Won't validate...why not?

[Beauregard T. Shagnasty]

Why does everyone assume I'm saying things I'm not saying? Nowhere
have I said that that my homepage has CSS.

I said I've "just started" to use CSS.

I realize that.

I started my site four years ago. Only the most recent section in
both sites has CSS, and the ones which are currently in production.
I have no intention of converting the older ones on my own site:
there are over 300 pages: by the time browsers won't support them,
any useful info in them will be dated, so I'll probably just delete
them.

Browsers will never stop supporting tables, as table markup is what
you use for .. tabular data. The validator and the browser doesn't
know if what you display is actually related data, or if you just used
the tables for layout and positioning.

This is my point. Since it looks essentially similar in all four RO
browsers as it does in the pc and Mac browsers I've looked in, I
have a wider range of similarity than do sites using only CSS for
presentation.

I'd agree with that, since the Risc browsers are at about version 4
state. IE 4, Netscape 4 ... Those did some CSS but not well at all.

I understand the principle. But to be totally correct, I have to
add quite a lot of stuff I didn't have before. I actually can't
remember what (apart from <p> </p> which I didn't previously put
in), but I do remember when I converted my recent subsection, I was
gaining bytes rather than losing them on each page.

With a full conversion, you would have removed all the <table><tr><td>
elements, and replaced them with about three <div>s, some <p>s, and

Oh, right, I see what you mean. But it would mean abandoning a few
deprecated, but still widely supported, elements which I choose to
use, like v-align="top". Of course, I don't *have* to use these,
which I think is your point.

Exactly. ;-)

 

[Toby Inkster]

I chose a light green background colour and a dark rose foreground colour
[...] If the bgcolor and text are set in the HTML, I'll see what the
author has set in Fresco. If not, I'm usually right in assuming the
author has used CSS.

What about on this page?

http://examples.tobyinkster.co.uk/no-css
(Please ignore all the broken links.)

If you view it in your browser, it will have a light green background and
a dark rose foreground. But there's no CSS.

As a page author, I simply assume that if you've set your own customised
foreground and background colours, it must be because you really love
those colours. Why should I over-ride them with anything else?

 

[Big Bill]

In message <[email protected]>

Actually, I wasn't strictly correct.
Lots of sites don't validate but seem to read perfectly well in Fresco.
As you suggest, the problem is generally with older sites written in Front
Page.
Nowadays, the problem is sites written totally in Flash.
You're obviously not advocating doing that.
Even though at least one college round here is actually teaching future 'web
designers' to do just that :-(((

Goody! More work for me!

BB

 

[Liz]

In message <[email protected]>

Design width is another (well-discussed) topic, not relevant here. On
your PC, reduce the browser window width to something comfortable.

You obviously test your site(s) with more computer-savvy people than I do.

Whereas you and I know how to resize a window, I've never seen one of my
'volunteers' do it. (Not a random group: they're friends and acquaintances I
would consider to be in my 'target' audience (age/income/interest) for
content: all of them are 'on the internet', but most of them don't seem to
be computer savvy. My Mac tester's a geek: he tests my site on whatever Mac
browsers he has installed at the time, I test his on four of the ROS ones (I
don't think the oldest one is used much now, apparently it doesn't even
render Frames).

Actually, when I watch weans at school on the net (teenagers), about half
seem not to resize windows either, even when too small.

Also quite a few people don't seem to notice the wee 'x' which shows you can
turn 'favourites' off, so that even twhen they're on a low res (like
800x600) they have favourites running down the side (in IE) so have to
scroll horizontally. Contrary to what all the 'pundits' say, I've never
heard anhone complain about having to do this! (They do complain about sites
being all spread out, and I usually give in eventually and show them how to
resize.)

My best friends have never had a desktop computer. They're not stupid: she's
a graduate, he ran a nuclear power station over a mainframe computer (well,
I suppose it was a mainframe, he was vague about the details - made by
Ferrari).

Anyway, they got BushInternetTV (a *horrendous* system - you have to write
your emails live online!) so by means of a welcome I emailed them and
included a link to a page in my site I knew they'd like.
They soon phoned and said they liked the page, didn't I have any more pages?
There were 14 links on the page, and I said so.
They phoned back and said they couldn't get anywhere else.
Eventually, having checked the links live and finding they all worked, I
invited myself round to their house and discovered that they didn't know
that when text was in a different colour and underlined that was a link, and
so were the pretty obvious buttons at the bottom of the page.
When I pointed this out, the woman (who's a retired teacher) said "I was
going to tell you off for that: I couldn't see why you were emphasising
these words -it looks really childish, but I didn't like to say over the
phone!"

Well, how would someone who's never used a desktop computer before know that
you had to click on links? She was apparently expecting something like
'next' or 'page 2' or somesuch.

I thought back to see if I could remember how I knew, and I think it was
because I started at the end of the 'click here' era.

Slainte

Liz

--

 

[Liz]

In message <[email protected]>

I chose a light green background colour and a dark rose foreground colour
[...] If the bgcolor and text are set in the HTML, I'll see what the
author has set in Fresco. If not, I'm usually right in assuming the
author has used CSS.

What about on this page?

http://examples.tobyinkster.co.uk/no-css
(Please ignore all the broken links.)

If you view it in your browser, it will have a light green background and
a dark rose foreground. But there's no CSS.

As a page author, I simply assume that if you've set your own customised
foreground and background colours, it must be because you really love
those colours. Why should I over-ride them with anything else?

You also must test your sites on more computer-savvy people than I do.
Fair enough if that's your target audience.
:-/
I think it wasn't until I began learning HTML that I discovered that I could
change my browsers' defaults - and I'd been online about 18 months before
then.

BTW - When I came off your page, I noticed that all the blacks in my desktop
had changed to light blue: that hasn't happened for *ages*. No matter, all I
have to do is change monitor mode momentarily then change back. Less
bothersome than changing my browser totally which one site posted here
recently did: that necessitates a soft-reset at least 25% of the time and an
alt-break the rest of the time to close the browser.

Slainte

Liz

 

[Titus A Ducksass - AKA broken-record]

While sitting in a puddle Spartanicus scribbled in the mud:

Why?

,----[ http://w3.org/TR/html4/sgml/loosedtd.html ]
| Authors should use the Strict DTD when possible, but may use the
| Transitional DTD when support for presentation attribute and elements is
| required.
`----

This is 2004. Browsers which lack support for style sheets generally also
lack support for presentational attributes and elements also. There is no
need to use Transitional this millenium.

Except when the writer doesn't know enough about CSS and has to resort
to some clunky code.

I know a fair bit about CSS but don't know enough and don't have the
time to learn everything all at once - I try to learn another bit
every week and am working towards the strict standard but will fall
back to transitional when needed.

<A message to top posters. Type your reply here>

 

[Lauri Raittila]

I thought you meant a browser which actually crashed when Transitional was
used.

I have seen every brower crash on perfectly correct and even strict code.
But that surely don't happen in all transitional pages.

 

[Liz]

In message <[email protected]>

I have seen every brower crash on perfectly correct and even strict code.
But that surely don't happen in all transitional pages.

Oh that's true.
On one of the assessments on the course I did, they asked in a
multiple-choice questions about which fragments of code were 'legal': all of
them validated, but only one was legal. I hadn't realised that there would
be pages which were 'illegal', but would validate, and would fail in most
browsers.

It's not something I'd ever read about before - or since!

Liz

 

[Lauri Raittila]

Since it looks essentially similar in all four RO browsers as it does in the
pc and Mac browsers I've looked in, I have a wider range of similarity than
do sites using only CSS for presentation.

Why would it need to do similar?

And sometimes 90%.

The stylesheet is only downloaded once, then cached.

The stylesheet is often smaller than bytes saved on one page...

But to be totally correct, I have to add quite a lot of stuff I didn't have
before. [snip] ,but I do remember when I converted my recent subsection,
I was gaining bytes rather than losing them on each page.

Well, if you care about bytes, there just is nothing better than strict
HTML + external CSS. Even plain text is heavier sometimes. (after all, it
needs much more line breaks...)

But if you just add stuff, whitout rethinking it, of course it will be
bigger.

 

[Starshine Moonbeam]

In message <[email protected]>


Why does everyone assume I'm saying things I'm not saying?
Nowhere have I said that that my homepage has CSS.
I said I've "just started" to use CSS.
I started my site four years ago.
Only the most recent section in both sites has CSS, and the ones which are
currently in production. I have no intention of converting the older ones on
my own site: there are over 300 pages: by the time browsers won't support
them, any useful info in them will be dated, so I'll probably just delete them.

This is my point.
Since it looks essentially similar in all four RO browsers as it does in the
pc and Mac browsers I've looked in, I have a wider range of similarity than
do sites using only CSS for presentation.


I understand the principle.
But to be totally correct, I have to add quite a lot of stuff I didn't have
before. I actually can't remember what (apart from <p> </p> which I didn't
previously put in), but I do remember when I converted my recent subsection,
I was gaining bytes rather than losing them on each page.

Oh, right, I see what you mean.
But it would mean abandoning a few deprecated, but still widely supported,
elements which I choose to use, like

v-align="top".

valign="top". (deprecated)

The <whatever> tag is an element.

valign="top" would be an attribute.

td {vertical-align: top}

 

[nice.guy.nige]

While the city slept, Liz ([email protected]) feverishly typed...

[Browsers on RiscOS]

All of them support tables used for presentation, which I realise is
deprecated; in fact hereabouts it is a capital offence.

Tables for presentation have not been deprecated, as they were never
intended for this purpose in the first place. They are for grouping tabular
data. Presentational HTML itself has been deprecated though.

I'm always grateful when a site uses Transitional (they look so awful
here otherwise, even if the information is 'legally' there [...]
There is clearly a need for me, and other RiscOS users, to use
Transitional.

No there isn't - rather you need to accept that properly authored code is
not going to render as nicely on your system as it will on more
standards-compliant software available for other OS's (although properly
authored code *should* degrade gracefully). Instead of asking the entire web
development community to take a step backwards for your benefit, maybe you
should be asking the developers of your UA's to take a few steps forward.

I'm sure I read somewhere that some huge proportion of
websites don't validate. Surely validating 4.01 Transitional is
better than not validating at all?

As I understand it, the Transitional doctype is a "grace" doctype, to assist
developers who may have a lot of legacy documents to convert to the current
standards. Think of it as a halfway house where you can get your documents
to a certain level of validation as quickly as possible before taking the
next leap to Strict code. It is still only intended as a temporary measure -
the W3C to not expect you to get your legacy pages to Transitional
validation and then leave them as they are, you do also need to go for the
next stage. Neither is it intended to be an "easier" doctype for people to
code new documents to because they can't get their heads around producing
Strict code. Again, this is how I understand it, based on a document I read
some years ago. YMMV.

Cheers,
Nige

 

[nice.guy.nige]

While the city slept, Beauregard T. Shagnasty ([email protected])
feverishly typed...

All Risc users in this group: please identify yourself! <g>

I take it the use of the singular was intentional? ;-)

Cheers,
Nige

 

[Beauregard T. Shagnasty]

While the city slept, Beauregard T. Shagnasty
([email protected]) feverishly typed...


I take it the use of the singular was intentional? ;-)

I was wondering if anyone was going to catch that. Kudos to nige!

 

[Liz]

In message <[email protected]>

No there isn't - rather you need to accept that properly authored code is
not going to render as nicely on your system as it will on more
standards-compliant software available for other OS's (although properly
authored code *should* degrade gracefully). Instead of asking the entire web
development community to take a step backwards for your benefit, maybe you
should be asking the developers of your UA's to take a few steps forward.

I have *not* asked the entire development community to take a step backwards.
As you can clearly read, I said "There is clearly a need for me, and other
RiscOS users, to use Transitional." I have said more than once in this
thread that I *don't* expect other people to do likewise. I also said earlier
in the thread that I have often complained in the RiscOS lists about
lack of development of browsers (they think I'm a troll!), but I can't
actually *force* someone to write one if they don't want to. It's thought to
be probably commercially unviable, because so many of the geeky-types don't
care what a page looks like as long as it has the info that they wouldn't
pay for a better browser anyway! I have every sympathy with the students who
are writing the pd one from scratch for it not being their top priority.
And I can't program and don't have the time or inclination to learn.

As I understand it, the Transitional doctype is a "grace" doctype, to assist
developers who may have a lot of legacy documents to convert to the current
standards. Think of it as a halfway house where you can get your documents
to a certain level of validation as quickly as possible before taking the
next leap to Strict code. It is still only intended as a temporary measure -
the W3C to not expect you to get your legacy pages to Transitional
validation and then leave them as they are, you do also need to go for the
next stage. Neither is it intended to be an "easier" doctype for people to
code new documents to because they can't get their heads around producing
Strict code. Again, this is how I understand it, based on a document I read
some years ago. YMMV.

That's how I understand it too.
Who said anything about 'not being able to get their heads around producing
Strict code'?
But you can't expect me or others like me to go around preparing pages which
look rubbish on our browsers. That's totally unreasonable. For everyone
else, I'd just rather they didn't have sites which fail when we want to buy
something: to my mind, if Amazon (up 'til now, the recent phenomenon is
worrying) Land's End and Victoria's Secret* can do it, everyone can. I hate
it when I spend 20 or 30 minutes on a site choosing an order and failing at
the very last hurdle. :-(

(*I have to admit that I was astonished that I was successful making an
order at VS: they don't actively market in the UK and I'd bet I was the
first RiscOS customer they've ever had. Certainly I've never seen it in a
list of RiscOS friendly sites - but it will be now!)

Slainte

Liz

 

[Leonard Blaisdell]

While sitting in a puddle Beauregard T. Shagnasty scribbled in the mud:

can i use my keyboard too?

Only at your own risc. Nyuk, nyuk!

leo

 

[Leonard Blaisdell]

<very much guessing here>
Perhaps the people utilizing CSS are different people, in nature I mean?
More likely to get into things, technical things. More eager to do thing
in the best way possible. More able to addept to new things.
Whereas the table-layout people may be just keep doing what they have
always done, because they found that hard enough to learn in the first
place. Switching to CSS scares them?
< />

I think you've identified both parts of the equation very successfully.
The table based layout people are following a tradition that started with
Gutenberg. For five hundred years, printers have been creating a specific
though evolutional layout for the printed page. It's been refined. It's
been accepted as state of the art, and it is for fixed media design. Table
based layout follows fixed media.
Now we have a "rubber band" page that stretches by browser window or
shrinks/increases dramatically by increase/decrease in computer
resolution.
The medium doesn't fit traditional ideas of design.
What is webdesign software to do? What do people expect? They expect what
they've seen all their lives. The web is scarcely more than ten years old.
Print designers go into web design. Software accomodates their ideas. The
W3C doesn't. It isn't beholding to Gutenberg. It's creating a new and very
different method of disseminating information.
There is a very big difference between content and presentation. It will
become apparent in the future. A table is a table. A list is a list. A
paragraph is a paragraph. A span separates us and a div is what we live
with now. I'm taking a br.

leo

 

[Duende]

While sitting in a puddle Leonard Blaisdell scribbled in the mud:

Only at your own risc. Nyuk, nyuk!

Ok Moe.
BTW, how was Hot August Nights this year. I went to The Sturgis Motorcycle
Rally again this year. Over 500,000 bikes this year. Enough to make me feel
like I'm gitting old.

 

[Leonard Blaisdell]

Ok Moe.
BTW, how was Hot August Nights this year. I went to The Sturgis Motorcycle
Rally again this year. Over 500,000 bikes this year. Enough to make me feel
like I'm gitting old.

They say it was a success as always. It must have been. I kept running
over those jalopies on the way to the store. If you ever come, let me
know, and I'll go to my first HAN ever.

leo

 

[Duende]

While sitting in a puddle Leonard Blaisdell scribbled in the mud:

I kept running over those jalopies

You shell go to hell for doing that!

on the way to the store. If you ever
come, let me know, and I'll go to my first HAN ever

I never can find a place to park my bus.

 

[Leonard Blaisdell]

I never can find a place to park my bus.

The Nugget! Big RV lot. Come early. Don't gamble. Eat well. See jalopies
before I run into them.

leo