Complicated Question?

[Mark Johnson]

Oddly enough, in searching Google for web and ngs, I could find only a
handful of articles on this problem. I wonder if it's one not
typically encountered?

I wanted to have three divs that could float, if need be. They are
inside another div. I put a background color on it. I float all three
inside it to the left.

IE works fine, just as you'd expect. But Firefox, while it renders
everything else the same, widens out the containing box and forces a
horizontal scrollbar.

I can't imagine why. If I remove the float, the scrollbar goes away.

[Mark Johnson]

Mark Johnson <10*******@compuserve.com> wrote:

Oddly enough, in searching Google for web and ngs, I could find only a
handful of articles on this problem. I wonder if it's one not
typically encountered?

I wanted to have three divs that could float, if need be. They are
inside another div. I put a background color on it. I float all three
inside it to the left.

IE works fine, just as you'd expect. But Firefox, while it renders
everything else the same, widens out the containing box and forces a
horizontal scrollbar.

I can't imagine why. If I remove the float, the scrollbar goes away.

If I might answer my own question, it probably is due simply to the
difference between 'quirks' and 'strict' mode in Firefox .8 or .9.

I prefer quirks, but would probably be able to render in Firefox using
strict. As it is Firefox kicks out extra-wide divs, spans, what have
you. IE isn't as bad. But switching to strict causes problems in IE,
as well.

Quirks mode is typically had by not including a URL in the DOCTYPE. If
using XSLT, that means including no "doctype-system" parameter. If you
do include the URL/URI then typically browsers kick into a strict
mode, which can completely change the rendering of your style sheets.

One day, I may decide to go transitional 4.01 strict, instead of
quirks as of now. But it's a lot of work to rework all those style
sheets just so I won't get a horizontal scroll bar in Firefox - which
browser hardly anyone uses. It'll take time. It might be worth it.

[Lauri Raittila]

In article <cl********************************@fe02.buzzardne ws.com>,
10*******@compuserve.com says...

Oddly enough, in searching Google for web and ngs, I could find only a
handful of articles on this problem. I wonder if it's one not
typically encountered?

I wanted to have three divs that could float, if need be. They are
inside another div. I put a background color on it. I float all three
inside it to the left.

IE works fine, just as you'd expect. But Firefox, while it renders
everything else the same, widens out the containing box and forces a
horizontal scrollbar.

I can't imagine why. If I remove the float, the scrollbar goes away.

URL?

--
Lauri Raittila <http://www.iki.fi/lr> <http://www.iki.fi/zwak/fonts>

[Mark Johnson]

Lauri Raittila <la***@raittila.cjb.net> wrote:

In article <cl********************************@fe02.buzzardne ws.com>,
10*******@compuserve.com says...

Oddly enough, in searching Google for web and ngs, I could find only a
handful of articles on this problem. I wonder if it's one not
typically encountered?

I wanted to have three divs that could float, if need be. They are
inside another div. I put a background color on it. I float all three
inside it to the left.

IE works fine, just as you'd expect. But Firefox, while it renders
everything else the same, widens out the containing box and forces a
horizontal scrollbar.

I can't imagine why. If I remove the float, the scrollbar goes away.

URL?

Box model differences between strict and quirks mode.

Frankly, I didn't realize that strict mode will NOT allow custom
attributes. I require attributes of my own choosing for certain
functionality. I literally can't use strict mode. And it seems to be
more trouble than it's worth. The only advantage would be that IE and
Firefox, and NN if they're still keeping that going, and various
versions more importantly (MAC anyone?) would all display the page the
same way, and have available the same DOM for javascript. But I'm told
it doesn't even always work that way.

That being said, I find in Firefox, for example, no real difference in
switching from strict to quirks except that one MUST use the 'px'
suffix for setting pixel values. IE isn't so well behaved.

The way you switch, at least in IE 6, is simply to use a doctype with
something for the type and something for the url. Could just be a
letter "x" in both cases. But using the second value kicks you into
strict mode, and all the problems that still presents in 2004. And
leaving off the second DOCTYPE argument keeps you in default quirks
mode, which you get without any DOCTYPE line at all.

[Lauri Raittila]

In article <al********************************@fe02.buzzardne ws.com>,
10*******@compuserve.com says...

Lauri Raittila <la***@raittila.cjb.net> wrote:

In article 10*******@compuserve.com says...

I can't imagine why. If I remove the float, the scrollbar goes away.

URL?

Box model differences between strict and quirks mode.

Frankly, I didn't realize that strict mode will NOT allow custom
attributes.

Can't think what box modell has to do with custom attributes. Solve
problem by getting rid of custom attributes.

http://www.hut.fi/u/hsivonen/doctype.html

--
Lauri Raittila <http://www.iki.fi/lr> <http://www.iki.fi/zwak/fonts>

[Mark Johnson]

Lauri Raittila <la***@raittila.cjb.net> wrote:

In article <al********************************@fe02.buzzardne ws.com>,
10*******@compuserve.com says...

Lauri Raittila <la***@raittila.cjb.net> wrote:

>In article 10*******@compuserve.com says...
>> I can't imagine why. If I remove the float, the scrollbar goes away.
>
>URL?

Box model differences between strict and quirks mode.

Frankly, I didn't realize that strict mode will NOT allow custom
attributes.

Can't think what box modell has to do with custom attributes. Solve
problem by getting rid of custom attributes.

Nobody said it had anything to do with it. I said the difference in
display was due to different models. But part of the strict mode, as I
understand it, is that custom attributes are not allowed. That's
unacceptable, and apparently is seen as such by many on the web - from
what I've read. One may well require attributes for functionality,
which is more efficient than using nested elements, as well.

Many may not approve of javascript, itself, of vbscript or java or
whatever else as anything but rude hacks. But short of compiling
plug-ins for everything, which you have to get people to download,
client side solutions are often very elegant.

Strict mode, in my opinion, doesn't seem 'ready for primetime'.

[Lauri Raittila]

Mark Johnson wrote;

Lauri Raittila <la***@raittila.cjb.net> wrote:

In article <al********************************@fe02.buzzardne ws.com>,
10*******@compuserve.com says...

Lauri Raittila <la***@raittila.cjb.net> wrote:
>In article 10*******@compuserve.com says...
>> I can't imagine why. If I remove the float, the scrollbar goes away.
>
>URL?

Box model differences between strict and quirks mode.

Frankly, I didn't realize that strict mode will NOT allow custom
attributes.
Can't think what box modell has to do with custom attributes. Solve
problem by getting rid of custom attributes.

Nobody said it had anything to do with it. I said the difference in
display was due to different models. But part of the strict mode, as I
understand it, is that custom attributes are not allowed.

Well, they aren't allowed in any mode. I think you have confused few
things. For example, what you mean by custom attributes?
That's
unacceptable, and apparently is seen as such by many on the web - from
what I've read. One may well require attributes for functionality,
which is more efficient than using nested elements, as well. Many may not approve of javascript, itself, of vbscript or java or
whatever else as anything but rude hacks. But short of compiling
plug-ins for everything, which you have to get people to download,
client side solutions are often very elegant.
URL? Browsers conserned.
Strict mode, in my opinion, doesn't seem 'ready for primetime'.

What you mean by strict mode?

It doesn't help anybody that you answer yourself on questions that are
uncomprehansible for anyone else. I have no idea what you are doing, or
why the hell you need custom attributes. Of why you can't tricker right
mode.

--
Lauri Raittila <http://www.iki.fi/lr> <http://www.iki.fi/zwak/fonts>

[Brian]

Mark Johnson wrote:

I said the difference in display was due to different models.
Maybe. But since you still haven't provided a url, we can't be sure.
But part of the strict mode, as I understand it, is that custom
attributes are not allowed.
Custom attributes -- by that I assume you mean ones that you've made up
for your own use -- are by definition not allowed in any version of
HTML. But what has this to do with strict or quirks rendering mode? You
seem to be conflating HTML document type definitions with browser
rendering modes. You may want to read how browsers switch between modes.

http://www.hut.fi/u/hsivonen/doctype
http://www.mozilla.org/docs/web-developer/quirks/
Strict mode, in my opinion, doesn't seem 'ready for primetime'.

Until you have a clear idea of the issues involved, your opinion is not
likely to hold much sway around here.

--
Brian (remove "invalid" to email me)

[Mark Johnson]

Lauri Raittila <la***@raittila.cjb.net> wrote:

Mark Johnson wrote;

Lauri Raittila <la***@raittila.cjb.net> wrote:

>In article <al********************************@fe02.buzzardne ws.com>,
>10*******@compuserve.com says...
>> Lauri Raittila <la***@raittila.cjb.net> wrote:
>> >In article 10*******@compuserve.com says...
>> >> I can't imagine why. If I remove the float, the scrollbar goes away.

Nobody said it had anything to do with it. I said the difference in
display was due to different models. But part of the strict mode, as I
understand it, is that custom attributes are not allowed.

Well, they aren't allowed in any mode. I think you have confused few
things. For example, what you mean by custom attributes?
Define terms, you mean? That should come first.

Custom attributes. Meta-attributes. Attributes you've never seen. Etc.

unacceptable, and apparently is seen as such by many on the web - from
what I've read. One may well require attributes for functionality,
which is more efficient than using nested elements, as well. Many may not approve of javascript, itself, of vbscript or java or
whatever else as anything but rude hacks. But short of compiling
plug-ins for everything, which you have to get people to download,
client side solutions are often very elegant. URL? Browsers conserned.
You lost me. Maybe you could elaborate beyond just three words.

Strict mode, in my opinion, doesn't seem 'ready for primetime'.

What you mean by strict mode?

Standards mode, then, as opposed to quirks mode. Not strict,
necessarily. Standards mode. But I might just loosely call it, strict,
because it seems more constricting, less flexible, and so more
descriptive.

[Mark Johnson]

Brian <us*****@julietremblay.com.invalid> wrote:

Mark Johnson wrote:

But part of the strict mode, as I understand it, is that custom
attributes are not allowed.

Custom attributes -- by that I assume you mean ones that you've made up
for your own use -- are by definition not allowed in any version of
HTML. But what has this to do with strict or quirks rendering mode?
I understood, and excuse me for that but standards mode if that's
better, that standards mode did not allow custom attributes. If they
aren't accessible by code, then functionality - necessary
functionality - might be lost.

Until you have a clear idea of the issues involved, your opinion is not
likely to hold much sway around here.

It wouldn't be in any case. And oddly enough, I'm not bothered by
that.

[Brian]

Mark Johnson wrote:

Brian wrote:

Mark Johnson wrote:

But part of the strict mode, as I understand it, is that custom
attributes are not allowed.

Custom attributes -- by that I assume you mean ones that you've
made up for your own use -- are by definition not allowed in any
version of HTML. But what has this to do with strict or quirks
rendering mode?

I should not have adopted your choice of words. To clarify: The
rendering modes are standards, almost standards, and quirks for Mozilla;
standards and quirks for MSIE 6/Win; and standards and quirks mode for
Opera. I don't think any other browser plays the dtd guessing game.

Strict and transitional (or loose) are terms used by the W3C to
distinguish between various flavors of HTML 4 and XHTML 1.0.
I understood, and excuse me for that but standards mode if that's
better, that standards mode did not allow custom attributes.

You missed my point. Browsers do not switch rendering modes based on
whether the document validates or not. They determine which mode to use
based on whether a doc type declaration is present, and if so, which one.

--
Brian (remove "invalid" to email me)

[Lauri Raittila]

Mark Johnson wrote;

Brian <us*****@julietremblay.com.invalid> wrote:

Mark Johnson wrote:

But part of the strict mode, as I understand it, is that custom
attributes are not allowed.

Custom attributes -- by that I assume you mean ones that you've made up
for your own use -- are by definition not allowed in any version of
HTML. But what has this to do with strict or quirks rendering mode?

I understood, and excuse me for that but standards mode if that's
better, that standards mode did not allow custom attributes. If they
aren't accessible by code, then functionality - necessary
functionality - might be lost.

Ever thought about about making it normal way?
Until you have a clear idea of the issues involved, your opinion is not
likely to hold much sway around here.

It wouldn't be in any case. And oddly enough, I'm not bothered by
that.

*blonk*

--
Lauri Raittila <http://www.iki.fi/lr> <http://www.iki.fi/zwak/fonts>

[Mark Johnson]

Brian <us*****@julietremblay.com.invalid> wrote:

You missed my point. Browsers do not switch rendering modes based on
whether the document validates or not. They determine which mode to use
based on whether a doc type declaration is present, and if so, which one.

You're repeating back to me what _I_ originally wrote.

[Mark Johnson]

Lauri Raittila <la***@raittila.cjb.net> wrote:

Mark Johnson wrote;

Brian <us*****@julietremblay.com.invalid> wrote:

>Mark Johnson wrote:

>> But part of the strict mode, as I understand it, is that custom
>> attributes are not allowed.

>Custom attributes -- by that I assume you mean ones that you've made up
>for your own use -- are by definition not allowed in any version of
>HTML. But what has this to do with strict or quirks rendering mode?

I understood, and excuse me for that but standards mode if that's
better, that standards mode did not allow custom attributes. If they
aren't accessible by code, then functionality - necessary
functionality - might be lost.

Ever thought about about making it normal way?

Sometimes to add a 'thinnest client' functionality, you might find it
important to store information in HTML/XHMTL, not in elements but in
attributes.

By the way, I've nothing against a 'tried and true' method. And often
people tend to converge on one or a couple of methods in any field,
for certain tasks and situations. But such 'normalcy' is generally
understood and defined. What do YOU mean?

[Neal]

On Fri, 15 Oct 2004 19:13:28 -0700, Mark Johnson
<10*******@compuserve.com> wrote:

Brian <us*****@julietremblay.com.invalid> wrote:

You missed my point. Browsers do not switch rendering modes based on
whether the document validates or not. They determine which mode to use
based on whether a doc type declaration is present, and if so, which
one.

You're repeating back to me what _I_ originally wrote.

But it seemed from your messages that you misunderstood.

What actually happens to my knowledge is that the browser, say IE6, looks
for the precise string <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
"http://www.w3.org/TR/html4/strict.dtd"> and if it sees that, bingo, we're
in standards mode. If it sees <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01
Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd"> it does
something different. And so on. If it sees nothing like this, it is quirks
mode.

What happens if you use a custom DTD? In the case of IE, it won't match
anything in memory, so it cannot use standards mode, it will revert to
some other mode.

So all we're saying is that if you don't have a standardized strict
doctype definition at the start of the document, you can't expect a
browser to use standards mode, it well may not.

[Fred Oz]

Mark Johnson wrote:

Mark Johnson <10*******@compuserve.com> wrote:
sheets just so I won't get a horizontal scroll bar in Firefox - which
browser hardly anyone uses. It'll take time. It might be worth it.

According to W3schools, 17% of their visitors last month were using
browsers identifying as "Mozilla", a fair proportion of which would be
Firefox.

http://www.w3schools.com/browsers/browsers_stats.asp

So even if Firefox was only half of the 'zillas, that's 8% of surfers
you are deciding to ignore - and that's a lot of potential visitors.

Cheers, Fred.

[Mark Johnson]

Fred Oz <oz****@iinet.net.au> wrote:

Mark Johnson wrote:

Mark Johnson <10*******@compuserve.com> wrote:
sheets just so I won't get a horizontal scroll bar in Firefox - which
browser hardly anyone uses. It'll take time. It might be worth it.

According to W3schools, 17% of their visitors last month were using
browsers identifying as "Mozilla", a fair proportion of which would be
Firefox.

http://www.w3schools.com/browsers/browsers_stats.asp

So even if Firefox was only half of the 'zillas, that's 8% of surfers
you are deciding to ignore - and that's a lot of potential visitors.

I wouldn't ignore anyone. You serve out a page without anything. And
they see that. Maybe slightly better than plain text. But not much
more. Bullets. Heading. Paragraphs. Basic, presentable. It's fine. But
in fact, I'm trying to upgrade to Firefox, specifically. It's actually
better 'behaved' than IE. I notice no real difference between
'standards' and quirks mode in Firefox for either my javascript or
style sheets. IE wasn't the same. But I had to develop for IE,
initially, because it's still IE's game. Be realistic.

[Mark Johnson]

Neal <ne*****@yahoo.com> wrote:

On Fri, 15 Oct 2004 19:13:28 -0700, Mark Johnson
<10*******@compuserve.com> wrote: What actually happens to my knowledge is that the browser, say IE6, looks
for the precise string <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
"http://www.w3.org/TR/html4/strict.dtd"> and if it sees that, bingo, we're
in standards mode.

No, it kicks into standards mode even if you use "x" "x". Any text in
that second argument kicks it into standards. Try it and see.

Now, if you want to say that specific functionality is not either
available or suppressed in THIS way, and IS by specific strings as you
suggest, then be specific. What specific functionality? What specific
DOM elements? What specific CSS?

[Brian]

Mark Johnson wrote:

Brian wrote:

Browsers do not switch rendering modes based on whether the
document validates or not. They determine which mode to use based
on whether a doc type declaration is present, and if so, which one.

You're repeating back to me what _I_ originally wrote.

No, I did not. You were confusing standards mode with document
validation, as if the two were conntected. Here's what you wrote
elsewhere in this thread:

<quote>
....standards mode did not allow custom attributes.
</quote>

A browser's rendering mode is not affected by your use of invented
attributes. It is determined solely by the dtd.

--
Brian (remove "invalid" to email me)

[Mark Johnson]

Brian <us*****@julietremblay.com.invalid> wrote:

Mark Johnson wrote:

Brian wrote: You're repeating back to me what _I_ originally wrote.

No, I did not.

I was saying that you did. Guess we'll have to agree to disagree, and
both reread the previous messages.