By using this site, you agree to our updated Privacy Policy and our Terms of Use. Manage your Cookies Settings.
424,946 Members | 744 Online
Bytes IT Community
+ Ask a Question
Need help? Post your question and get tips & solutions from a community of 424,946 IT Pros & Developers. It's quick & easy.

Reasons to use HTML 4.01 Transitional

P: n/a
A web page can be created using either strict or transitional HTML.

The DOCTYPE for strict is:

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
"http://www.w3.org/TR/html4/strict.dtd">

And transitional:

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"
"http://www.w3.org/TR/html4/loose.dtd">

According to this document (written in 1999):
http://www.w3.org/TR/html401/sgml/loosedtd.html

"the HTML 4.01 Transitional DTD ... includes presentation attributes and
elements that W3C expects to phase out as support for style sheets matures."

My questions are:

Can someone please summarize reasons why transitional HTML would be used
over strict?

Of the elements which the W3C plans to phase out, are any still in common
use?
Thanks in advance.

Oct 30 '06 #1
Share this Question
Share on Google+
41 Replies


P: n/a
deko wrote:
Can someone please summarize reasons why transitional HTML would be
used over strict?
Use Transitional for those times you are working on a legacy page,
perhaps doing minor updates, and you don't have time to remove all those
silly <fonttags throughout the page.

Use Strict when creating new documents, using CSS for presentation and
not 1995-style coding.

--
-bts
-Motorcycles defy gravity; cars just suck
Oct 30 '06 #2

P: n/a
>Can someone please summarize reasons why transitional HTML would be
>used over strict?

Use Transitional for those times you are working on a legacy page,
perhaps doing minor updates, and you don't have time to remove all those
silly <fonttags throughout the page.

Use Strict when creating new documents, using CSS for presentation and
not 1995-style coding.
I've been looking for a good summary of attributes not supported in HTML strict.
As you've mentioned, <fontis one.

The only one I might want to use is <uso folks posting replies on a blog site
can mark up their comments with underlined text.

In regard to legacy sites, IE7 apparently had to walk a fine line here so as not
to break sites that display properly in IE6, yet still move IE7 towards
standards compliance. As I understand it, IE7 will interpret a page as strict
even if the DOCTYPE specifies transitional, as long as the URL is included.

Oct 30 '06 #3

P: n/a
deko wrote:
As I understand it, IE7 will interpret a page as strict even if the
DOCTYPE specifies transitional, as long as the URL is included.
That would be a mistake, in my opinion.

(No, I don't have IE7...)

--
-bts
-Motorcycles defy gravity; cars just suck
Oct 30 '06 #4

P: n/a
On Mon, 30 Oct 2006, Beauregard T. Shagnasty wrote:
deko wrote:
>As I understand it, IE7 will interpret a page as strict even if the
DOCTYPE specifies transitional, as long as the URL is included.

That would be a mistake, in my opinion.
deko still doesn't understand the difference between
[HTML 4] "Strict" and "Standards compliance mode".

Oct 30 '06 #5

P: n/a
Andreas Prilop wrote:
On Mon, 30 Oct 2006, Beauregard T. Shagnasty wrote:
>deko wrote:
>>As I understand it, IE7 will interpret a page as strict even if the
DOCTYPE specifies transitional, as long as the URL is included.

That would be a mistake, in my opinion.

deko still doesn't understand the difference between
[HTML 4] "Strict" and "Standards compliance mode".
Strict/Transitional vs. Standards/Quirks ... ;-)

--
-bts
-Motorcycles defy gravity; cars just suck
Oct 30 '06 #6

P: n/a
>On Mon, 30 Oct 2006, Beauregard T. Shagnasty wrote:
>>
>>deko wrote:

As I understand it, IE7 will interpret a page as strict even if the
DOCTYPE specifies transitional, as long as the URL is included.

That would be a mistake, in my opinion.

deko still doesn't understand the difference between
[HTML 4] "Strict" and "Standards compliance mode".

Strict/Transitional vs. Standards/Quirks ... ;-)
So HTML has different *document types* (strict and transitional), while browsers
have different *layout modes* (standards and quirks). Newer browsers can
operate in quirks mode to accommodate legacy pages (tag soup) that were authored
with dependence on a browser's quirks layout mode.

So to correct my statement regarding IE7, I should say that IE7 will accommodate
a legacy page (that violates W3C Recommendations) by operating in quirks mode
*unless* the DOCTYPE includes the URL (in which case IE7 operates in standards
compliance mode).

But getting back to my original question, are there any attributes or tags that
one should be aware of when switching from HTML transitional to HTML strict?

I mentioned the <utag, which some blog pages might be using to let people
underline parts of their comments. Aside from this, I don't think I have
anything to worry about. But I'm still concerned that something might break if
I just change all my DOCTYPEs.

Oct 30 '06 #7

P: n/a
deko wrote:
But getting back to my original question, are there any attributes or
tags that one should be aware of when switching from HTML transitional
to HTML strict?
Check <http://www.w3.org/TR/html4/index/elements.htmlfor which
elements are deprecated and which are only in the transitional DTD
(called loose DTD there to have one more term to confuse you :)).
There is also a similar table for attributes
<http://www.w3.org/TR/html4/index/attributes.html>.
--

Martin Honnen
http://JavaScript.FAQTs.com/
Oct 30 '06 #8

P: n/a
On Mon, 30 Oct 2006 09:09:12 -0800, "deko" <de**@nospam.comwrote:
>
The only one I might want to use is <uso folks posting replies on a blog site
can mark up their comments with underlined text.
You don't need the <u< tag for that.

Oct 30 '06 #9

P: n/a
On Mon, 30 Oct 2006 10:06:36 -0800, "deko" <de**@nospam.comwrote:

>I mentioned the <utag, which some blog pages might be using to let people
underline parts of their comments. Aside from this, I don't think I have
anything to worry about. But I'm still concerned that something might break if
I just change all my DOCTYPEs.
For pre-existing pages the transitional DTD exists to accomidate
depreciated tags.

For new pages, strict is recommended and best practice.
Oct 30 '06 #10

P: n/a
>But getting back to my original question, are there any attributes or tags
>that one should be aware of when switching from HTML transitional to HTML
strict?

Check <http://www.w3.org/TR/html4/index/elements.htmlfor which elements are
deprecated and which are only in the transitional DTD (called loose DTD there
to have one more term to confuse you :)).
There is also a similar table for attributes
<http://www.w3.org/TR/html4/index/attributes.html>.
Thanks, those pages help.

I may have a colspan or align out there somewhere...

As for authoring new pages, there appears to be no good reason to use
transitional HTML.

As for IE7 (standards-compliant layout mode), this browser is supposed to be
behaving itself now. But I've still seen some IE7 peculiarities in regards to
margins/padding. This may not have anything to do with W3C Recommendations, but
it still requires additional coding if I want pages to display properly.

Oct 30 '06 #11

P: n/a
>>The only one I might want to use is <uso folks posting replies on a blog
>>site
can mark up their comments with underlined text.

You don't need the <u< tag for that.
What is the alternative?

Oct 30 '06 #12

P: n/a
On Mon, 30 Oct 2006 11:01:33 -0800, "deko" <de**@nospam.comwrote:
>>>The only one I might want to use is <uso folks posting replies on a blog
site can mark up their comments with underlined text.
>You don't need the <utag for that.
>What is the alternative?
Best alternative is not to use underlines except on a hyperlink. To the
vast majority of users underlined text means that text is a hyperlink
and most people will click on it and get annoyed when it goes nowhere.

You can underline text in any element by applying the "text-decoration:
underline;" CSS rule to it.

As to what element you should use, this depends on the semantic meaning
of the text in question. If the underline is meant to express emphasis
(a bad practice, but if you must...) then surround it with an <emtag
and style accordingly. If it is meant to provide strong emphasis then
use the <strongtag and style accordingly.

But it's still best to avoid underlining anything that isn't a hyperlink
on a web page.

On a blog page you should probably not let your users underline anything
but a hyperlink. Unless perhaps there is a site-wide convention that
underline means something else and your users have learned to expect
that.

Oct 30 '06 #13

P: n/a
>>>>The only one I might want to use is <uso folks posting replies on a blog
>>>>site can mark up their comments with underlined text.
>>You don't need the <utag for that.
>>What is the alternative?

Best alternative is not to use underlines except on a hyperlink. To the
vast majority of users underlined text means that text is a hyperlink
and most people will click on it and get annoyed when it goes nowhere.

You can underline text in any element by applying the "text-decoration:
underline;" CSS rule to it.
understood.
As to what element you should use, this depends on the semantic meaning
of the text in question. If the underline is meant to express emphasis
(a bad practice, but if you must...) then surround it with an <emtag
and style accordingly. If it is meant to provide strong emphasis then
use the <strongtag and style accordingly.

But it's still best to avoid underlining anything that isn't a hyperlink
on a web page.

On a blog page you should probably not let your users underline anything
but a hyperlink. Unless perhaps there is a site-wide convention that
underline means something else and your users have learned to expect
that.
Avoiding underlined text where it might be confused with a hyperlink is a good
idea for usability sake, but underlining in the appropriate context, such as
underlining book titles, should not be discarded.

The more important question may be:

How can we let people markup their comments when responding in a Strict
HTML/XHTML formatted forum?

The nice thing about <u>, <iand <btags is that they provide non-technical
responders with an intuitive way to format their comments.

How about redefining the tags?

..user-comments u {
text-decoration: underline;
}

..user-comments b {
font-weight: bold;
}

..user-comments i {
font-style: italic;
}
Oct 30 '06 #14

P: n/a
VK
deko still doesn't understand the difference between
[HTML 4] "Strict" and "Standards compliance mode".
I guess you do though ;-)
In such case you may have time to explain to me why the presence of DTD
link in DOCTYPE affects on Global scope in javascript (with ID's added
as variables w/o DTD). I once twisted my mind around trying to find a
slightest relation between script Global space, DTD, DOCTYPE and the
sanity - no luck... :-) :-|
That was a great marketing move Microsoft did once in IE6 and now
everyone set on this pil. Even I did see the catch right away. Namely
they gave to other producers a way to implement "IE-like mode" while
leaving them an opportunity to keep swearing in commitment to standards
- and to be formally right. "What is your problem? We are standard,
just add DTD link. *Any* fricken link, we don't care."

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"
"http://www.sex.com">

puts Firefox into "Standard-compliant" mode as well as

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"
"http://www.w3.org/TR/html401/loose.dtd">

Maybe DOCTYPE's and DTD's are useless in (X)HTML in application to
their real purpose and mechanics, but at least no one was doing some
crazy show out of it as it is now all around (talking about UA's
producers).

To OP: browsers do not care what actually DTD points to. They do not
load DTD neither they validate against it. It means that if you use tag
that is not in the current DTD, the page will not crash and W3C cops
will not come knocking your door. :-)

At the same time it is a good idea to keep your document at least
formally valid (minimum for the personal satisfaction). Both HTML
Transitional and Strict is the last century crap (1999) so neither one
is more modern than other. The "burden of proof" is iframe. This
element became so commonly used that it's difficult to imagine that
some DTD may not have it - yet HTML Strict has not. So if you have to
use iframe on your page or if you foresee using it - then better stay
on Transitional with DTD link:

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"
"http://www.w3.org/TR/html401/loose.dtd">

Another possible reason for Transitional can be <utag as it became
widely used again for access keys. Unless that use HTML Strict:

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
"http://www.w3.org/TR/html401/strict.dtd">
P.S. Just to keep IE in "pseudo-standard-whatever" mode it is enough to
set Strict w/o DTD:
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN">
But formally a document is not valid until validated, and it cannot be
validated w/o DTD link to validate against. Again: it has nothing to do
with any practical outcome, but to be formally correct you may choose
to use Strict in its full form (with DTD linked).

Oct 30 '06 #15

P: n/a
To OP: browsers do not care what actually DTD points to. They do not
load DTD neither they validate against it. It means that if you use tag
that is not in the current DTD, the page will not crash and W3C cops
will not come knocking your door. :-)
10-4
But formally a document is not valid until validated, and it cannot be
validated w/o DTD link to validate against. Again: it has nothing to do
with any practical outcome, but to be formally correct you may choose
to use Strict in its full form (with DTD linked).
I suppose as long as a page displays correctly in target browsers, who cares
what W3C thinks. But my guess is Google and other search engines will adjust
your rank based on how miserably you fail, or pass validation.

Oct 30 '06 #16

P: n/a
In our last episode, <-P******************************@comcast.com>, the
lovely and talented deko broadcast on comp.infosystems.www.authoring.html:
Avoiding underlined text where it might be confused with a hyperlink is a
good idea for usability sake, but underlining in the appropriate context,
such as underlining book titles, should not be discarded.
Underlining book titles is a typewriter/handwriting hack to indicate
italics. Book titles belong in CITE and cite should styled according to
context (i.e. italics in a roman context, roman -- that is, normal -- in an
italics context), unless of course for some reason you are attempting to
imitate a typewriter document. Underlining still is the indication of
italics for documents which will be typeset by human beings (i.e. in paper
manuscripts) but it is unlikely that HTML is the ideal way to prepare such
documents.

--
Lars Eighner <http://larseighner.com/ <http://myspace.com/larseighner>
Behaviorism is the art of pulling habits out of rats. -- O'Neill
Oct 30 '06 #17

P: n/a
VK
But my guess is Google and other search engines will adjust
your rank based on how miserably you fail, or pass validation.
You think that Google bases their rank on W3C validation results as
well? That doesn't seem true given that they disregard perfectly valid
XHTML pages served (as should, actually) with application/xhtml+xml
Content-Type.
Was it your guess or some inside?

Oct 30 '06 #18

P: n/a
>But my guess is Google and other search engines will adjust
>your rank based on how miserably you fail, or pass validation.

You think that Google bases their rank on W3C validation results as
well? That doesn't seem true given that they disregard perfectly valid
XHTML pages served (as should, actually) with application/xhtml+xml
Content-Type.
Was it your guess or some inside?
It's a guess. But it's not hard to imagine that Google runs it's downloaded
pages through various tests, including W3C validation tests. If you were a
decision maker at Google, would you include W3C test results in the algorithm
that determines page rank? I sure would. It may not be as significant as other
factors, but I'll bet it's part of the calculation.
Oct 30 '06 #19

P: n/a
>Avoiding underlined text where it might be confused with a hyperlink is a
>good idea for usability sake, but underlining in the appropriate context,
such as underlining book titles, should not be discarded.

Underlining book titles is a typewriter/handwriting hack to indicate
italics. Book titles belong in CITE and cite should styled according to
context (i.e. italics in a roman context, roman -- that is, normal -- in an
italics context), unless of course for some reason you are attempting to
imitate a typewriter document. Underlining still is the indication of
italics for documents which will be typeset by human beings (i.e. in paper
manuscripts) but it is unlikely that HTML is the ideal way to prepare such
documents.
I just checked the MLA Formatting and Style Guide ... No more underlining?!
hmmm. I think this is a sign of the apocalypse...

Oct 31 '06 #20

P: n/a
VK wrote:

[snip]
To OP: browsers do not care what actually DTD points to. They do not
load DTD neither they validate against it. It means that if you use
tag that is not in the current DTD, the page will not crash and W3C
cops will not come knocking your door. :-)
Quite. Browsers have their own idea of what they can represent, and
anything other than that is subject to arbitrary error correction.
However, it is for this reason that valid markup is a good idea: no
structural changes will need to be imposed by the user agent.
At the same time it is a good idea to keep your document at least
formally valid (minimum for the personal satisfaction).
One does have to wonder why some people are content in producing markup
that is so blatantly poor.
Both HTML Transitional and Strict is the last century crap (1999) so
neither one is more modern than other. The "burden of proof" is
iframe. This element became so commonly used that it's difficult to
imagine that some DTD may not have it - yet HTML Strict has not.
Virtually every time I've seen in-line frames in use, it's been totally
unnecessary. That doesn't discount any possibility of legitimate use, of
course, but it can hardly be a good choice for "proof".

[snip]
But formally a document is not valid until validated, and it cannot
be validated w/o DTD link to validate against.
And that URI may be inferred by the formal public identifier. If the
validating processor has access to a local DTD catalogue, it need only
use the FPI to find that DTD. In this sense, the system identifier is a
backup of sorts: if the FPI is of no help, the cited resource will be
used instead.

[snip]

Mike
Oct 31 '06 #21

P: n/a
deko wrote:

[snip]
But it's not hard to imagine that Google runs it's downloaded pages
through various tests, including W3C validation tests. If you were a
decision maker at Google, would you include W3C test results in the
algorithm that determines page rank?
No. The point of ranking results is to find the most relevant sites.
Popularity is significant here as a site is less likely to be popular if
it has no relevance to its field, though that doesn't provide any guarantee.

The page rank doesn't indicate quality, nor are validation tests
particularly adept at determining quality. One may still write a valid
document that's absolutely abysmal at either a technical or aesthetic level.

[snip]

Mike
Please attribute your quotes.
Oct 31 '06 #22

P: n/a
In our last episode,
<f7******************************@comcast.com>,
the lovely and talented deko
broadcast on comp.infosystems.www.authoring.html:
>>Avoiding underlined text where it might be confused with a hyperlink is a
good idea for usability sake, but underlining in the appropriate context,
such as underlining book titles, should not be discarded.

Underlining book titles is a typewriter/handwriting hack to indicate
italics. Book titles belong in CITE and cite should styled according to
context (i.e. italics in a roman context, roman -- that is, normal -- in an
italics context), unless of course for some reason you are attempting to
imitate a typewriter document. Underlining still is the indication of
italics for documents which will be typeset by human beings (i.e. in paper
manuscripts) but it is unlikely that HTML is the ideal way to prepare such
documents.
I just checked the MLA Formatting and Style Guide ... No more underlining?!
hmmm. I think this is a sign of the apocalypse...
Actually, you can do many things toward MLA format with CSS. For
example, you can double-space using the line-height property. This, of
course, only goes to illustrate that manuscript style is very different
from publication style. The desktop publishing abilities of much word
processing software has rather blurred the distinctions in the minds of many
idio-- users. In fact it is rather difficult with some word processors to
set up formatting suitable for manuscripts, because they are geared to
average ...er... user, who wants things to look just like they would in
books and magazines, not the way the need to look to get published in real
books and magazines. How things should look to get published is what MLA
is all about (although only for some academic manuscripts). The authority
most often cited for how things should look in publications (in the US)
is the Chicago Manual of Style.
--
Lars Eighner <http://larseighner.com/ <http://myspace.com/larseighner>
Someday you will look back on this moment and plow into a parked car.
Oct 31 '06 #23

P: n/a
>But it's not hard to imagine that Google runs it's downloaded pages
>through various tests, including W3C validation tests. If you were a
decision maker at Google, would you include W3C test results in the
algorithm that determines page rank?

No. The point of ranking results is to find the most relevant sites.
Popularity is significant here as a site is less likely to be popular if it
has no relevance to its field, though that doesn't provide any guarantee.

The page rank doesn't indicate quality, nor are validation tests particularly
adept at determining quality. One may still write a valid document that's
absolutely abysmal at either a technical or aesthetic level.
That's a reasonable argument, but I think Google is more sophisticated than
that. It's like saying college admissions should be based purely on merit, yet
we all know there is social engineering (however misguided) involved.

If there are two pages that rate the same in other search criteria, why not
break the tie with a W3C validation test (or some derivative thereof)? If a
page author constructs a well-coded page, some credit is due, all other things
being equal.

As I said, this is only a guess. Discovering the exact algorithm Google uses is
on my list of things to do - just after finding the formula for Coca Cola.

Oct 31 '06 #24

P: n/a
On Mon, 30 Oct 2006 13:01:43 -0800, "deko" <de**@nospam.comwrote:
>Avoiding underlined text where it might be confused with a hyperlink is a good
idea for usability sake, but underlining in the appropriate context, such as
underlining book titles, should not be discarded.
One of the sites I work on is a Library catalogue and all the titles
*are* hyperlinks. I don't see any other reason to underline book
titles, so I think you've picked a bad example.

Oct 31 '06 #25

P: n/a
>>>Avoiding underlined text where it might be confused with a hyperlink is a
>>>good idea for usability sake, but underlining in the appropriate context,
such as underlining book titles, should not be discarded.

Underlining book titles is a typewriter/handwriting hack to indicate
italics. Book titles belong in CITE and cite should styled according to
context (i.e. italics in a roman context, roman -- that is, normal -- in an
italics context), unless of course for some reason you are attempting to
imitate a typewriter document. Underlining still is the indication of
italics for documents which will be typeset by human beings (i.e. in paper
manuscripts) but it is unlikely that HTML is the ideal way to prepare such
documents.
>I just checked the MLA Formatting and Style Guide ... No more underlining?!
hmmm. I think this is a sign of the apocalypse...

Actually, you can do many things toward MLA format with CSS. For
example, you can double-space using the line-height property. This, of
course, only goes to illustrate that manuscript style is very different
from publication style. The desktop publishing abilities of much word
processing software has rather blurred the distinctions in the minds of many
idio-- users. In fact it is rather difficult with some word processors to
set up formatting suitable for manuscripts, because they are geared to
average ...er... user, who wants things to look just like they would in
books and magazines, not the way the need to look to get published in real
books and magazines. How things should look to get published is what MLA
is all about (although only for some academic manuscripts). The authority
most often cited for how things should look in publications (in the US)
is the Chicago Manual of Style.
I'm fairly conversant with CSS and have used line-height on a number of
occasions. I remember having trouble with vertical-align:super when trying to
create a reference to a footnote, but have not used it lately.

MLA was the authority I used in both college and grad school, so I figure that's
the gold standard. As I remember, the Chicago Manual of Style was an
alternative source of truth, so to speak, but MLA was more common, at least in
academia.

This idea of using italics instead of underlining book citations is something
new. I asked a friend, who is a high school teacher, and she said the shift
took place only a few years ago.

Oct 31 '06 #26

P: n/a
>>Avoiding underlined text where it might be confused with a hyperlink is a good
>>idea for usability sake, but underlining in the appropriate context, such as
underlining book titles, should not be discarded.

One of the sites I work on is a Library catalogue and all the titles
*are* hyperlinks. I don't see any other reason to underline book
titles, so I think you've picked a bad example.
It looks like the poor underline is being displaced, relegated to the dustbin of
typography.

Oct 31 '06 #27

P: n/a
deko wrote :
A web page can be created using either strict or transitional HTML.

The DOCTYPE for strict is:

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
"http://www.w3.org/TR/html4/strict.dtd">

And transitional:

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"
"http://www.w3.org/TR/html4/loose.dtd">

According to this document (written in 1999):
http://www.w3.org/TR/html401/sgml/loosedtd.html

"the HTML 4.01 Transitional DTD ... includes presentation attributes and
elements that W3C expects to phase out as support for style sheets
matures."

My questions are:

Can someone please summarize reasons why transitional HTML would be
used over strict?
I think there are none actually: even <appletcan now be replaced with
<object>. There are nowadays lots of excellent tools to upgrade old
documents or create new ones: transitional HTML really makes no sense
nowadays.

Of the elements which the W3C plans to phase out, are any still in
common use?

<font>, <u>, <s>, <applet>, <basefont>, <center>, <dir>, <isindex>,
<menu>, <strike>, align, archive, background, bgcolor, nowrap, target,
etc... They all can be replaced by other elements or can be conveniently
replaced by CSS properties.

In my opinion, still often seen are <font>, <center>, bgcolor but don't
misunderstand me: they all can be replaced with CSS without a problem.

Gérard
--
remove blah to email me
Oct 31 '06 #28

P: n/a
deko wrote:
A web page can be created using either strict or transitional HTML.

The DOCTYPE for strict is:

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
"http://www.w3.org/TR/html4/strict.dtd">

And transitional:

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"
"http://www.w3.org/TR/html4/loose.dtd">

According to this document (written in 1999):
http://www.w3.org/TR/html401/sgml/loosedtd.html

"the HTML 4.01 Transitional DTD ... includes presentation attributes and
elements that W3C expects to phase out as support for style sheets
matures."

My questions are:

Can someone please summarize reasons why transitional HTML would be
used over strict?

Of the elements which the W3C plans to phase out, are any still in
common use?
Thanks in advance.
As I understand, Strict does not allow the use of deprecated tags and
attributes while Transitional allows them.

I hand-code my HTML. I specify Transitional in my <!DOCTYPEbecause
it's easier to code with some of the deprecated items. For example:

I use <fontwhere I want to specify a CSS class that is entirely
font-based. If I later have to edit that portion of a page, this
provides a mnemonic aid to me so that I recognize the class as only
affecting fonts.

I use the "width" attribute on <hrbecause I have many different widths
of the horizontal rule. I don't want to have many different classes,
one for each width.

David E. Ross
<http://www.rossde.com/>

Concerned about someone (e.g., Pres. Bush) snooping
into your E-mail? Use PGP.
See my <http://www.rossde.com/PGP/>
Oct 31 '06 #29

P: n/a
deko wrote:
>>>>Avoiding underlined text where it might be confused with a
hyperlink is a
good idea for usability sake, but underlining in the appropriate
context,
such as underlining book titles, should not be discarded.

Underlining book titles is a typewriter/handwriting hack to indicate
italics. Book titles belong in CITE and cite should styled
according to
context (i.e. italics in a roman context, roman -- that is, normal
-- in an
italics context), unless of course for some reason you are
attempting to
imitate a typewriter document. Underlining still is the indication of
italics for documents which will be typeset by human beings (i.e. in
paper
manuscripts) but it is unlikely that HTML is the ideal way to
prepare such
documents.
>>I just checked the MLA Formatting and Style Guide ... No more
underlining?!
hmmm. I think this is a sign of the apocalypse...

Actually, you can do many things toward MLA format with CSS. For
example, you can double-space using the line-height property. This, of
course, only goes to illustrate that manuscript style is very different
from publication style. The desktop publishing abilities of much word
processing software has rather blurred the distinctions in the minds
of many
idio-- users. In fact it is rather difficult with some word
processors to
set up formatting suitable for manuscripts, because they are geared to
average ...er... user, who wants things to look just like they would in
books and magazines, not the way the need to look to get published in
real
books and magazines. How things should look to get published is what MLA
is all about (although only for some academic manuscripts). The
authority
most often cited for how things should look in publications (in the US)
is the Chicago Manual of Style.

I'm fairly conversant with CSS and have used line-height on a number of
occasions. I remember having trouble with vertical-align:super when
trying to create a reference to a footnote, but have not used it lately.

MLA was the authority I used in both college and grad school, so I
figure that's the gold standard. As I remember, the Chicago Manual of
Style was an alternative source of truth, so to speak, but MLA was more
common, at least in academia.

This idea of using italics instead of underlining book citations is
something new. I asked a friend, who is a high school teacher, and she
said the shift took place only a few years ago.
Actually, book titles were italicized in other books when I was in high
school 50 years ago. We only underlined (as indicated earlier) when
using a typewriter. As soon as I started using word processing on a
computer (Interleaf on a UNIX host about 12 years ago), I was
italicizing book titles.

--

David E. Ross
<http://www.rossde.com/>

Concerned about someone (e.g., Pres. Bush) snooping
into your E-mail? Use PGP.
See my <http://www.rossde.com/PGP/>
Oct 31 '06 #30

P: n/a
VK
As I said, this is only a guess. Discovering the exact algorithm Google uses is
on my list of things to do - just after finding the formula for Coca Cola.
As yourself then, I can talk only for myself as my last strike force
team got nuked before anyone reached the Flag :-)
>From few survivers of previously failed operations :-) I may tell that
Google -also- analizes your source of information as it would be
delivered to an average HTML browser. This way (anything else equal) a
page with single <htmland single <bodymay get advantage over a page
with say two heads and three bodies. At the same time even the latter
case takes an advantage over a perfectly valid XHTML page delivered
from the server with application/xhtml+xml Content-Type which is simply
disregarded. This way I see some accessability ranking in action but no
standard-compliance of any kind.

Oct 31 '06 #31

P: n/a
VK

VK wrote:
Even I did see the catch right away.
Even I did *not* see the catch right away.

A sense-changing typo, sorry - just noticed.

Oct 31 '06 #32

P: n/a
On 2006-10-31, deko wrote:
>>>Avoiding underlined text where it might be confused with a hyperlink is a good
idea for usability sake, but underlining in the appropriate context, such as
underlining book titles, should not be discarded.

One of the sites I work on is a Library catalogue and all the titles
*are* hyperlinks. I don't see any other reason to underline book
titles, so I think you've picked a bad example.

It looks like the poor underline is being displaced, relegated to the dustbin of
typography.
The underline never had anything to do with typography. It was a
workaround for the limitations of the typewriter. Now that very few
people use a typewriter, its use is unnecessary.
--
Chris F.A. Johnson <http://cfaj.freeshell.org>
================================================== =================
Author:
Shell Scripting Recipes: A Problem-Solution Approach (2005, Apress)
Nov 1 '06 #33

P: n/a
On Mon, 30 Oct 2006, Ed Seedhouse wrote:
Best alternative is not to use underlines except on a hyperlink.
Still, <uis the best and most compatible way to write "_kh_"
as in http://www.unics.uni-hannover.de/nht...-alphabet.html
to distinguish it from "kh" without underline.
http://www.loc.gov/catdir/cpso/romanization/urdu.pdf

Nov 2 '06 #34

P: n/a
In message <0a************@xword.teksavvy.com>, Tue, 31 Oct 2006
22:18:56, Chris F.A. Johnson <cf********@gmail.comwrites
>
The underline never had anything to do with typography. It was a
workaround for the limitations of the typewriter. Now that very few
people use a typewriter, its use is unnecessary.
That's anti-elitist.

Underline, and many other forms of marking surrounding a normal letter
or other character(s), are commonly used in science and mathematics.

One should not limit consideration to the common or garden needs of
administrative business.

--
(c) John Stockton, Surrey, UK. ?@merlyn.demon.co.uk Turnpike v6.05 IE 6
<URL:http://www.jibbering.com/faq/>? JL/RC: FAQ of news:comp.lang.javascript
<URL:http://www.merlyn.demon.co.uk/js-index.htmjscr maths, dates, sources.
<URL:http://www.merlyn.demon.co.uk/TP/BP/Delphi/jscr/&c, FAQ items, links.
Nov 2 '06 #35

P: n/a
In our last episode,
<Lz**************@invalid.uk.co.demon.merlyn.inval id>,
the lovely and talented Dr J R Stockton
broadcast on comp.infosystems.www.authoring.html:
In message <0a************@xword.teksavvy.com>, Tue, 31 Oct 2006
22:18:56, Chris F.A. Johnson <cf********@gmail.comwrites
>>
The underline never had anything to do with typography. It was a
workaround for the limitations of the typewriter. Now that very few
people use a typewriter, its use is unnecessary.
That's anti-elitist.
Underline, and many other forms of marking surrounding a normal letter
or other character(s), are commonly used in science and mathematics.
Underlining is blackboard italics. In print, underlining is replaced
with italics (just as the blackboard wavy underlining is replaced with
bold). HTML is very poor suited for mathematics -- no one is claiming
otherwise. Even so, HTML could be used to markup much more mathematics,
but it pointless to do so when there are not browsers that could render
the markup appropriately.
One should not limit consideration to the common or garden needs of
administrative business.
--
Lars Eighner <http://larseighner.com/ <http://myspace.com/larseighner>
My last cow just died, so I won't need your bull anymore.
Nov 2 '06 #36

P: n/a
On 2 Nov 2006, Lars Eighner wrote:
Underlining is blackboard italics. In print, underlining is replaced
with italics (just as the blackboard wavy underlining is replaced with
bold).
Underlining is not the same as italics!
Sometimes underlining in manuscripts *may* denote italics.
But not every underlining is really italic text.

I have already mentioned
http://www.loc.gov/catdir/cpso/romanization/urdu.pdf
http://www.unics.uni-hannover.de/nht...-alphabet.html
with underlined letters.

Specialized mathematical and physical notation even uses underlining
*in addition* to italics. For example,

<i>Z</i>

may denote (real) impedance and

<u><i>Z</i></u>

may denote complex impedance.

Nov 3 '06 #37

P: n/a
deko schrieb:
[...]
My questions are:

Can someone please summarize reasons why transitional HTML would be
used over strict?
I can see one reason: Your client/boss/designer asks you to realize a
layout that can't be done without tables or frames. You have already
spent some hours explaining him/her the disadvantages of this layout and
presenting alternatives, but without success. For whatever reason you
don't want, or you are not in the position to resign from the job.

--
Markus
Nov 4 '06 #38

P: n/a
>Can someone please summarize reasons why transitional HTML would be used over
>strict?

I can see one reason: Your client/boss/designer asks you to realize a layout
that can't be done without tables or frames. You have already spent some hours
explaining him/her the disadvantages of this layout and presenting
alternatives, but without success. For whatever reason you don't want, or you
are not in the position to resign from the job.
I don't use frames (and probably never will) but what is it about tables that
would require Transitional HTML? I have all kinds of tables used all over the
place. Why would a design layout containing tables require Transitional HTML?

Nov 4 '06 #39

P: n/a
deko wrote:
>>Can someone please summarize reasons why transitional HTML would be
used over strict?

[Someone else penned this:]
I can see one reason: Your client/boss/designer asks you to realize a
layout that can't be done without tables or frames. You have already
spent some hours explaining him/her the disadvantages of this layout
and presenting alternatives, but without success. For whatever
reason you don't want, or you are not in the position to resign from
the job.

I don't use frames (and probably never will) but what is it about
tables that would require Transitional HTML? I have all kinds of
tables used all over the place. Why would a design layout containing
tables require Transitional HTML?
I think Markus was simply noting that one should eschew tables for
layout, per the 21st century mantra. Tables are certainly allowed with
Strict, and Strict is what should be used for all new documents.
Transitional is for converting/working on legacy documents.

If one were to use Strict, would the boss ever really know? <g>

Please don't snip attributes from your posts. Thanks.

--
-bts
-Motorcycles defy gravity; cars just suck
Nov 4 '06 #40

P: n/a
deko schrieb:
>>Can someone please summarize reasons why transitional HTML would be
used over strict?

I can see one reason: Your client/boss/designer asks you to realize a
layout that can't be done without tables or frames. You have already
spent some hours explaining him/her the disadvantages of this layout
and presenting alternatives, but without success. For whatever reason
you don't want, or you are not in the position to resign from the job.

I don't use frames (and probably never will) but what is it about tables
that would require Transitional HTML? I have all kinds of tables used
all over the place. Why would a design layout containing tables require
Transitional HTML?
You are right, in the case that was in my mind it was actually not the
table that required the transitional doctype. The layout required
horizontal centering of the main contents area, and also of some blocks
inside other elements. I was only able to achieve this using a layout
table, and also only with a transitional doctype, as strict doctype does
not seem to link table height of 100% to the viewport height. So I
should rather have written "You are asked to realize a layout that can't
be realized with a strict doctype".

Anyway maybe my posting was not extremely relevant - I just read the
original question after having spent many hours trying to do this kind
of thing the clean way, and ended up doing it as I could have done it
years ago...

--
Markus
Nov 6 '06 #41

P: n/a
On Mon, 30 Oct 2006 08:30:34 -0800, "deko" <de**@nospam.comwrote:
>My questions are:

Can someone please summarize reasons why transitional HTML would be used
over strict?
When ones HTML experiences were learned in the mid 90s. I can deal
with making a page that will validate using HTML 4.0 transitional, but
that's where I "dropped out" of HTML school. I work with what I have,
and don't do enough web design to warrant learning more, at this point
in time. But, CSS is next on my list of things to learn. Maybe next
year. :)

--
Zilbandy - Tucson, Arizona USA <zi*@zilbandyREMOVETHIS.com>
Dead Suburban's Home Page: http://zilbandy.com/suburb/
PGP Public Key: http://zilbandy.com/pgpkey.htm
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~
Nov 6 '06 #42

This discussion thread is closed

Replies have been disabled for this discussion.