By using this site, you agree to our updated Privacy Policy and our Terms of Use. Manage your Cookies Settings.
438,493 Members | 1,936 Online
Bytes IT Community
+ Ask a Question
Need help? Post your question and get tips & solutions from a community of 438,493 IT Pros & Developers. It's quick & easy.

Observe: Who uses CSS?

P: n/a
Is there information content in this observation?

use html tables for layout:

http://www.microsoft.com and http://www.netscape.com

uses styles:

http://www.opera.com

MasonC
Jul 20 '05 #1
Share this Question
Share on Google+
59 Replies


P: n/a
Mason A. Clark wrote:
Is there information content in this observation?

use html tables for layout:
http://www.microsoft.com and http://www.netscape.com

uses styles:
http://www.opera.com


What is your question?

I have browsed for a long period of time with the following local style:
table { border: 1px dotted blue; }
It puts a dotted blue rectangle round every table.

I estimate that at least 95% (I actually believe 99%, but I could never prove
it!) of the web pages (indeed web sites) I see use table-layout. In all the
100s (by now, perhaps 1000s) of news sites I have come across, only 2 (+)
don't use table layout. (Wired, C|News, some ESPN). Nearly all news sites do,
nearly all academic sites do, nearly all government sites do, nearly all major
business sites do. The vast majority of private sites do. *Worldwide*. For the
last *several years*. Probably between 100,00 and 1 million new table-layout
pages each day. The vast majority of the world's major new global information
repository! On and on, year after year!

This isn't a problem. Simple layout tables are very effective. They can be
accessible, efficient, effective, standards-compliant, future-proof, very
flexible, and adaptable to content and viewport width. Because they are
standards-compliant, and represent the dominant layout-technique on the
planet, new browsers will obviously try to support them well, and so reduce
the need to switch to other methods. (I assume no one here wants new new
browsers *not* to be standards-compliant!)

Right from late 1993, tables were explicitly intended to layout complex
material in horizontal and vertical manner. Tables cells were, from that time,
intended to include text, headers, paragraphs, and lists; and nested tables
followed soon after. There has never, as far as I know, been a credible
intention to restrict the contents of tables to simpler material, such as you
would find in a spreadsheet. The latest XHTML 2.0 proposals continue to allow
cells to be complex mixtures of headers, block-level elements, and inline
material. Therefore these will be valid on the web for decades.

I have written, in some *tableless-layout pages*, 5 articles on this theme:
"Layout tables considered valuable":
http://www.barry.pearson.name/articles/layout_tables/

--
Barry Pearson
http://www.Barry.Pearson.name/photography/
http://www.BirdsAndAnimals.info/
http://www.ChildSupportAnalysis.co.uk/
Jul 20 '05 #2

P: n/a
Barry, At your web page
http://www.barry.pearson.name/articles/layout_tables/
you mention that CSS positioning will not be effective for several years
when MS finally gets their act together.

So, why is everybody still using IE (which is very old on other support -
javascript limitied to 1.2, etc.) when they could be using Mozilla which has
the most complete current support of all of W3C's standards?

Don't limit web pages because a $$$ company can't keep up, tell everyone who
visits your sites to switch (at no cost - ever) and get far superior results
and far fewer security issues!

John
---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.560 / Virus Database: 352 - Release Date: 1/8/2004
Jul 20 '05 #3

P: n/a
John Bowling wrote:
Barry, At your web page
http://www.barry.pearson.name/articles/layout_tables/
you mention that CSS positioning will not be effective for several
years when MS finally gets their act together.

So, why is everybody still using IE (which is very old on other
support - javascript limitied to 1.2, etc.) when they could be using
Mozilla which has the most complete current support of all of W3C's
standards?

Don't limit web pages because a $$$ company can't keep up, tell
everyone who visits your sites to switch (at no cost - ever) and get
far superior results and far fewer security issues!


Unfortunately, it is not even that much of Microsoft's fault as it is of the
average users out there. Most people use Windows, and on most Windows
platforms IE comes preinstalled.

Most users expect Web to work like a TV: if one "program" (website) can't be
seen right, and others are OK, than that one is "broken". And since most are
OK, they just ignore the "broken" site and move on.

That is why most of us can't afford to dismiss such casual visitors.

Berislav
Jul 20 '05 #4

P: n/a
Barry Pearson wrote:
Mason A. Clark wrote:
Is there information content in this observation?

use html tables for layout:
http://www.microsoft.com and http://www.netscape.com

uses styles:
http://www.opera.com
What is your question?

I have browsed for a long period of time with the following local
style: table { border: 1px dotted blue; }
It puts a dotted blue rectangle round every table.

I estimate that at least 95% (I actually believe 99%, but I could
never prove it!) of the web pages (indeed web sites) I see use
table-layout.


I would say the figure is around 99.99%

This isn't a problem.


Have you tried printing those pages? Endless navigation garbage and the
content pushed out of the right edge? Happens.
Have you tried browsing with your hand phone display? Makes the Web
completely useless because of table-layouty schemes (table-layout is
not the only problem, but a big part of it, e.g. navigation usually
comes left and thus must come first if you linearize it -- not so in
CSS).

--
Google Blogoscoped
http://blog.outer-court.com
Jul 20 '05 #5

P: n/a
Berislav Lopac wrote:
Unfortunately, it is not even that much of Microsoft's fault as it is of the
average users out there. Most people use Windows, and on most Windows
platforms IE comes preinstalled.


http://www.webstandards.org/buzz/arc...1.html#a000278

--
Stanimir
Jul 20 '05 #6

P: n/a
> Unfortunately, it is not even that much of Microsoft's fault as it is of the
average users out there. Most people use Windows, and on most Windows
platforms IE comes preinstalled.


Apparently, new versions of IE will not be freely available for
download, but will only be available with new versions of Windows -
surely good news for mozilla and opera and others. If people have to
have the latest OS to get the latest IE, then they are more likely to
swap to another browser.
Jul 20 '05 #7

P: n/a
*Ian Watts*:

Apparently, new versions of IE will not be freely available for
download, but will only be available with new versions of Windows -
surely good news for mozilla and opera and others. If people have to
have the latest OS to get the latest IE, then they are more likely to
swap to another browser.


Wish-thinking. The majority of users still use the browser that came with
their OS and they use the OS that came with their computer. New computers
come with the most recent OS with the most recent browser. The majority of
computers sold are new and the total number is constantly growing, i.e. the
growing percentage of IE6 is (almost) not based on people downloading and
installing it, but by people buying new computers with Windows XP
preïnstalled, then there are some who accidently get it by installing their
ISP's required software, i.e. inserting the CD into the drive and clicking
OK several times as they were told.
For that reason IE7 will have almost the same adoption rate as IE6, at most
two years after the Longhorn launch it will be the most used browser if
nothing serious happens until then (e.g. in court).

That's however off topic in ciwas.
F'up2 poster

--
"When a thing has been said and well, have no scruple. Take it and copy it."
Anatole France
Jul 20 '05 #8

P: n/a
Mason A. Clark <ma*******@THISix.netcom.comQ> wrote in message news:<03********************************@4ax.com>. ..
Is there information content in this observation?

use html tables for layout:

http://www.microsoft.com and http://www.netscape.com

uses styles:

http://www.opera.com

MasonC


Maybe the first two have more connection to the real world?
I don't want to be inflammatory, nor do I want to even seem for an
instant to approve of M$'s hostility to the spirit of standards. What
I mean is that CSS has been botched.
In the early days of the web, when frames appeared, you could learn in
about an hour how to do web-site layout with frames. Of course, people
eventually realized that frames cause a number of problems. So a lot
of those web sites converted to frameless layouts using tables. Again,
tables have a clear conceptual model as a layout tool. You could pick
it up in an hour or less.

But converting those sites to use CSS is a time-consuming nightmare.
There are lots of web sites out there which explain how to use CSS for
2-column layouts ... how to make the columns look the same length ...
how to create footers ... etc. But the very proliferation of such
sites is an indication that something is wrong. Tasks that were
trivial with frames or tables (because of their clear conceptual
model) are difficult with CSS.

This indicates that the people who created CSS botched the job. CSS is
not being accepted, and it's not because the browser implementors
aren't implementing it (on the whole they are doing a good job), it's
because CSS is no good. Separating layout from content is a good idea
- but it has not been done right in CSS.
Jul 20 '05 #9

P: n/a
On 26 Jan 2004 14:25:02 -0800, Retlak <re****@go.com> wrote:

But converting those sites to use CSS is a time-consuming nightmare.
There are lots of web sites out there which explain how to use CSS for
2-column layouts ... how to make the columns look the same length ...
how to create footers ... etc. But the very proliferation of such
sites is an indication that something is wrong. Tasks that were
trivial with frames or tables (because of their clear conceptual
model) are difficult with CSS.

This indicates that the people who created CSS botched the job. CSS is
not being accepted, and it's not because the browser implementors
aren't implementing it (on the whole they are doing a good job), it's
because CSS is no good. Separating layout from content is a good idea
- but it has not been done right in CSS.


I think a large part of the problem is the misconception that CSS is a
drop-in replacement for tabular layout.

The misuse of the table markup has led to some rather attractive and
popular designs. Naturally, the web author wants to accomplish this type
of layout with CSS, and there are those who mislead by saying "you can do
anything with CSS." So the author half-learns CSS, cannot accomplish the
identical table layout, and gives up.

Facts are facts. CSS is the correct way to do layout, not tables. Table
markup must be reserved for tabular data, lest we lose the meaning of that
markup entirely. And anything worth doing (including CSS) is worth taking
enough time and experimentation with to eventually make it possible to
work with it fluently.

Over time students of CSS can see that it's totally different from table
layout, and works nothing like it, but the resulting product is far
superior in many ways. And they see that the CSS model is not a
replacement for the table layout model, but the first really correct way
to do the job of layout where tables were misused to get a layout "effect".

Yes, I agree that the learning curve for CSS is steep, and that has an
effect. But the learning curve for Javascript seems steep as well, and we
see it used all over. Somehow I can't accept the learning curve as being a
primary reason for the problem. A CSS-Lite might be a worthwhile
objective, perhaps...
Jul 20 '05 #10

P: n/a
Neal <ne*****@spamrcn.com> wrote in news:op**************@news.rcn.com:
Yes, I agree that the learning curve for CSS is steep, and that has an
effect. But the learning curve for Javascript seems steep as well, and
we see it used all over. Somehow I can't accept the learning curve as
being a primary reason for the problem. A CSS-Lite might be a
worthwhile objective, perhaps...


I suspect that the learning curve for CSS is actually a lot less steep for
people who are learning it as the first way to do layout than for people
who are already used to using tables for layout. Creating a new mental
model from scratch is generally much easier (though still hard) than trying
to modify an existing, partially incorrect mental model (the easiest of all
is to incorporate new, compatible details into an existing correct mental
model; in fact that's what we're usually doing when we say that something
is "intuitive").

As for Javascript, I'm afraid that the majority of Javascript out on the
web was copied without understanding rather than written from scratch, so
the learning curve was probably seldom entered. CSS most definitely can't
be used in that fashion.
Jul 20 '05 #11

P: n/a
It seems "Neal" wrote in comp.infosystems.www.authoring.stylesheets:
Yes, I agree that the learning curve for CSS is steep, and that has an
effect


The problem is that there's no reliable documentation. There's good
documentation of how CSS _should_ work, but not of how it _does_
work. As a PP said, the very fact that there are so many sites
giving conflicting advice on something as simple as a two-column
layout shows that there's a problem.

I've been struggling with this specific problem, occasionally, for
more than a year. Writing a CSS layout for my site map
http://oakroadsystems.com/sitemap.htm
that works in reasonably standard browsers is trivial. Writing one
that doesn't break badly in various (mutually incompatible) versions
of MSIE is a nightmare, and I've failed to do it. I came close with
http://oakroadsystems.com/sitemap2.htm
but MSIE 4 screws it up, catastrophically; see the screen shot at
http://oakroadsystems.com/ie4.gif
(Solutions are welcome!)

Even something as simple as centering a table is a major pain in the
a-- (or a--- for our British friends :-). Look at
http://www.acad.sunytccc.edu/instruct/sbrown/
in almost any browser, including IE4 and IE6, and you'll see the
table is centered. Now look at it in IE 5 for Windows and you'll see
the table is sitting at the left. I mean, I ask you!

--
Stan Brown, Oak Road Systems, Cortland County, New York, USA
http://OakRoadSystems.com/
HTML 4.01 spec: http://www.w3.org/TR/html401/
validator: http://validator.w3.org/
CSS 2 spec: http://www.w3.org/TR/REC-CSS2/
2.1 changes: http://www.w3.org/TR/CSS21/changes.html
validator: http://jigsaw.w3.org/css-validator/
Jul 20 '05 #12

P: n/a
Neal <ne*****@spamrcn.com> wrote in message news:<op**************@news.rcn.com>...
On 26 Jan 2004 14:25:02 -0800, Retlak <re****@go.com> wrote:

But converting those sites to use CSS is a time-consuming nightmare.
There are lots of web sites out there which explain how to use CSS for
2-column layouts ... how to make the columns look the same length ...
how to create footers ... etc. But the very proliferation of such
sites is an indication that something is wrong. Tasks that were
trivial with frames or tables (because of their clear conceptual
model) are difficult with CSS.

This indicates that the people who created CSS botched the job. CSS is
not being accepted, and it's not because the browser implementors
aren't implementing it (on the whole they are doing a good job), it's
because CSS is no good. Separating layout from content is a good idea
- but it has not been done right in CSS.
I think a large part of the problem is the misconception that CSS is a
drop-in replacement for tabular layout.


I expect CSS to provide the capability to structure the visual
appearance of a page. If I want my page to have a header bar at the
top, a menu down the left hand side, and a footer bar at the bottom,
the whole thing being nicely centered, I should be able to do it
easily in CSS. This is a *very simple* layout.

Facts are facts. CSS is the correct way to do layout, not tables.
Of course. Separating layout from meaning is the whole point of XML
and CSS.
Table
markup must be reserved for tabular data, lest we lose the meaning of that
markup entirely. And anything worth doing (including CSS) is worth taking
enough time and experimentation with to eventually make it possible to
work with it fluently.
Well, it depends on *how much* time. I remember spending about an hour
getting up to speed with frames, years and years ago. Once you'd got
the concept, you could accomplish pretty much anything with nested
framesets. Same for table-based layout. The exact same brain (mine)
has spent about a week trying to understand CSS well enough to do
simple layouts. Yes, I'd accept a longer learning period for a better
tool, but not 10 times longer!
Over time students of CSS can see that it's totally different from table
layout, and works nothing like it, but the resulting product is far
superior in many ways.
The notion of separating layout from semantics is superior to the de
facto conceptual model of HTML, which tangles them.

But you are confusing that statement with the specific means of
accomplishing it that is CSS. The goal is worthy and good. CSS does
not seem to be the right way to accomplish it
Yes, I agree that the learning curve for CSS is steep, and that has an
effect. But the learning curve for Javascript seems steep as well, and we
see it used all over. Somehow I can't accept the learning curve as being a
primary reason for the problem. A CSS-Lite might be a worthwhile
objective, perhaps...


No. Hell, no. The *very last thing* anybody needs is dialects and
subsets.
We need a layout mechanism that has a clear set of concepts underlying
it, not a mishmash of features. Any reasonably competent person should
be able to absorb the key concepts of how to do layout in an hour or
two. It should then be clear, in outline, how to accomplish any
layout. Getting the detail right will involve working with a reference
manual or the specification. But the principles should be clear.
I'm not sure why you bring Javascript in here; there's a lot of detail
work in using Javascript well, but the underlying concepts are clear,
unlike CSS.
Jul 20 '05 #13

P: n/a
On 26 Jan 2004 23:26:19 -0800, re****@go.com (Retlak) wrote:
I'm not sure why you bring Javascript in here; there's a lot of detail
work in using Javascript well, but the underlying concepts are clear,
unlike CSS.


Clear to you, perhaps. Most definitely not clear to the majority of the
people who use it.

--
Stephen Poley

http://www.xs4all.nl/~sbpoley/webmatters/
Jul 20 '05 #14

P: n/a
Stephen Poley wrote:
On 26 Jan 2004 23:26:19 -0800, re****@go.com (Retlak) wrote:

I'm not sure why you bring Javascript in here; there's a lot of detail
work in using Javascript well, but the underlying concepts are clear,
unlike CSS.

Clear to you, perhaps. Most definitely not clear to the majority of the
people who use it.


Amateurs. Bleedin' amateurs.
Matthias

Jul 20 '05 #15

P: n/a
CJM

"Matthias Gutfeldt" <sa************@gmx.net> wrote in message
news:bv************@ID-16734.news.uni-berlin.de...
Amateurs. Bleedin' amateurs.


I was going to post this under Eric's post, but I think this kinda sums it
up!

Amateurs, eh?! lol

I'm dead against being elitist and stuck-up. And I think the open nature of
the technology is a major benefit in so many ways, but with this comes the
downside.

Because it's so easy to kick out a few pages that do *something*, every
bloke and his dog thinks he's a web developer! And with this comes a
proliferation of bad how-to guides, and bad examples. What is more, the most
readily available development environments (although improving) dont always
help the situation. Plus, it's always harder to unlearn a bad habit than to
learn to do it right the first time.

Having said that, the web wouldn't be too bad a place if the only worry was
about Tables vs CSS-P. There are so many more worse crimes going on - just
look that the tag-soup that passes for the average web page.

But I'm optimistic that the situation will improve. And we can help by
continuing to try and build 'good' pages, hopefully improving the
signal-to-noise ratio...

CJM
Jul 20 '05 #16

P: n/a
Retlak wrote:

[Trying to make table-like layout with CSS]
This indicates that the people who created CSS botched the job. CSS is
not being accepted, and it's not because the browser implementors
aren't implementing it (on the whole they are doing a good job), it's
because CSS is no good. Separating layout from content is a good idea
- but it has not been done right in CSS.


As I told you in c.t.x, CSS offers a way to do "table-like" layout:

display: table/table-row/table-cell

But it's not implemented in several browser versions. And this is not
the CSS spec authors's fault. So people try to achieve similar effects
using features implemented more widely in today's browsers. And this is,
as you saw, not trivial.
--
Johannes Koch
In te domine speravi; non confundar in aeternum.
(Te Deum, 4th cent.)
Jul 20 '05 #17

P: n/a
On Tue, 27 Jan 2004 10:21:17 +0100, Matthias Gutfeldt
<sa************@gmx.net> wrote:

Amateurs. Bleedin' amateurs.


Don't forget that "amateur" means "people who do it for love, not for
money." That makes me an amateur too. Amateurs can rise to the same
heights of quality as professionals, given the same dedication and work.

Of course, I understand you were not intending to slight all amateurs -
there was more presentation than meaning in that statement ;)
Jul 20 '05 #18

P: n/a
On Mon, 26 Jan 2004 17:49:17 -0500, Neal <ne*****@spamrcn.com> wrote:
Facts are facts. CSS is the correct way to do layout, not tables. Table
markup must be reserved for tabular data, lest we lose the meaning of that
markup entirely.


Some prefer CSS and some tables. I do not think it is a huge crime to
use tables for layout.

Two quotes from World Wide Web Consortium's website:

"The HTML table model allows authors to arrange data -- text,
preformatted text, images, links, forms, form fields, other tables, etc.
into rows and columns of cells."

"Tables should not be used purely as a means to layout document content
as this may present problems when rendering to non-visual media.
Additionally, when used with graphics, these tables may force users to
scroll horizontally to view a table designed on a system with a larger
display. To minimize these problems, authors should use style sheets to
control layout rather than tables."

World Wide Web Consortium does courage making layouts with CSS, but it
does not say that tables are only for tabular data.
Jul 20 '05 #19

P: n/a
On Tue, 27 Jan 2004 20:10:36 GMT, Verdoux <ve*****@hotmail.com> wrote:
Two quotes from World Wide Web Consortium's website:

"The HTML table model allows authors to arrange data -- text,
preformatted text, images, links, forms, form fields, other tables, etc.
into rows and columns of cells."
Yep. Data. Into cells.
"Tables should not be used purely as a means to layout document content
as this may present problems when rendering to non-visual media.
In other words, don't use tables for the sole reason of content layout.
Additionally, when used with graphics, these tables may force users to
scroll horizontally to view a table designed on a system with a larger
display. To minimize these problems, authors should use style sheets to
control layout rather than tables."
There you go. Use style sheets and not tables.
World Wide Web Consortium does courage making layouts with CSS, but it
does not say that tables are only for tabular data.


Uh, did you quote the right part? Because the above clearly says (1)
tables are for data and (2) tables are not for content layout.
Jul 20 '05 #20

P: n/a
On Tue, 27 Jan 2004 16:24:40 -0500, Neal <ne*****@spamrcn.com> wrote:
On Tue, 27 Jan 2004 20:10:36 GMT, Verdoux <ve*****@hotmail.com> wrote:
Two quotes from World Wide Web Consortium's website:

"The HTML table model allows authors to arrange data -- text,
preformatted text, images, links, forms, form fields, other tables, etc.
into rows and columns of cells."


Yep. Data. Into cells.
"Tables should not be used purely as a means to layout document content
as this may present problems when rendering to non-visual media.


In other words, don't use tables for the sole reason of content layout.
Additionally, when used with graphics, these tables may force users to
scroll horizontally to view a table designed on a system with a larger
display. To minimize these problems, authors should use style sheets to
control layout rather than tables."


There you go. Use style sheets and not tables.
World Wide Web Consortium does courage making layouts with CSS, but it
does not say that tables are only for tabular data.


Uh, did you quote the right part? Because the above clearly says (1)
tables are for data and (2) tables are not for content layout.


What is "content" if not " text, preformatted text, images, links, forms, form
fields, other tables, etc."

Neal, it's apparent that you are very expert in CSS. Why don't you
give us the URL's for one or more of the web pages you've designed?

Mason C
Jul 20 '05 #21

P: n/a
On Tue, 27 Jan 2004 21:46:41 GMT, Mason A. Clark
<ma*******@THISix.netcom.comQ> wrote:
What is "content" if not " text, preformatted text, images, links,
forms, form
fields, other tables, etc."
In this case, the quoted excerpt -
"The HTML table model allows authors to arrange data -- text,
preformatted text, images, links, forms, form fields, other tables,
etc.
into rows and columns of cells."


- indicates it is speaking of data, and is listing various kinds of data
that can be used. I don't read this as saying that anything that is in the
category of "text, preformatted text, images, links, forms, form fields,
other tables, etc." can be used, just when it's data.
Neal, it's apparent that you are very expert in CSS. Why don't you
give us the URL's for one or more of the web pages you've designed?


Mason, I'd be a liar if I characterized myself as an expert. I'm decidedly
an amateur. When I got into this forum I was very incompetent with
positioning. I would describe myself now as "getting it" - I know enough
to see the possibilities, not necessarily to always make it happen.

As far as sites, my latest project is the site I do for the orchestra I'm
involved with - my redesign is currently at http://www.opro.org/newdesign.
Still not finished, but it's getting there. Plus I'm working on my own
site at http://users.rcn.com/neal413 - that's when I have enough free
time. Haven't touched it in a few weeks, and it's not at all complete.

When I made my first website ever - back in '97 I believe - yep, frames
for layout. I know all about that. But the problems with them were
apparent to me even then. The current live orchestra site design (dating
from maybe 2000) uses frames - it's a really awful design in many ways. I
hope my new approach is an improvement.

So while I'm no expert in the execution of CSS, I am firmly committed to
the theory and practicality of it.
Jul 20 '05 #22

P: n/a
Neal wrote:
On Tue, 27 Jan 2004 10:21:17 +0100, Matthias Gutfeldt
<sa************@gmx.net> wrote:

Amateurs. Bleedin' amateurs.

Don't forget that "amateur" means "people who do it for love, not for
money." That makes me an amateur too. Amateurs can rise to the same
heights of quality as professionals, given the same dedication and work.

Of course, I understand you were not intending to slight all amateurs -
there was more presentation than meaning in that statement ;)


OK, bad choice of word, I guess. Just replace "amateur" with a suitable
antonym for "expert" :-).
Matthias

Jul 20 '05 #23

P: n/a
On Tue, 27 Jan 2004 16:24:40 -0500, Neal <ne*****@spamrcn.com> wrote:
"The HTML table model allows authors to arrange data -- text,
preformatted text, images, links, forms, form fields, other tables, etc.
into rows and columns of cells."


Yep. Data. Into cells.


What else is there?
"Tables should not be used purely as a means to layout document content
as this may present problems when rendering to non-visual media.


In other words, don't use tables for the sole reason of content layout.
Additionally, when used with graphics, these tables may force users to
scroll horizontally to view a table designed on a system with a larger
display. To minimize these problems, authors should use style sheets to
control layout rather than tables."


Uh, did you quote the right part? Because the above clearly says (1)
tables are for data and (2) tables are not for content layout.


To my understanding of English language, "should use" does not mean
that the second option is completely wrong. It means that it would be
preferable to use the other one.

The two problems that are mentioned above depends on the situation. For
instance, would a blind person browse a picture gallery site (I don't
think so)?

In addition, table widths do not have to be in pixels, they can also be
in percents, that way it "lives" with the screen size and those nasty
horizontal scrollbars do not appear.

This table vs. CSS issue should not be so black and white (because it
is not). Both ways can coexist.
Jul 20 '05 #24

P: n/a
On Tue, 27 Jan 2004 17:41:03 -0500, Neal <ne*****@spamrcn.com> wrote:
On Tue, 27 Jan 2004 21:46:41 GMT, Mason A. Clark
<ma*******@THISix.netcom.comQ> wrote:
What is "content" if not " text, preformatted text, images, links,
forms, form
fields, other tables, etc."


In this case, the quoted excerpt -
"The HTML table model allows authors to arrange data -- text,
preformatted text, images, links, forms, form fields, other tables,
etc.
into rows and columns of cells."


- indicates it is speaking of data, and is listing various kinds of data
that can be used. I don't read this as saying that anything that is in the
category of "text, preformatted text, images, links, forms, form fields,
other tables, etc." can be used, just when it's data.
Neal, it's apparent that you are very expert in CSS. Why don't you
give us the URL's for one or more of the web pages you've designed?


Mason, I'd be a liar if I characterized myself as an expert. I'm decidedly
an amateur. When I got into this forum I was very incompetent with
positioning. I would describe myself now as "getting it" - I know enough
to see the possibilities, not necessarily to always make it happen.

As far as sites, my latest project is the site I do for the orchestra I'm
involved with - my redesign is currently at http://www.opro.org/newdesign.
Still not finished, but it's getting there. Plus I'm working on my own
site at http://users.rcn.com/neal413 - that's when I have enough free
time. Haven't touched it in a few weeks, and it's not at all complete.

When I made my first website ever - back in '97 I believe - yep, frames
for layout. I know all about that. But the problems with them were
apparent to me even then. The current live orchestra site design (dating
from maybe 2000) uses frames - it's a really awful design in many ways. I
hope my new approach is an improvement.

So while I'm no expert in the execution of CSS, I am firmly committed to
the theory and practicality of it.


Neal, thanks. In the CSS world, you're an expert. Your pages are elegant and
the coding is simple and clean. Not that I'm an expert to be judging -- just
learning. ( I have a collection of ancient html table-formatted pages needing
updating. )

Mason C
Jul 20 '05 #25

P: n/a
John Bowling wrote:
Barry, At your web page
http://www.barry.pearson.name/articles/layout_tables/
you mention that CSS positioning will not be effective for several
years when MS finally gets their act together.
Not quite my position. I believe that CSS positioning is highly dependent on
the collective skills of people to discover and publicise what works in
practice. This changes year by year. Everyone who discovers something new and
publicises it helps.

For example, I have publicised some tableless-layouts & their CSS that I
developed recently. Perhaps no one will ever copy them. Perhaps 10 people
will. Or a 100. But I did what I could. Please have a look:
http://www.barry.pearson.name/articles/templates/

The browsers won't change for years. I don't see new authoring tools appearing
and taking the pain away. So each of us who makes some progress should
publish. But this is not the way a proper system would work. How are newbies
expected to handle this?
So, why is everybody still using IE (which is very old on other
support - javascript limitied to 1.2, etc.) when they could be using
Mozilla which has the most complete current support of all of W3C's
standards?

Don't limit web pages because a $$$ company can't keep up, tell
everyone who visits your sites to switch (at no cost - ever) and get
far superior results and far fewer security issues!

[snip]

Don't we keeping saying "the user is in control"?

Let's act that way. I will react according to the way my target audience
chooses to act. I may be able to influence them over a year or two. But not
immediately.

--
Barry Pearson
http://www.Barry.Pearson.name/photography/
http://www.BirdsAndAnimals.info/
http://www.ChildSupportAnalysis.co.uk/
Jul 20 '05 #26

P: n/a
Berislav Lopac wrote:
[snip]
Unfortunately, it is not even that much of Microsoft's fault as it is
of the average users out there. Most people use Windows, and on most
Windows platforms IE comes preinstalled.

Most users expect Web to work like a TV: if one "program" (website)
can't be seen right, and others are OK, than that one is "broken".
And since most are OK, they just ignore the "broken" site and move on.

[snip]

And they are right! Why didn't everyone do what the OK site did?

Do we believe we are in control? Or do we accept that the users are in
control?

--
Barry Pearson
http://www.Barry.Pearson.name/photography/
http://www.BirdsAndAnimals.info/
http://www.ChildSupportAnalysis.co.uk/
Jul 20 '05 #27

P: n/a
Philipp Lenssen wrote:
Barry Pearson wrote:

[snip]
I estimate that at least 95% (I actually believe 99%, but I could
never prove it!) of the web pages (indeed web sites) I see use
table-layout.


I would say the figure is around 99.99%


Gosh! I would never have dared to say that!
This isn't a problem.


Have you tried printing those pages? Endless navigation garbage and
the content pushed out of the right edge? Happens.
Have you tried browsing with your hand phone display? Makes the Web
completely useless because of table-layouty schemes (table-layout is
not the only problem, but a big part of it, e.g. navigation usually
comes left and thus must come first if you linearize it -- not so in
CSS).


OK. But what is your point? Where does that take us?

Which is more likely - all those users change their browsers; or browser
developers change to meet the desires of those users?

Those table-layout pages will be part of the planet's information resource for
a very long time. Users will expect to access those pages. When enough people
have a demand, developers will respond. Anyone who have browsed lots of sites
using the Opera 7.2 "small screen mode" will know that incredible things are
possible. I was astonished to see pages of mine that I though could never be
shown on 240px screens showing quite well.

Where will we be in 5 years time?

--
Barry Pearson
http://www.Barry.Pearson.name/photography/
http://www.BirdsAndAnimals.info/
http://www.ChildSupportAnalysis.co.uk/
Jul 20 '05 #28

P: n/a
Retlak wrote:
[snip]
Maybe the first two have more connection to the real world?
I don't want to be inflammatory, nor do I want to even seem for an
instant to approve of M$'s hostility to the spirit of standards. What
I mean is that CSS has been botched.
CSS for *styles* is excellent. CSS for *page layout* is botched. Or, rather, I
believe it was simply never intended for that purpose. So why should we be
surprised if it can't do it very well?
In the early days of the web, when frames appeared, you could learn in
about an hour how to do web-site layout with frames. Of course, people
eventually realized that frames cause a number of problems. So a lot
of those web sites converted to frameless layouts using tables. Again,
tables have a clear conceptual model as a layout tool. You could pick
it up in an hour or less.
People treat "horizontal" and "vertical" as very important. We even have
neurons in the brain that react to these. We are sensitive to "sloping
horizons" and "converging verticals" in pictures.

Why did people put things in tabular format long before the web? Becase it is
so natural to us. HTML tables are a *consequence* of the way we think about
layout. People think and demand grid-like layout. Tables were designed with
our expectations in mind. They are not an accident.

http://www.barry.pearson.name/articl...es/history.htm
But converting those sites to use CSS is a time-consuming nightmare.
There are lots of web sites out there which explain how to use CSS for
2-column layouts ... how to make the columns look the same length ...
how to create footers ... etc. But the very proliferation of such
sites is an indication that something is wrong. Tasks that were
trivial with frames or tables (because of their clear conceptual
model) are difficult with CSS.
CSS is not a page layout language. If it had been designed for that purpose,
would we still be struggling more than 5 years after the CSS2 Recommendation
was published? Wouldn't it have been accompanied by a demonstration that it
had succeeded with its aims - if page layout were ever its aims?

Try designing a page layout language. It would look much more like tables than
CSS 1+2! (But it wouldn't actually be either).
This indicates that the people who created CSS botched the job. CSS is
not being accepted, and it's not because the browser implementors
aren't implementing it (on the whole they are doing a good job), it's
because CSS is no good. Separating layout from content is a good idea
- but it has not been done right in CSS.


Correct. We actually have 2 (or more) imperfect schemes, neither of which was
designed for the purpose. We are better off taking the best of both, until
something better comes along.

--
Barry Pearson
http://www.Barry.Pearson.name/photography/
http://www.BirdsAndAnimals.info/
http://www.ChildSupportAnalysis.co.uk/
Jul 20 '05 #29

P: n/a
Neal wrote:
[snip]
I think a large part of the problem is the misconception that CSS is a
drop-in replacement for tabular layout.
It shouldn't be thought of as a drop-in replacement. But it *should* be
thought of as a way that people with a design-concept for a web page can
achieve that concept. And it was not really designed for that.

We must not say "CSS can't do X, therefore it is wrong to expect web pages to
look like X". That is a pathetic cop-out. The web exists for people to
communicate with one another. And if that involves some well-considered
layouts that are not contrained by the limitations of CSS, so be it. CSS
should *serve* people's needs to communicate in what they consider to be the
most effective way, it should *not* constrain it.
The misuse of the table markup has led to some rather attractive and
popular designs. Naturally, the web author wants to accomplish this
type of layout with CSS, and there are those who mislead by saying
"you can do anything with CSS." So the author half-learns CSS, cannot
accomplish the identical table layout, and gives up.
Turn that round. If those designs were popular, there is a reason. For
example, *people* want it! And the web exists for people.

It is foolish to believe that human nature can be thwarted. Human beings
wanted elaborate page layouts, with lots of colours & images, etc. So they got
it. They had to use tables to get it. If tables hadn't existed they would have
used other means. Perhaps that would have involved forcing browser developers
to put those features in. But people, en masse, will not be thwarted by purist
principles. Have they ever?
Facts are facts. CSS is the correct way to do layout, not tables.
Table markup must be reserved for tabular data, lest we lose the
meaning of that markup entirely. And anything worth doing (including
CSS) is worth taking enough time and experimentation with to
eventually make it possible to work with it fluently.
Who says CSS is the correct way to do layout? Where is your evidence? Can you
show that CSS1+2 was ever intended to be a page layout language? Can you show
that it would be competent at the task even if all browsers supported it?
Don't make the dreadful mistake of thinking that because it is what we have,
it must be right! When has that *ever* been the case?

Right from the start (say late 1993), tables were intended to layout complex
material in a grid-like formation. That was long before browsers implemented
tables (say late 1994). The discussion of "tabular data" came years later!
(Say 1997). It is tempting to try to re-write history. But see:

"A brief history of tables"
http://www.barry.pearson.name/articl...es/history.htm
Over time students of CSS can see that it's totally different from
table layout, and works nothing like it, but the resulting product is
far superior in many ways. And they see that the CSS model is not a
replacement for the table layout model, but the first really correct
way to do the job of layout where tables were misused to get a layout
"effect".
No. It is different. No more, no less. We have 2 imperfect systems, and only
people willing to stand on one foot should shoot themselves in the other foot
by casting-out a sensible technique.
Yes, I agree that the learning curve for CSS is steep, and that has an
effect. But the learning curve for Javascript seems steep as well,
and we see it used all over. Somehow I can't accept the learning
curve as being a primary reason for the problem. A CSS-Lite might be
a worthwhile objective, perhaps...


W3C published a Recommendation in December 1997 that CSS should be used for
layout. (It first appeared in a working draft in September 1997). Yet we are
still struggling! That is more than a "learning curve". That is a desperate
struggle of many brilliant people to try to adapt a style system for the
purposes of page layout.

When Sony & Philips published the specification for music CDs, it didn't take
many years for people to discover how to put music onto a CD. That was written
into the specification from the start. You just followed the instructions, and
people could put the CD into a player, and even on day 1 it worked. (I bought
my first CD player & a couple of CDs on day 2! And they worked, of course).

If CSS1+2 had really been designed for page layout, we would have known how to
do basic layouts by early 1998, and would just have been waiting for the
technology to arrive. But even in 2000, W3C used tables for some of their
pages because they couldn't make alternatives work. It isn't simply a problem
with some browsers - no one appears to have thought it through.

--
Barry Pearson
http://www.Barry.Pearson.name/photography/
http://www.BirdsAndAnimals.info/
http://www.ChildSupportAnalysis.co.uk/
Jul 20 '05 #30

P: n/a
Barry Pearson wrote:

CSS for *styles* is excellent. CSS for *page layout* is botched. Or, rather, I
believe it was simply never intended for that purpose.


Hmmm... I wonder what the intended purpose of the "float" and "position"
properties were, if not for layout?

--
To email a reply, remove (dash)un(dash). Mail sent to the un
address is considered spam and automatically deleted.
Jul 20 '05 #31

P: n/a
On Fri, 30 Jan 2004 21:42:22 -0000, Barry Pearson
<ne**@childsupportanalysis.co.uk> wrote:
Why did people put things in tabular format long before the web? Becase
it is
so natural to us. HTML tables are a *consequence* of the way we think
about
layout. People think and demand grid-like layout. Tables were designed
with
our expectations in mind. They are not an accident.
I'll never argue that a grid method isn't sensible. I don't agree that the
HTML table model is the proper tool for the job. I don't recall ever
seeing a layout in tables that cannot be at least approximated in CSS,
losing some precision in placement but no usability.
CSS is not a page layout language. If it had been designed for that
purpose,
would we still be struggling more than 5 years after the CSS2
Recommendation
was published? Wouldn't it have been accompanied by a demonstration that
it
had succeeded with its aims - if page layout were ever its aims?
I'm not really struggling, Barry. And the float property, the position
property - gee, seems like they TRIED to design it to be a layout protocol.
Correct. We actually have 2 (or more) imperfect schemes, neither of
which was
designed for the purpose. We are better off taking the best of both,
until
something better comes along.


I'll agree to this degree - CSS isn't 'perfect'. But it's still pretty
new. A decade from now, we'll have long forgotten about any benefits from
table-markup layout.
Jul 20 '05 #32

P: n/a
On Fri, 30 Jan 2004 22:18:17 -0000, Barry Pearson
<ne**@childsupportanalysis.co.uk> wrote:
We must not say "CSS can't do X, therefore it is wrong to expect web
pages to
look like X".
I've never said that. But I do say that until they invent a car I can buy
that runs without engine oil, I'm putting oil in my car even if it's not
exactly convenient. It's the right way to do things, and it's good for me
in the long run.

I think you will agree that 10 years down the road, table markup will not
be the best tool for layout. It is far more likely any progress in layout
tools will be besed on the CSS model than on the HTML model. So, by doing
a table layout you're guaranteeing that down the road you'll have to
redesign the whole site to take advantage of the best that web layout
options have to offer. By using CSS for layout, in time the layout model
will be based on what I'm already using.

I can expect web pages to look like X or Y. But that doesn't mean I can
make it happen without shooting myself in the foot.
Turn that round. If those designs were popular, there is a reason. For
example, *people* want it! And the web exists for people.
People want to ploonk my girlfriend too.
Who says CSS is the correct way to do layout? Where is your evidence?
Read the W3C specifications. They explain that though table layout is
common, one should use CSS for layout.
Can you
show that CSS1+2 was ever intended to be a page layout language?
float. position.
Can you show
that it would be competent at the task even if all browsers supported it?
Give me a design you can do in tables that I cannot at least get the
spirit of in CSS.
W3C published a Recommendation in December 1997 that CSS should be used
for
layout. (It first appeared in a working draft in September 1997).
Oh good, you read that.
Yet we are
still struggling!
Who's this 'we', man? I'm doin' great.
That is more than a "learning curve". That is a desperate
struggle of many brilliant people to try to adapt a style system for the
purposes of page layout.
No, that's the result of a functioning yet incorrect way of layout making
it seem to the average web authoir that it's not worthwhile to learn a new
system. Why should I learn to run a chainsaw when I'm getting these trees
down just fine with my chisels?
When Sony & Philips published the specification for music CDs, it didn't
take
many years for people to discover how to put music onto a CD. That was
written
into the specification from the start. You just followed the
instructions, and
people could put the CD into a player, and even on day 1 it worked. (I
bought
my first CD player & a couple of CDs on day 2! And they worked, of
course).
Right, and Sony and Phillips made or designed all the CD players too. We
have a lot of different browser makers all following the specs a little
differently. I don't see that as a negative - a hassle sometimes, sure,
but it's not a negative. The positive effect of innovation and the
discovery of ways to grow this technique of making web pages far outweighs
the negative of having to figure out the relatively few incompatibility
issues.
If CSS1+2 had really been designed for page layout, we would have known
how to
do basic layouts by early 1998, and would just have been waiting for the
technology to arrive.
Nah. Tables are easy. People like to use them. And when CSS was first
invented, the table model was the only layout tool we had, and it worked
pretty good. The fault isn't that CSS cannot do layout - it can, and well.
The problem is simply that you need to give people a reason to change, and
apparently "Stop abusing markup and use this new method of layout" wasn't
enough.
But even in 2000, W3C used tables for some of their
pages because they couldn't make alternatives work. It isn't simply a
problem
with some browsers - no one appears to have thought it through.


Set up a challenge for the newsgroup. Make a table layout and invite
people to come as close as they can with CSS. Go ahead.
Jul 20 '05 #33

P: n/a
In article <op**************@news.rcn.com>, Neal <ne*****@spamrcn.com>
wrote:
I think a large part of the problem is the misconception that CSS is a
drop-in replacement for tabular layout.
Indeed.
Facts are facts. CSS is the correct way to do layout, not tables. Table
markup must be reserved for tabular data, lest we lose the meaning of that
markup entirely.


Another common misconception is that the table layout model is more evil
than CSS positioning. However, the original argument was that <table>
*markup* should be reserved for tabular data. The table layout model
(and CSS floats) scale better to different view ports and font sizes
than CSS positioning.

Moving from tables to CSS positioning (as opposed to float or display:
table) is jumping from the frying pan into the fire.

--
Henri Sivonen
hs******@iki.fi
http://iki.fi/hsivonen/
Mozilla Web Author FAQ: http://mozilla.org/docs/web-developer/faq.html
Jul 20 '05 #34

P: n/a
Barry Pearson wrote:

Why did people put things in tabular format long before the web?
Because they had tabular data.
HTML tables are a *consequence* of the way we think about layout.
No. HTML tables are a consequence of wanting to present tables using
HTML. Just like HTML lists are a consequence of wanting to present
lists using HTML.
People think and demand grid-like layout.
Grid layout is not tabular layout. DTP programs don't offer table
data blocks to layout e.g. a flyer. They use blocks with a grid and
rulers. grid <> table
http://www.barry.pearson.name/articl...es/history.htm
As unconvincing on the web as they are in usenet.
CSS is not a page layout language.
float; clear; position; margin; padding; text-align; vertical-align.
I'd have to say that, as is so often the case, you're determined to
see things your way in the face of the obvious.
If it had been designed for that purpose, would we still be
struggling more than 5 years after the CSS2 Recommendation was
published? Wouldn't it have been accompanied by a demonstration
that it had succeeded with its aims - if page layout were ever its
aims?


Wouldn't that rely on implementation? Wouldn't there be some time
between a recommendation and its implementation?

--
Brian (follow directions in my address to email me)
http://www.tsmchughs.com/

Jul 20 '05 #35

P: n/a
Barry Pearson wrote:

We must not say "CSS can't do X, therefore it is wrong to expect
web pages to look like X".
But we should say, "the web cannot do that, or cannot do that well,
therefore it is wrong to expect a web technology to produce that result.
Can you show that CSS1+2 was ever intended to be a page layout
language?
Yes, but simply looking at the spec.
Can you show that it would be competent at the task even if all
browsers supported it?
Since they do not, how do you propose one make such a demonstration?
If CSS1+2 had really been designed for page layout, we would have
known how to do basic layouts by early 1998, and would just have
been waiting for the technology to arrive. But even in 2000, W3C
used tables for some of their pages because they couldn't make
alternatives work.


And that was because of faults in the css recommendations? Or because
their implementation was found wanting?

--
Brian (follow directions in my address to email me)
http://www.tsmchughs.com/

Jul 20 '05 #36

P: n/a
On 1/30/2004, in comp.infosystems.www.authoring.stylesheets, Barry
Pearson wrote:
Right from the start (say late 1993), tables were intended to
layout complex material in a grid-like formation. That was long
before browsers implemented tables (say late 1994). The discussion
of "tabular data" came years later! (Say 1997). It is tempting to
try to re-write history.
But, on 1/19/2004, in comp.infosystems.www.authoring.html, re: tables
for layout, Barry Pearson wrote:
- What were the original intentions? They are interesting from a
historical perspective. They are irrelevant for how we should
behave today. We can't afford to re-fight lost battles. (But
history is still interesting! That is why I am maintaining a page
on the subject).

- How do things behave today, and in the future? This is what
matters.


So, if "historical perspective...[is] irrelevant to how we should
behave today," why do you keep bringing it up?

--
Brian (follow directions in my address to email me)
http://www.tsmchughs.com/

Jul 20 '05 #37

P: n/a
On Sat, 31 Jan 2004 17:18:10 GMT, Brian
<us*****@julietremblay.com.invalid-remove-this-part> wrote:
Barry Pearson wrote:

People think and demand grid-like layout.


Grid layout is not tabular layout. DTP programs don't offer table
data blocks to layout e.g. a flyer. They use blocks with a grid and
rulers. grid <> table


Problem is, there is no easy 'grid' capability in CSS. CSS 2 introduced
something that goes a long way (but witpout spanning cells across rows or
columns), which works OK in Opera and Mozilla. But it doesn't work in MSIE.

Positioning can be made to work, but it is a lot more work to get it
right. So it is no wonder that manage designers cling to their trusted
table methods for creating a basic page grid. I don't understand why
anyone would still want to complex nested tables though. CSS is *so* much
easier to create colored background, borders, spacing etc.

--
Rijk van Geijtenbeek

The Web is a procrastination apparatus:
It can absorb as much time as is required to ensure that you
won't get any real work done. - J.Nielsen
Jul 20 '05 #38

P: n/a
On Sat, 31 Jan 2004 21:55:10 +0100, Rijk van Geijtenbeek <ri**@opera.com>
wrote:
But it doesn't work in MSIE.


- which, incidentally, is the real reason CSS has been slow to catch on.
It's not necessarily a fault of CSS that a major browser fails to
implement it.
Jul 20 '05 #39

P: n/a
In article Verdoux wrote:
On Tue, 27 Jan 2004 16:24:40 -0500, Neal <ne*****@spamrcn.com> wrote:
"The HTML table model allows authors to arrange data -- text,
preformatted text, images, links, forms, form fields, other tables, etc.
into rows and columns of cells."
Yep. Data. Into cells.


What else is there?


Things that aren't tabular data. Of course one bit can be thought to be
table data, or just normal content. But not all bits of normal flow can
be thought as tabular.
"Tables should not be used purely as a means to layout document content
as this may present problems when rendering to non-visual media.


In other words, don't use tables for the sole reason of content layout.


Right. That pretty much disallows using tables for layout.
To my understanding of English language, "should use" does not mean
that the second option is completely wrong. It means that it would be
preferable to use the other one.
Your understanding and specs are different. Go look definition of
"should" in that spec, and then RFC it refers. You might be surprised.
The two problems that are mentioned above depends on the situation. For
instance, would a blind person browse a picture gallery site (I don't
think so)?
No, but person with browser that is not capable to show images (nor
tables), but is able to launch other application to do showing. I have
tried that, as I needed one image quickly, and first machine on my way
was dump terminal. Couldn't do that, table layout and inadequate alt
texts took care of that.
In addition, table widths do not have to be in pixels, they can also be
in percents, that way it "lives" with the screen size and those nasty
horizontal scrollbars do not appear.
Percents aren't good for width of text. Especially, if less than 50% is
used. Why? Because people may use big font, and still have small
screen/window. Of course, it is even more important with CSS, as tables
has build in min-width.

Table layouts has one advantage on CSS layouts that work in IE: it
doesn't need widths. If you use width, it is practically never better
than CSS layout.
This table vs. CSS issue should not be so black and white (because it
is not). Both ways can coexist.


Yes. The problem is that people insisting on that tables are good thing
are the same that use them when there is better alternatives. And when
there isn't, they still use them stupid way.

And using advanged CSS and tables for layout at same time, is very
stupid, of course, unless it is very carefully done, then it actually can
be better.

--
Lauri Raittila <http://www.iki.fi/lr> <http://www.iki.fi/zwak/fonts>
Saapi lähettää meiliä, jos aihe ei liity ryhmään, tai on yksityinen
tjsp., mutta älä lähetä samaa viestiä meilitse ja ryhmään.

Jul 20 '05 #40

P: n/a
In article Rijk van Geijtenbeek wrote:
On Sat, 31 Jan 2004 17:18:10 GMT, Brian
<us*****@julietremblay.com.invalid-remove-this-part> wrote:
Barry Pearson wrote:
People think and demand grid-like layout.


Grid layout is not tabular layout. DTP programs don't offer table
data blocks to layout e.g. a flyer. They use blocks with a grid and
rulers. grid <> table


Problem is, there is no easy 'grid' capability in CSS. CSS 2 introduced
something that goes a long way (but witpout spanning cells across rows or
columns), which works OK in Opera and Mozilla. But it doesn't work in MSIE.


The problem is that authors want to have grid. That is biggest problem
with most table layouts - they wan't it to be grid. Grid is much less
flexible as almost anything else. The reason is that it can't wrap.

And, it is usually possible to do table layout whitout needing to grid
everything.
Positioning can be made to work, but it is a lot more work to get it
right.


I used to think it was easy. Then I understood I'm amateur only in sence
that I'm not paid...

50% of pages done using positioning are done badly. 90% of done using
layout tables are done badly. The best thing in CSS is that I can cure
80% of incorrectly made CSS layouts (those with correct HTML markup), but
only about 50% of table layouts - client side.

So, it leaves only 10% nonworking CSS layouts, compared to 45% of non-
working table layouts. For those, I need to make my window bigger. It
requires twice as much work as closing page, so latter is more likely
option...

(Figures from my new South-American hat I got for X-mas.)

--
Lauri Raittila <http://www.iki.fi/lr> <http://www.iki.fi/zwak/fonts>
Saapi lähettää meiliä, jos aihe ei liity ryhmään, tai on yksityinen
tjsp., mutta älä lähetä samaa viestiä meilitse ja ryhmään.

Jul 20 '05 #41

P: n/a
On Wed, 28 Jan 2004 13:21:34 GMT, Verdoux <ve*****@hotmail.com> wrote:
The two problems that are mentioned above depends on the situation. For
instance, would a blind person browse a picture gallery site (I don't
think so)?


Mary's husband loves seeing snow. Unfortunately he's stationed in the
desert. So she goes online searching for images of snowscapes, preferably
not far from home, to send to him.

She soon finds that if she were searching for pictures of "Click here to
enlarge" she'd have had better luck. Until she finds your gallery, where
the alt tags explain what the pictures are.

Don't be so sure blind folks aren't interested in images.
Jul 20 '05 #42

P: n/a
Brian wrote:
On 1/30/2004, in comp.infosystems.www.authoring.stylesheets, Barry
Pearson wrote:
Right from the start (say late 1993), tables were intended to
layout complex material in a grid-like formation. That was long
before browsers implemented tables (say late 1994). The discussion
of "tabular data" came years later! (Say 1997). It is tempting to
try to re-write history.


But, on 1/19/2004, in comp.infosystems.www.authoring.html, re: tables
for layout, Barry Pearson wrote:
- What were the original intentions? They are interesting from a
historical perspective. They are irrelevant for how we should
behave today. We can't afford to re-fight lost battles. (But
history is still interesting! That is why I am maintaining a page
on the subject).

- How do things behave today, and in the future? This is what
matters.


So, if "historical perspective...[is] irrelevant to how we should
behave today," why do you keep bringing it up?


Because, as I said earlier (see above) "They are interesting from a historical
perspective", and "history is still interesting"!

So I continue to update my page: "A brief history of tables"
http://www.barry.pearson.name/articl...es/history.htm

And because I respond to others who believe that original intentions should
dictate how people behave today. I believe they are wrong, so I say so.

--
Barry Pearson
http://www.Barry.Pearson.name/photography/
http://www.BirdsAndAnimals.info/
http://www.ChildSupportAnalysis.co.uk/
Jul 20 '05 #43

P: n/a
Neal wrote:
On Fri, 30 Jan 2004 22:18:17 -0000, Barry Pearson
<ne**@childsupportanalysis.co.uk> wrote:
We must not say "CSS can't do X, therefore it is wrong to expect web
pages to
look like X". [snip]
I think you will agree that 10 years down the road, table markup will
not be the best tool for layout. It is far more likely any progress
in layout tools will be besed on the CSS model than on the HTML
model. So, by doing a table layout you're guaranteeing that down the
road you'll have to redesign the whole site to take advantage of the
best that web layout options have to offer. By using CSS for layout,
in time the layout model will be based on what I'm already using.
In 10 years time, perhaps XHTML 2.0 will be the best standard for web pages.
Does that mean we should all be using XHTML 2.0 now?

I hope that indeed we will see progress in development tools based on the CSS
model. And when we get those tools, it will become much more sensible to use
CSS positioning for everything, or nearly everything, we do. But if such tools
are desirable, then perhaps it is because we *need* them! Why shouldn't we
therefore wait for them? (*If*, indeed, it is possible to develop them).

Why will anyone have to redesign their sites? Table-layout sites will work at
least as well in 10 years (and 20 years time) time as they do today. Authors
who use the modern table-layout techniques, where only the very minimum table
is in HTML, and the rest of the controls are in CSS, will be able to
continually update just by changing the CSS.

I've been able to turn many pages designed without standards in mind, and no
use of CSS, into, first 4.01 Transitional + CSS, then 4.01 Strict + CSS. I
admit that this has needed hours of "find & replaces". So I may need to spend
more hours on "find & replaces" within a decade. Big deal!

[snip] Who says CSS is the correct way to do layout? Where is your evidence?


Read the W3C specifications. They explain that though table layout is
common, one should use CSS for layout.


I repeat my question: "Where is your evidence?" It doesn't exist!

[snip] Give me a design you can do in tables that I cannot at least get the
spirit of in CSS.


Saying "spirit of" is a cop-out! Some of us have been seeing the claims of
snake-oil merchants for decades. We recognise the tactics of salespeople who
say "your real requirements are not what you say, they are really these
different requirements, and we can satisfy your real-real requirements".

The instant you try to change the desired layout into something different, you
may be following in the footsteps of those snake-oil merchants. Why not,
instead, say "yes, we will deliver your exact layout using CSS positioning
techniques"? Do you lack the confidence to do that? I assume that the
table-layout pages on the web were the result of some design process, good or
bad. Can you confidently claim that CSS positioning can deliver the same
result to the target audience as table-layout?

I'll allow you to refuse to match any table-layout that you can show is simply
failing to communicate with the target audience. You don't have to show that
CSS positioning can deliver results as bad as the worst results of
table-layout! I'm not playing pedantic games here.

Here are some quotes that are too important to ignore:

- Float: "... floats were not originally intended to be used for layout, some
hacks may be necessary to make them behave as intended.... Looking to the
future, there have been a variety of proposed enhancements to CSS that would
allow an author to declare that an element should stretch to contain any
floated elements within itself.... but as of this writing, support for such
abilities is likely to be a long time coming".
Eric Meyer, Complex Spiral Consulting, Containing Floats
http://www.complexspiral.com/publica...aining-floats/

- Position: "And much of the blame for this must be directed at CSS itself....
And while, thanks to their efforts, the main problems of how to produce
workable layouts from the CSS specification have now been cracked, there are
still innumerable smaller issues ready to raise their head and make the
designer's life a misery. Such as: how to consistently control margins and
padding to ensure alignment, how to set up a full-width footer section below
columns of varying length; how to vertically centre content; how to give a
coloured background to a side column that matches the height of a longer
content column; and so on. These are all givens with table-based layout and
should certainly have been addressed before CSS2 was allowed out of the door".
Tom Arah, The Future of Web Layout
http://www.designer-info.com/Web/abs...ositioning.htm

- Position: "There were really only three things we couldn't do with this
redesign: properly position our partner-specific footer at the bottom of the
page, use vertical alignment within divs, and achieve validation. Positioning
footers is a huge Achilles heel of absolute positioning. It is ridiculous that
you cannot embed three absolutely positioned columns within a master div and
then position a footer below that master div. This is a well known problem of
absolute positioning and there are a few workarounds, none of which are very
elegant".
Mike Davidson of ESPN.com
http://devedge.netscape.com/viewsour...-interview/02/

[snip]
That is more than a "learning curve". That is a desperate
struggle of many brilliant people to try to adapt a style system for
the purposes of page layout.


No, that's the result of a functioning yet incorrect way of layout
making it seem to the average web authoir that it's not worthwhile to
learn a new system. Why should I learn to run a chainsaw when I'm
getting these trees down just fine with my chisels?


You do what you like! I'm not trying to stop you using CSS positioning. Heck!
I use CSS positioning myself, where I feel it satisfies the requirements!
Between 1/3rd and half (I think at least 350 to 400 pages) of my web pages use
tableless-layout.

But the difference is that I am not prejudiced against table-layout. I use
engineering analysis to decide what is best in particular cases, then act
accordingly. So I have a hammer & and screwdriver in my toolkit. I am not
going to discard either of them.

[snip]
If CSS1+2 had really been designed for page layout, we would have
known how to
do basic layouts by early 1998, and would just have been waiting for
the technology to arrive.


Nah. Tables are easy. People like to use them. And when CSS was first
invented, the table model was the only layout tool we had, and it
worked pretty good. The fault isn't that CSS cannot do layout - it
can, and well. The problem is simply that you need to give people a
reason to change, and apparently "Stop abusing markup and use this
new method of layout" wasn't enough.

[snip]

First, see the quotes I provided above, CSS *can't* do layout well! Do you
think the people quoted are naive people who haven't studied this? No - these
are people who *have* to deliver, and find that CSS sucks.

Second - obviously unsupported b*llsh*t like "Stop abusing markup and use this
new method of layout" won't get anywhere! Who are those people to say that,
when it is obvious that the web works very well indeed with table-layout? We
have a web that may well be 99% table-layout. What would a visitor from Mars
say is demonstrably the most viable layout system on the planet? It worked in
1995. It works in 2004. There is so much material around, that browser
developers will ensure that it works in 2014. And probably 2024. And ...

Table layout is part of the future. It is useful to have a discussion about
what else is part of the future. But .... as well, not instead of.

--
Barry Pearson
http://www.Barry.Pearson.name/photography/
http://www.BirdsAndAnimals.info/
http://www.ChildSupportAnalysis.co.uk/
Jul 20 '05 #44

P: n/a
Neal wrote:
On Fri, 30 Jan 2004 21:42:22 -0000, Barry Pearson
<ne**@childsupportanalysis.co.uk> wrote:
Why did people put things in tabular format long before the web?
Becase it is
so natural to us. HTML tables are a *consequence* of the way we think
about
layout. People think and demand grid-like layout. Tables were
designed with
our expectations in mind. They are not an accident.


I'll never argue that a grid method isn't sensible. I don't agree
that the HTML table model is the proper tool for the job. I don't
recall ever seeing a layout in tables that cannot be at least
approximated in CSS, losing some precision in placement but no
usability.


Irrelevant! Can CSS do what the designer wanted? (Not "approximated").
Designers are not constrained - they can, and *should*, use whatever is needed
to achieve their designs.

They don't just look at tables & CSS. They look at lots more besides. They
always will. Look at this from David Siegel in November 1995:
http://www.w3.org/Style/951106_Works...t1.html#siegel
"David described himself as a `radical designer', i.e., one that doesn't care
about style sheets but about results. He would use whatever means where
available to achieve the desired appearance."

The web *can* do these things. If CSS can't, then designers will obviously
(and rightly) go elsewhere. The web isn't, and won't be, constrained by the
limitations in CSS positioning. (Who said that is the limit of the web?)
Neither should authors be constrained.
CSS is not a page layout language. If it had been designed for that
purpose,
would we still be struggling more than 5 years after the CSS2
Recommendation
was published? Wouldn't it have been accompanied by a demonstration
that it
had succeeded with its aims - if page layout were ever its aims?


I'm not really struggling, Barry. And the float property, the position
property - gee, seems like they TRIED to design it to be a layout
protocol.


Possibly. I still can't decide whether they were a terrible botch, or not an
attempt to do page layout. It is one or the other.

I am influenced by section 9.6.1 of the CSS2 specification. (Aren't we all?) I
thought "yes, they understand page layout, and are trying to do something
about it!" They showed an image of a header, a footer, a sidebar, and a main
area. They also provided the HTML and the CSS needed to achieve this. So they
*really* appreciated the need to be able to position a footer below all the
columns! But ... have a look:

"9.6.1 Fixed positioning.... Authors may use fixed positioning to create
frame-like presentations":
http://www.w3.org/TR/REC-CSS2/visure...ed-positioning

The CSS2 people knew about headers, footers, and sidebars, etc. They put some
stuff into their Recommendation. But we all know that "absolute positioning"
is hopeless at handling footers! So perhaps we should assume that "absolute
positioning" was not intended to handle page layout? Or was it intended to
handle footers, but they botched it? You tell me!

Just look at the 9.6.1. of the CSS2 specification! Fixed positioning suggested
as an alternative to Frames for laying out the viewport. But does it even hint
at absolute positioning as an alternative for layout tabes for laying out a
page? I think not.
Correct. We actually have 2 (or more) imperfect schemes, neither of
which was designed for the purpose. We are better off taking the best of both, until something better comes along.


I'll agree to this degree - CSS isn't 'perfect'. But it's still pretty
new. A decade from now, we'll have long forgotten about any benefits
from table-markup layout.


No we won't! Let's be clear. When did W3C say "Tables should not be used
purely as a means to layout document content ..."? It appeared in the
September 1997 working draft for the HTML 4.0 specification (which became a
formal recommendation in that December). It is now well over 6 years since
they said that. That is not "pretty new"!

"A brief history of tables"
http://www.barry.pearson.name/articl...es/history.htm

Tables were proposed by Dave Raggett late in 1993, and appeared in a browser
late in 1994. They "took off" as a layout technique in 1995/6. That is how
fast (2 years or so) things work when there is a drive for them to do so.

Yet, perhaps 5.5 years or more after CSS2 became a recommendation, somewhere
between 95% and 99% or so of all pages uploaded to the web use table-layout.

I don't believe CSS positioning is simply a displacement technology. It is
taking too long. I think it may be an additional technology. So it may live
side-by-side for a long time.Until something else, perhaps CSS3, changes the
balance.

--
Barry Pearson
http://www.Barry.Pearson.name/photography/
http://www.BirdsAndAnimals.info/
http://www.ChildSupportAnalysis.co.uk/
Jul 20 '05 #45

P: n/a
On Sun, 01 Feb 2004 11:56:26 -0500, Neal <ne*****@spamrcn.com> wrote:
Mary's husband loves seeing snow. Unfortunately he's stationed in the
desert. So she goes online searching for images of snowscapes, preferably
not far from home, to send to him.

She soon finds that if she were searching for pictures of "Click here to
enlarge" she'd have had better luck. Until she finds your gallery, where
the alt tags explain what the pictures are.

Don't be so sure blind folks aren't interested in images.


OK. I stand corrected ;) but to be honest this has nothing to do with
CSS. This is just a case of an HTML tag gone AWOL.
Jul 20 '05 #46

P: n/a
In article <gp**************@newsfep3-gui.server.ntli.net>,
"Barry Pearson" <ne**@childsupportanalysis.co.uk> writes:
always will. Look at this from David Siegel in November 1995:


You are aware that David Siegel later retracted that view and issued
a public apology for the damage he'd done in teaching bad practice?

--
Nick Kew
Jul 20 '05 #47

P: n/a
"Barry Pearson" ...
Neal wrote: .... [snip]
I think you will agree that 10 years down the road, table markup will
not be the best tool for layout.
...... Why will anyone have to redesign their sites? Table-layout sites will work at least as well in 10 years (and 20 years time) time as they do today.


Surely not when they are being rendered
on a PDA.

I think the statement 'in 10 years time' is taking
into account cases where the html is being rendered
on devices that are not 17" monitors with min
800x600 resolution...

--
Andrew Thompson
* http://www.PhySci.org/ Open-source software suite
* http://www.PhySci.org/codes/ Web & IT Help
* http://www.1point1C.org/ Science & Technology
Jul 20 '05 #48

P: n/a
Barry Pearson wrote:
Neal wrote:
10 years down the road, table markup will not be the best tool
for layout. It is far more likely any progress in layout tools
will be besed on the CSS model than on the HTML model.
Especially since HTML has had little to say about layout outside of
the 3.2 spec. And that version was essentially abandoned in favor of
html 4 and css for presentation suggestions.
So, by doing a table layout you're guaranteeing that down the
road you'll have to redesign the whole site to take advantage of
the best that web layout options have to offer. By using CSS for
layout, in time the layout model will be based on what I'm
already using.
In 10 years time, perhaps XHTML 2.0 will be the best standard for
web pages. Does that mean we should all be using XHTML 2.0 now?


Given that XHTML 2.0 is not currently supported, no. CSS, by contrast,
is a currently implemented technology.
Some of us have been seeing the claims of snake-oil merchants for
decades. We recognise the tactics of salespeople who say "your real
requirements are not what you say, they are really these different
requirements, and we can satisfy your real-real requirements".
Someone hires a physicist to consult, and tells the scientist that he
needs to defy gravity for a project. The only ethically correct reply
is that the client cannot defy gravity, and must use something else to
achieve his goals.
Why not, instead, say "yes, we will deliver your exact layout using
CSS positioning techniques"? Do you lack the confidence to do
that?
He lacks the ability to do that. As do you and I. It is not possible
in the www/html context to deliver an "exact" replication of a
layout. Anyone who claims that he can is the real snake oil salesman.
obviously unsupported b*llsh*t like "Stop abusing markup and use
this new method of layout" won't get anywhere!


Using blockquote for an indent is abusing markup. That is not bull. It
is quite obvious to anyone with a cursory knowledge of html.

--
Brian (follow directions in my address to email me)
http://www.tsmchughs.com/

Jul 20 '05 #49

P: n/a
It seems "Nick Kew" wrote in
comp.infosystems.www.authoring.stylesheets:
You are aware that David Siegel later retracted that view and issued
a public apology for the damage he'd done in teaching bad practice?


I was intrigued enough to did through Google for this, and here's a
link to "The Web is Ruined and I Ruined it":
http://www.xml.com/pub/a/w3j/s1.people.html
But it doesn't read like an apology, it reads like an attack. He
sure seems to be saying that <i>...</i> is a better way to mark up a
newspaper title because the alternative is
<newspapertitle>....</newspapertitle>.

Did you have an actual apology of Siegel's in mind?

--
Stan Brown, Oak Road Systems, Cortland County, New York, USA
http://OakRoadSystems.com/
HTML 4.01 spec: http://www.w3.org/TR/html401/
validator: http://validator.w3.org/
CSS 2 spec: http://www.w3.org/TR/REC-CSS2/
2.1 changes: http://www.w3.org/TR/CSS21/changes.html
validator: http://jigsaw.w3.org/css-validator/
Jul 20 '05 #50

59 Replies

This discussion thread is closed

Replies have been disabled for this discussion.