROTECTED]  
wrote:

Why not include a 'browser-build' selector in css?
And then some browsers will start to offer option:
Change browser-build information
...and it would fail as any other browser-detection mechanism
has failed. Except the hacks -- there are no good reasons to
simulate some odd behaivior of browser X in browser Y. And also
it's not that simple too.

Rene Saarsoo
**
The discussion list for  http://webstandardsgroup.org/
See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**


Re: [WSG] Help on file extensions used

2005-01-27 Thread Rene Saarsoo
Hi,
I think the only advantages of these extensions are dis-advantages.
The must-read on this topic is from Tim Berners-Lee Cool URIs don't  
change:

  http://www.w3.org/Provider/Style/URI.html
If you must use any extension at all - please consider something that
anyone finds meaningful. The strange extensions may give maybe some
security through obscurity, but in longer term you will curse them.
Rene Saarsoo
**
The discussion list for  http://webstandardsgroup.org/
See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**


Re: [WSG] how to mark up h3 header for list

2005-01-27 Thread Rene Saarsoo
Why not use a definition list:
dl
  dtHeading/dt
  dd
ul
liitem1/li
liitem2/li
liitem3/li
/ul
  /dd
/dl
Using some Hx for list heading when there actually doesn't exist a level
for that heading doesn't make much sense... just pollutes the site with
strange headings imho.

Rene Saarsoo
**
The discussion list for  http://webstandardsgroup.org/
See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**


Re: [WSG] Popups

2005-01-13 Thread Rene Saarsoo
Hi,
I think there are two main possibilities, both quite equal:
[1] Information on the same page
In that case you have link to the helping paragraph on the same page.
The onclick and other event handlers are added to the links with
javascript and also the help section (probably at the bottom of the
page) is being hidden only when scripting is available.
a href=#foo title=More info on foo onclick=display(foo);help/a
...
dl
dt id=foofoo/dt
ddfoo is actually bar.../dd
...
/dl
...
!-- at the bottom of page goes code
 which generates the event handlers
 and hides help section. --
script type=text/javascript
!--
initalize();
// --
/scrip
[2] Information on external page
This time you add links to external html files and the same way add
event handlers only if scripting is available.
a href=foo.html title=More info on foo  
onclick=popup(foo.html);help/a

It probably depends on the amount of help text you need to show.
With method #2 you don't bloat your page with large help section,
which might not even be used, but the savings might be minor if
there's only a little amount of help available.
Hope it provided some help,
Rene
**
The discussion list for  http://webstandardsgroup.org/
See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**


Re: [WSG] when to post, was - asking a PC user for a page check

2005-01-06 Thread Rene Saarsoo
Hi,
I just really have to say: don't ask for PC users if you
really mean Windows users. I always check that kind of
posts just to find out they are talking about IE, which
does not run on my system (which still is PC).
And usually you are interested from software anyway -
doesn't really matter if I run Windows on PC or
emulate it on Macintosh or opposite way around...
regards,
Rene
PS. The page looks fine in Konqueror, except the small font,
but this is always the problem when font size is specified
relatively. The lines can become quite long also (why not
specify max-width at least)
And it looks like there is an error in your CSS-file:
#contents   
{
margin-top: 1.2em;;
margin-right: 10em;
margin-left: 12em;
}
**
The discussion list for  http://webstandardsgroup.org/
See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**


[WSG] Interesting Doctype: anyone seen something like that before?

2005-01-05 Thread Rene Saarsoo
I found a page, I tried to validate it (like I always do 8-) )
and suddenly there appeared an error message I had never seen before,
but more interesting then the error message was the code that
had caused it:
!DOCTYPE HTML generated by Dreamweaver, does not conform to W3C open  
standards

OK, they don't want to follow standards, that's okay with me, but
blaiming Dreamweaver for that... hey, isn't this the program
that many Web Standards activists use?
OK, they don't want to be so accessible to all those blind or
whatever users, that's ok with me too, but this is a homepage
of an health-care organisation... 8-o
The page is:
http://www.medicum.ee/
Rene.
PS. I'm probably going to send them some sort of note about this
crazyness.
**
The discussion list for  http://webstandardsgroup.org/
See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**


Re: [WSG] Slow loading of CSS

2005-01-03 Thread Rene Saarsoo
Hi,
Interestingly I saw the FOUC for about a minute before the style appeared,
and it was a good thing, because the site was fully usable, so I read it  
through
before any styles were applied :) I have mostly encountered this with
cssZenGarden (when the stylesheet is fetched from some far-far-away  
server),
althought there it's not so usuful thing, because I already know the  
content :P

Using Opera,
Rene
**
The discussion list for  http://webstandardsgroup.org/
See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**


Re: [WSG] Funky tags or XHTML2 Transistional

2005-01-01 Thread Rene Saarsoo
On Fri, 31 Dec 2004 15:36:58 -, Kornel Lesinski [EMAIL PROTECTED]  
wrote:

I'm thinking about creating my own doctype/DTD and adding some new tags.
I'm far from being an expert on this subject, but here's what I think:
First of all, current browsers don't really use the Document Type  
Definitions
provided by the W3C website, they just look for the correct
Document Type Declarations. So, they probably will recognise your
DT Declaration as another junk and will probably fall back to the
quirksmode (I wouldn't certanly want that).

As you are refering to XHTML2, I must note that this should be served
as application/xhtml+xml which isn't accepted by IE - so that's
clearly off.
And if you are using some other elements than those in the specification,
or defining your own DT Definitions, then clearly you aren't following
the standards - but trieng to come up with one by yourself.
I think the right way to define your own tags is to use XML - which is
ment for it. And then using some XSL Transformations or some other
server-side technology to transform your custom markup to correct XHTML.
Best wishes, and Happy New Year,
Rene Saarsoo
**
The discussion list for  http://webstandardsgroup.org/
See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**


Re: [WSG] putting code into a web page

2004-12-27 Thread Rene Saarsoo
Well, code comes especially useful compared to pre when you
need to reference some code inline. For example:
  pUse codelen(varA/var);/code do determine the
  size of an array varA/var./p
Like you see, I have used the var element to denote variables -
actually I've never seen this being done (usually sites who
offer colored source code just use font-tags to acieve the effect).
pre is for any kind of preformatted text, but it is
owerwhelmingly used for computer code. The precode combination
is probably the best you can get.
Althought you can make code to behave like pre with CSS,
it wouldn't be wise, because non-css browsers would see a mess
instead of preformatted text.
There is also the samp element, which can be used for sample
output of programs, like:
  pre
  [EMAIL PROTECTED] $ uptime
  samp01:18:12 up 11:01,  1 user,  load average: 0.81, 0.59, 0.50/samp
  /pre
question remains... should the uptime-command I gave be marked up as
code because it's the one-line shell-script I wrote, or as
samp, because it's the output of Bash? There probably is no correct
answer - just to show how hard it can be to distinguish between
program and it's output.
Hey, but there's also the kbd element for user-entered text, so maybe:
  pre
  [EMAIL PROTECTED] $ kbduptime/kbd
  samp01:18:12 up 11:01,  1 user,  load average: 0.81, 0.59, 0.50/samp
  /pre
Well, in real life people probably wont bother to parse the shell output
the way I have done - I think pre and samp inside there should be  
enough.

Hopefully you got the idea, that there might be a lot of different  
semantical
meanings inside one pre.

There's one thing bothering me - for some strange reason Opera
renders code by default in almost invisible color (anyone
knows why?)

Rene Saarsoo
**
The discussion list for  http://webstandardsgroup.org/
See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**


Re: [WSG] accessible image form buttons

2004-12-16 Thread Rene Saarsoo
What if at the first place you write in the document:
input type=submit value=Search
Which should be OK for any klient.
And, if the browser happens to support DOM correctly,
you remove it with Javascript and replace with:
a href=javascript:submitForm()Search/a
input type=image src=button.gif alt=Search
**
The discussion list for  http://webstandardsgroup.org/
See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**


Re: [WSG] styling :first-line Pseudo-element

2004-12-16 Thread Rene Saarsoo
On Fri, 17 Dec 2004 09:02:25 +0930, [EMAIL PROTECTED] wrote:
.pmi p:first-line {
font-size: 1.2em;
}
this should be:
p.pmi:first-line {
  font-size: 1.2em;
}
But using the 'br' isn't any good too. Maybe this line with
the br should be instead a heading followed with a paragraph?
Rene
**
The discussion list for  http://webstandardsgroup.org/
See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**


Re: [WSG] Visual rendering in gecko with app/xhtml

2004-12-15 Thread Rene Saarsoo
On Thu, 16 Dec 2004 10:40:49 +1100, Lindsay Evans [EMAIL PROTECTED]  
wrote:

Your two example pages look identical to me.
Running Firefox 1.0 on Windows XP
No difference in my Linux version of Firefox 1.0 too...
**
The discussion list for  http://webstandardsgroup.org/
See http://webstandardsgroup.org/mail/guidelines.cfm
for some hints on posting to the list  getting help
**