IBM doctype?

T

Toby A Inkster

Sean said:
<!DOCTYPE html SYSTEM "http://www.ibm.com/data/dtd/v11/ibmxhtml1-
transitional.dtd">

Are they trying to make their own web standards or something? Will browsers
actually read this DTD?
Click to expand...

Well, there is nothing to stop anyone making up their own doctypes.

e.g.
http://www.foad.org/~abigail/abigail.dtd
http://www.w3.org/Style/HTML40-plus-blink.dtd

You will notice that these DTDs tend to restrict themselves only to what
browsers can already do.

For instance, they don't try to introduce some element '<POTATOPICTURE>'
to insert a picture of a potato, because they know that no browsers will
recognise it.

On the other hand, they might add the '<BLINK>' element, because browsers
tend to already understand that.

Why do this?

Well, if you use '<BLINK>Important</BLINK>' in an HTML 4.0 document, you
will soon find yourself with an invalid document, because HTML 4.0 doesn't
have any element '<BLINK>'. However, if you change your DOCTYPE to the
W3C's HTML40-plus-blink.dtd, then '<BLINK>' is allowed, but your page is
still valid. (It's not valid HTML 4.0, but it's valid something!)
 
L

Leif K-Brooks

Toby said:
Well, if you use '<BLINK>Important</BLINK>' in an HTML 4.0 document, you
will soon find yourself with an invalid document, because HTML 4.0 doesn't
have any element '<BLINK>'. However, if you change your DOCTYPE to the
W3C's HTML40-plus-blink.dtd, then '<BLINK>' is allowed, but your page is
still valid. (It's not valid HTML 4.0, but it's valid something!)
Click to expand...

I've never understood that. Why is being valid so important if you
aren't validating against real standards?
 
J

Jukka K. Korpela

Leif K-Brooks said:
Why is being valid so important if you
aren't validating against real standards?
Click to expand...

Being valid, in the SGML sense, means correspondence between the actual
markup and a formalized description of markup syntax. This was relevant
years before HTML, or the Web, was invented. In fact, it is far less
useful for HTML documents than one might expect, since hardly any browser
actually implements HTML as an SGML application.

By the way, there is only one real standard (an ISO standard) for HTML,
and (virtually) nobody uses it.
 
J

Joel Shepherd

Leif said:
I've never understood that. Why is being valid so important if you
aren't validating against real standards?
Click to expand...

Maybe you're validating to something more strict than the usual DTD
(e.g., maybe your DTD requires closing tags that are optional in the
"standard" DTD). Or maybe you've decided for your purposes that use of
a deprecated or non-standard element is acceptable, and you want to
silence that particular warning.

The latter is no better or worse than, say, ignoring a C-compiler
warning, or writing using sloppier-than-schoolbook grammar. While in
general neither one is an especially good idea, if you know what
you're doing there are situations where it's acceptable.

If you don't know what you're doing, then you've got bigger problems
than the validator or compiler can fix, though your grammar teacher
might give it a shot... ;-)
 
T

Toby A Inkster

Jukka said:
By the way, there is only one real standard (an ISO standard) for HTML,
and (virtually) nobody uses it.
Click to expand...

Some things have been standardised by ISO, some by ECMA, some by ANSI. So
what? Just because something has "ISO" stamped on it, doesn't make it any
more standard than a W3C standard.
 
D

Dylan Parry

Toby said:
Just because something has "ISO" stamped on it, doesn't make it any
more standard than a W3C standard.
Click to expand...

The W3C does not publish standards, it publishes "recommendations".
 
L

Leif K-Brooks

Jukka said:
The word "real" was relevant here. As long as you use the word "standard"
informally, you can use it rather sloppily and call everyone's and his
dog's specification, or even a mere industry practice, a "standard".
But when you slap the word "real" in front of it, it is fair to require
that you be able to specify the ISO, IEC, or ITU number, or at least the
number of a national standard as issued by a national authorized standards
body, such as ANSI. (In the Internet context, you might get away with it
if you can specify the STD number assigned by the IESG, but those few
"Internet standards" are not really real standards - not issued by an
international standards body authorized by national members that have as
official a status in their countries as any body can, depending on
national situation.)
Click to expand...

And what makes a country a country? I can call my house a country if I
want to. All that makes a country real is force. Do we really need guns
and bombs to create web standards now?
 
M

Matthias Gutfeldt

Jukka K. Korpela said:
Surely. And you can call a piece of bread an HTML program,
Click to expand...

Jukka, I would like a slice of whole wheat HTML bread with my tag soup,
please :).


Matthias
 
N

Nick Theodorakis

Jukka, I would like a slice of whole wheat HTML bread with my tag soup,
please :).
Click to expand...

I've always thought that "tag soup" was a German-English pidgin word
for "soup of the day." ;-)

Nick
 
L

Leif K-Brooks

Jukka said:
Surely. And you can call a piece of bread an HTML program, or pointless
play with words an argument. I would appreciate if you did not do that
here, and, as the second best option, if you continued using a forged From
field as long as you just babble. TIA.
Click to expand...

Then explain why Canada or Britian is more of a country than what I
decide to call a country. The only real answer is the number of people
who recognize it as such, which makes sense.

By the same logic, a web standard is determined by how many people
recognize it as such. Even if no country or international body
recognizes W3C standards, they are loosely used by 99.9% of web authors
and browsers.
 
E

Eric B. Bednarz

Leif K-Brooks said:
By the same logic, a web standard is determined by how many people
recognize it as such.
Click to expand...

By the same logic, that's how M$ Internet Exploder became the only
relevant standard on the web, and the rest obscurity for its own sake.
Even if no country or international body
recognizes W3C standards, they are loosely used by 99.9% of web
authors and browsers.
Click to expand...

You have a serious typo there.

s/used by/unknown or irrelevant to/

Short of a couple of weblogs, nobody takes W3C prose serious, apparently
not even its members. Maybe because of the tradition of keeping the
nose close to the ground, following the trail of the big Redmond dog,
and wagging with the tail when they encounter its warm and steamy
droppings.
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Similar Threads

What doctype should I use ? 26
CSS Newbie - CSS Works With Invalid DOCTYPE. Fails With Valid DOCTYPE. 19
newbie, trouble with doctype update 7
DOCTYPE related question 7
DOCTYPE html PUBLIC and document.onmousemove 3
Help with Visual Lightbox: Scripts 2
I need help making an html website 2
Can't make table the size of the screen 12

Members online

No members online now.
Total: 26 (members: 0, guests: 26)
Robots: 240

Forum statistics

Threads
473,756
Messages
2,569,535
Members
45,008
Latest member
obedient dusk

Latest Threads

Top