Why this html doesn't work?

**Jukka K. Korpela** · Jun 27 '08, 07:16 PM

Re: Why this html doesn't work?

Scripsit ray:

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"

Post a URL, not a code extract.

Don't mix HTML and XHTML. Don't use XHTML on the web until you
understand what XHTML actually is. (And when you do understand it, you
don't need to be told not to use XHTML on the web.)

Specifically, do you know what the above tag means in (non-XHTML) HTML
and how it is processed by browsers treating your document as HTML?

<SCRIPT type="text/javascript">
function onload(){
alert("aaa");
}
</SCRIPT>

The content of the element defines a function in JavaScript. Do you
expect the function to be called (invoked) when the page is loaded? Why?

(You did not actually cut and paste the _real_ page, did you? This is
one of the reasons why you should post a URL.)

If I remove the line that includes the test.js file, the html file
works, otherwise the alert dialog doesn't appear even if the test.js
is empty.

If "works" means that the alert("aaa") invocation is executed, I don't
believe you.

--
Jukka K. Korpela ("Yucca")

http://www.cs.tut.fi/~jkorpela/

**Gus Richter** · Jun 27 '08, 07:16 PM

Re: Why this html doesn't work?

ray wrote:

Your document is an HTML document per the doctype declaration.
The META element is an empty element without a closing tag.
In XML, empty elements require a (fwd slash) terminator.
In XHTML, open elements require (space + fwd slash) terminator.
In HTML the META element should look so:
<META http-equiv="Content-Type" content="text/html; charset=UTF-8">

The SCRIPT element requires a closing tag in both HTML and XHTML,
whether empty or not.
In XML (I believe, since not into XML), an empty element has the closing
tag dropped and the opening tag has a (fwd slash) terminator added.
In HTML the SCRIPT element should look so:
<SCRIPT type="text/javascript" src="test.js"></SCRIPT>

If I remove the line that includes the test.js file, the html file
works, otherwise the alert dialog doesn't appear even if the test.js
is empty.

With the closing tag it works.

--
Gus

**Roy A.** · Jun 27 '08, 07:16 PM

Re: Why this html doesn't work?

On 26 Mai, 04:53, ray <hanl...@gmail. comwrote:

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
"http://www.w3.org/TR/html4/strict.dtd">
>
<HTML>
<HEAD>
<META http-equiv="Content-Type" content="text/html; charset=UTF-8"/
>
<TITLE>HTML to Text Test Page</TITLE>
<SCRIPT type="text/javascript" src="test.js"/>

You should use a proper document type.

It's considered best practise to read htmldog.com.

<SCRIPT type="text/javascript">
function onload(){
alert("aaa");
}
</SCRIPT>
</HEAD>
<BODY>
<DIV id="case1">
<P>abc</P>
<P>def</P>
</DIV>
</BODY>
</HTML>
The test.js is located in the same directory as that of the html file.
The alert dialog doesn't appear.
If I remove the line that includes the test.js file, the html file
works, otherwise the alert dialog doesn't appear even if the test.js
is empty.
Thanks in advance.

**Eric B. Bednarz** · Jun 27 '08, 07:16 PM

Re: Why this html doesn't work?

"Roy A." <royarneskar@gm ail.comwrites:

On 26 Mai, 04:53, ray <hanl...@gmail. comwrote:

>Â Â <META http-equiv="Content-Type" content="text/html; charset=UTF-8"/
>>

You should use a proper document type.

Like PDF? :)

It's considered best practise to read htmldog.com.

Is that in the FAQ?

(I think it is generally considered good pratice to use a nntp client
for posting on Usenet, preferably one that does not mangle quoted code
snippets, however inadequate they may be, but that is an entirely
different problem)

--
"Note that the DOCTYPE tag is a bit of a rebel and demands to be written
in upper case and adorned with an exclamation mark. It also breaks the
rules in that it is the only tag that doesn't need closing."
<http://htmldog.com/guides/htmladvanced/declarations/>
[good stuff, no emphasis added â€”Ed.]

**Jukka K. Korpela** · Jun 27 '08, 07:16 PM

Group FAQ & Bogosities in htmldog.com

Sub titulo Re: Why this html doesn't work?
scripsit Eric B. Bednarz:

>It's considered best practise to read htmldog.com.

>
Is that in the FAQ?

Why do you ask? Because you are uncertain about the existence of a FAQ
for this group? According to faqs.org, for whatever it's worth, the
group has a FAQ at http://www.faqs.org/faqs/www/authoring-faq/ and the
FAQ is posted monthly, last modified in 2001, last posted in 2002; ref.:

http://www.faqs.org/faqs/by-newsgroup/comp/comp.infosystems.www.authoring.html.html

But the FAQ there refers to

Web Authoring FAQ (Web Design Group)

http://htmlhelp.com/faq/html/

The Web Design Group's Web Authoring FAQ addresses frequently asked questions related to HTML, images, style sheets, and other Web authoring issues.

which was last updated on April 26, 2007.

Generally, let's all sing "Oh where have all the FAQs gone", and let us
be thankful to people who contributed to the FAQ and especially to Darin
McGrew, who is apparently the last person keeping the FAQ somewhat
alive.

(I think it is generally considered good pratice to use a nntp client
for posting on Usenet, preferably one that does not mangle quoted code
snippets, however inadequate they may be, but that is an entirely
different problem)

Right. But I think it's not good practice to hide the most important
content of a message into a sig. You sig says:

"Note that the DOCTYPE tag is a bit of a rebel and demands to be written
in upper case and adorned with an exclamation mark. It also breaks the
rules in that it is the only tag that doesn't need closing."
<http://htmldog.com/guides/htmladvanced/declarations/>
[good stuff, no emphasis added â€”Ed.]

Technically, it exceeds the four-line limit. More importantly, it is
apparently ironic, but not apparently enough, since most HTML authors,
even including those who read the group, are more or less clueless about
doctype declarations. Sorry if this insults someone, but it's still
true. I am clueless about most things in world, and I don't mind if
people point that out. (Strangely enough, though, I'm usually accused
for being clueless in matters that I actually know pretty well. Strange
world, isn't it?)

Anyway, let's be explicit about the goofs in the quoted text:
1) The doctype declaration is not a tag at all (so the last sentence in
the quoted text is nonsense).
2) It is not case-sensitive, though some parts of it are.
3) The "funny" wordings aren't funny, and an explanation like that
wouldn't be a right place to be funny.

Otherwise, the page from which the quotation was taken tells what the
XHTML doctype declarations are and babbles about them. This is
counter-productive. It also says:

"Without specifying a doctype, your HTML just isn't valid HTML and most
browsers viewing them will switch to 'quirks mode', which means they
will think that you don't know what the hell you're doing and make up
their own mind on what to do with your code."

The text makes a point and then quickly loses it, describing "quirks
mode" in wrong terms and without citing any information on what "quirks
mode" actually means. It does not even warn that adding a doctype
declaration to an existing page may completely ruin the page (for
example, all content turns invisible), unless the page is fixed to use
proper HTML and proper CSS, which is rather nontrivial task.

I'm not saying that the author is "a short-sighted, one-eyed infant
gibbon with learning difficulties", as he describes a hypothetical web
author, but that's mainly due to my respect for people with learning
difficulties and for our simian relatives.

I had a positive impression of htmldog.com, from other people's comments
and from a quick look at the surface. This surely opened my eyes.

--
Jukka K. Korpela ("Yucca")

http://www.cs.tut.fi/~jkorpela/

**Jonathan N. Little** · Jun 27 '08, 07:16 PM

Re: Group FAQ & Bogosities in htmldog.com

Jukka K. Korpela wrote:

Sub titulo Re: Why this html doesn't work?
scripsit Eric B. Bednarz:
>

>>It's considered best practise to read htmldog.com.

< snip>

Otherwise, the page from which the quotation was taken tells what the
XHTML doctype declarations are and babbles about them. This is
counter-productive. It also says:
>
"Without specifying a doctype, your HTML just isn't valid HTML and most
browsers viewing them will switch to 'quirks mode', which means they
will think that you don't know what the hell you're doing and make up
their own mind on what to do with your code."
>
The text makes a point and then quickly loses it, describing "quirks
mode" in wrong terms and without citing any information on what "quirks
mode" actually means. It does not even warn that adding a doctype
declaration to an existing page may completely ruin the page (for
example, all content turns invisible), unless the page is fixed to use
proper HTML and proper CSS, which is rather nontrivial task.
>
I'm not saying that the author is "a short-sighted, one-eyed infant
gibbon with learning difficulties", as he describes a hypothetical web
author, but that's mainly due to my respect for people with learning
difficulties and for our simian relatives.
>
I had a positive impression of htmldog.com, from other people's comments
and from a quick look at the surface. This surely opened my eyes.
>

There is little out there either online or in print that gives newbies
the proper modern approach. It is disappointing section on their site.
However, other parts are best that I have seen online. Guess we cannot
have perfection. Would welcome the discovery of anything better...but
until that I would say that www.htmldog.com tutorials are best(*
although not perfect) to be found, and far better then most.

--
Take care,

Jonathan
-------------------
LITTLE WORKS STUDIO

Little Works Studio Site Maintainance

http://www.LittleWorksStudio.com

**Eric B. Bednarz** · Jun 27 '08, 07:16 PM

Re: Group FAQ & Bogosities in htmldog.com

"Jukka K. Korpela" <jkorpela@cs.tu t.fiwrites:

scripsit Eric B. Bednarz:
>

>>It's considered best practise to read htmldog.com.

>>
>Is that in the FAQ?

>
Why do you ask? Because you are uncertain about the existence of a FAQ
for this group?

It was more of a joke (these days you never know, though :).

Right. But I think it's not good practice to hide the most important
content of a message into a sig.

I wouldnâ€™t say itâ€™s the most important content. Mind me, if itâ€™s not
just plain old good but best practice, Iâ€™d really like to learn more
about the whereabouts of those who appear to consider that. You may
consider the question ironic too, if you must. :) Frankly, I find
htmldog a pretty lousy resource, but didnâ€™t feel pressed to say that in
a straightforward manner.

Besides the excellent advice to exclusively declare language on the web
with the xml:lang MacGuffin, did you notice the PHP BS at the end of
that page, by the way?

| <?
| if(stristr(\$_S ERVER["HTTP_ACCEP T"],"applicatio n/xhtml+xml")){
| header("Content-Type: application/xhtml+xml; charset=UTF-8");
<http://htmldog.com/guides/htmladvanced/declarations/>

Thereâ€™s so much wrong with it that you could build an entire Web 2.0
conference around it. As to the conclusion

| Other than peace of mind that you know you're doing the right thing
| and preparing yourself for the way to do things in the future, the
| immediate benefit of using this method is that Mozilla will treat your
| files as XML applications [â€¦]

Iâ€™m glad to say that this *is* in the FAQ.

<http://www.mozilla.org/docs/web-developer/faq.html#accept >

(Strangely enough, though, I'm usually
accused for being clueless in matters that I actually know pretty
well. Strange world, isn't it?)

Dunno. Actually, one of the reasons why I despise all the web standards
yada yada is that the most vocal proponents typically didnâ€™t read any of
that boring stuff that they are evangelizing. It is pretty easy to get
into hopeless arguments with them, especially if you did your
homework. As I see it, that leads to either frustration or amusement. I
learned to prefer the latter; it was tough, but it worked out :)

--
||| hexadecimal EBB
o-o decimal 3771
--oOo--( )--oOo-- octal 7273
205 goodbye binary 111010111011

**Andy Dingley** · Jun 27 '08, 07:16 PM

Re: Group FAQ & Bogosities in htmldog.com

On 29 May, 11:27, "Jukka K. Korpela" <jkorp...@cs.tu t.fiwrote:

I am clueless about most things in world, and I don't mind if
people point that out. (Strangely enough, though, I'm usually accused
for being clueless in matters that I actually know pretty well.

So that makes you smarter than Plato?

Few dispute your cluefulness in the areas you discuss. If you're
accused of cluelessness, might I suggest that it's your
communication skills that are falling short, not your understanding.

**Jukka K. Korpela** · Jun 27 '08, 07:16 PM

Re: Group FAQ & Bogosities in htmldog.com

Scripsit Andy Dingley:

might I suggest that it's your
communication skills that are falling short, not your understanding.

You might, but then you would express something about _your_ manners and
skills in communication. Like I would if I started babbling my guesses
about your personality.

--
Jukka K. Korpela ("Yucca")

http://www.cs.tut.fi/~jkorpela/

**dorayme** · Jun 27 '08, 07:16 PM

Re: Group FAQ & Bogosities in htmldog.com

In article
<6bd907d2-3632-41f0-8ee8-eed86ec07a12@j2 2g2000hsf.googl egroups.com>,
Andy Dingley <dingbat@codesm iths.comwrote:

On 29 May, 11:27, "Jukka K. Korpela" <jkorp...@cs.tu t.fiwrote:
>

I am clueless about most things in world...

>
>
... I suggest that it's your
communication skills that are falling short, not your understanding.

The two things, in certain defined areas - namely understanding the
ordinary "other" human mind - are connected so tightly that one cannot
so easily separate them. So, in this respect, but not in any other, it
suggests a misplaced flattery.

--
dorayme

**dorayme** · Jun 27 '08, 07:16 PM

Re: Group FAQ & Bogosities in htmldog.com

In article <YoX%j.9613$_03 .3478@reader1.n ews.saunalahti. fi>,
"Jukka K. Korpela" <jkorpela@cs.tu t.fiwrote:

Scripsit Andy Dingley:
>

might I suggest that it's your
communication skills that are falling short, not your understanding.

>
You might, but then you would express something about _your_ manners and
skills in communication. Like I would if I started babbling my guesses
about your personality.

So much for the good natured aspect of AD's remark. Not only what is it
not seen to be flattering - the only respect, paradoxically, in which it
was false - but it simply fell on deaf ears and so enhanced its
credibility.

--
dorayme

**Jukka K. Korpela** · Jun 27 '08, 07:16 PM

Re: Group FAQ & Bogosities in htmldog.com

Scripsit Eric B. Bednarz:

Besides the excellent advice to exclusively declare language on the
web with the xml:lang MacGuffin, did you notice the PHP BS at the end
of that page, by the way?

No, I didn't read that carefully. I usually read text carefully only if
it has been written carefully, unless I am specifically paid for reading
it.

Thanks for the amusement.

But I'm afraid people's irony detectors might not be working, so they
might take your praise at face value. After all, they keep hearing that
XML and XHTML are solutions to everything and the way of the future we
should have taken years ago and they are semantic and free of errors and
extensible and everythingable.

Few people can tell a single _correct_ example of _actual_ benefits from
the use of language markup in HTML documents. One reason to that is that
the benefits are so limited. Another reason is that nobody ever
presented factually correct examples, still less demonstrated them.
People who can repeat the claim that language markup helps in speech
synthesis typically haven't ever actually _heard_ this is in action. So
no wonder they see nothing wrong with examples that use xml:lang (only).

Let us consider one actual benefit: if you open an HTML document in
Microsoft Word (or even copy and paste data from a web page to Word),
language markup gets transferred into Word's internal information about
language. This is useful for spelling and grammar checking (Word
automatically applies the rules of the languages used), for automatic
hyphenation, etc. Now guess (twice, if needed) whether Word recognizes
lang="..." or xml:lang="...". Or just try it.

--
Jukka K. Korpela ("Yucca")

http://www.cs.tut.fi/~jkorpela/

**ray** · Jun 27 '08, 07:16 PM

Re: Why this html doesn't work?

On May 26, 5:41 pm, Gus Richter <gusrich...@net scape.netwrote:

ray wrote:

>
Your document is an HTML document per the doctype declaration.
The META element is an empty element without a closing tag.
In XML, empty elements require a (fwd slash) terminator.
In XHTML, open elements require (space + fwd slash) terminator.
In HTML the META element should look so:
<META http-equiv="Content-Type" content="text/html; charset=UTF-8">
>

>
The SCRIPT element requires a closing tag in both HTML and XHTML,
whether empty or not.
In XML (I believe, since not into XML), an empty element has the closing
tag dropped and the opening tag has a (fwd slash) terminator added.
In HTML the SCRIPT element should look so:
<SCRIPT type="text/javascript" src="test.js"></SCRIPT>
>

If I remove the line that includes the test.js file, the html file
works, otherwise the alert dialog doesn't appear even if the test.js
is empty.

>
With the closing tag it works.
>
--
Gus

Great! Thanks.

Why this html doesn't work?

Why this html doesn't work?

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment