who uses virtual pc for testing webpages

[Eric B. Bednarz]

1) Bugs rarely if ever effect STANDARD HTML.

Maybe. But ISO/IEC 15445:2000 is hardly used, or even known to the
public.

2) rendering is illrelevent.

How sad. Frankly, I'd rather have my illrelevant stuff rendered one or
the other way nevertheless.

The sooner the dimwits that call themselves HTML writers understand HTML is
NOT nor NEVER will be a page layout format the better it will be for
everyone else.

The fact that it started out that way might have something to do with
that.

Any modern brower out there is HTML 4.01 complient.

Surprising. I haven't seen a single one yet (as an estimate, I have
about 30, not all of them modern, though).

I challange someone to show a *valid* (ie W3c complient) online HTML file
doing something improper

Yawn.

<http://sandbox.bednarz.nl/sgml/html/version_information.html>

Mozilla can't even display the source properly. Opera 7 at least has
made some improvements IMS.

 

[Bruce Grubb]

You seem to be assuming that browsers perfectly support the specs. They
don't.

Display quirks aside just what spec non-support are well know bugs?

 

[Bruce Grubb]

?? fatigue?

Huh? Without the alt tah you get the [image] [image] [image] crap all the
time. Now what on earth does fatigue have to do with this?

 

[Duende]

While sitting in a puddle Bruce Grubb scribbled in the mud:

?? fatigue?

Huh? Without the alt tah you get the [image] [image] [image] crap all
the time. Now what on earth does fatigue have to do with this?

What is a tag?

 

[Toby A Inkster]

Display quirks aside just what spec non-support are well know bugs?

Almost any modern browser has problems with its parser: most will not
allow:

<b<i>This text is bold and italic</></>

even though that is valid HTML. Many have problems with comments.

<!-- This is a comment ->
<><><>
This is still a comment, but many browsers will render it. ---->
<-><-><->
This is still a comment, but many browsers will render it. -->

And Internet Explorer completely ignores <abbr>.

That's all I can think of off the top of my head first thing in the
morning, but I'm sure more will come to me through the day.

 

[Bruce Grubb]

Almost any modern browser has problems with its parser: most will not
allow:

<b<i>This text is bold and italic</></>

The specs specifically warn against using SGML SHORTTAG constructs as they
"unlikely to work with many existing HTML tools." Appendix B.3.7.

Therefore using this type of thing is NOT following the 4.01 spec.

even though that is valid HTML. Many have problems with comments.

<!-- This is a comment ->
Correct

<><><>
This is still a comment, but many browsers will render it. ---->
<-><-><->
This is still a comment, but many browsers will render it. -->

Not according to 3.2.4 Comments section. Again non spec garbage causing
problems.

And Internet Explorer completely ignores <abbr>.

Does anybody use <abbr> enough to knwo that this si really a problem? and
thent he question come as to why MS doesn't fix it?

 

[Toby A Inkster]

The specs specifically warn against using SGML SHORTTAG constructs as they
"unlikely to work with many existing HTML tools." Appendix B.3.7.

Therefore using this type of thing is NOT following the 4.01 spec.

Appendix B is non-normative. That is, it can be completely ignored and
you'll still be following the specification.

 

[Bruce Grubb]

Appendix B is non-normative. That is, it can be completely ignored and
you'll still be following the specification.

But the other sections use examples of non SGML SHORTTAG, so again using
SGML SHORTTAG constructs is NOT following the spec.

So far ALL the counter points to simply following the HTML 4.01 rather than
testing on every browser out there have consisted of:

1) Browser detection crap (javascript, java, etc)

2) Minor display hiccups (HTML is NOT a page layout format people)

3) NONstandard garbage.

Nothing really of merit. Give it up. You write the to spec and do not use
java or javascript (which have their own set of problems) and you will not
need to test on every browser you can get your hands on. Pure and simple.

 

[Toby A Inkster]

Give it up. You write the to spec and do not use
java or javascript (which have their own set of problems) and you will not
need to test on every browser you can get your hands on.

Which brings us back to my original example:

http://www.goddamn.co.uk/tobyink/scratch/example

Browsers often make mistakes, even when given valid HTML with, no Javascript,
no Java, etc -- so there is no substitute for testing.

 

[Mark Parnell]

Browsers often make mistakes, even when given valid HTML with, no Javascript,
no Java, etc -- so there is no substitute for testing.

And then when CSS gets involved...BOOM! ;-)

 

[Bruce Grubb]

Which brings us back to my original example:

http://www.goddamn.co.uk/tobyink/scratch/example

Which has javascript based browser detection crap hidden as a comment:

<!--[if gte IE 5]><input type example><![endif]

As I said before so far ALL the counter points to simply following the HTML
4.01 rather than testing on every browser out there have consisted of:

1) Browser detection crap (javascript, java, etc)

2) Minor display hiccups (HTML is NOT a page layout format people)

3) NONstandard garbage.

Nothing really of merit. Give it up. You write the to spec and do not use
java or javascript (which have their own set of problems [ie like being
executed even if commented out as per the Inksters's examples page) and you
will not need to test on every browser you can get your hands on. Pure and
simple.

 

[Mark Parnell]

Which has javascript based browser detection crap hidden as a comment:

There is no Javascript on Toby's page. Google for conditional comments.

 

[Eric B. Bednarz]

Which has javascript based browser detection crap hidden as a comment:

Please explain where Javascript[tm] or even active scriting is used on
Toby's page, file a bug report if necessary or RTFM for a change.
It's a feature -- and if your UA supports it, just drop it, if

3) NONstandard garbage.

is supported.

<http://www.w3.org/TR/html401/references.html#h-1.1>
Reading up on (especially) the *standards* part is mandatory. You are
not excused.

 

[Bruce Grubb]

Which has javascript based browser detection crap hidden as a comment:

Please explain where Javascript[tm] or even active scriting is used on
Toby's page, file a bug report if necessary or RTFM for a change.

I did read the manual that how i know the

<!--[if gte IE 5]><input type example><![endif]

garbage a Javascript command hidden as a comment.

The implimention of Javascript and Java is a know crossplatform headache.

It's a feature -- and if your UA supports it, just drop it, if

Good grief this is the drivel Mac users get when Windows or programs ported
form Windows does something royally stupid - 'its not a bug its a feature'

To parafrase the Bard 'A skunk by any other name still stinks to high
heaven.'

If you really want to see this stupitiy carried to its lowest level go to
google look for the 'CR2 Expansion Bug/Feature' thread in
rec.games.frp.dnd.

years ago Mac users had a bad joke:

How many Microsoft programmers does it take to change a light bulb?

None. They declare darkness the new standard.

Today there is a new line to the joke:

Which explains why Windows users and programers have been stumbling around
int eh dark for the last 10 years.

is supported.

<http://www.w3.org/TR/html401/references.html#h-1.1>
Reading up on (especially) the *standards* part is mandatory. You are
not excused.

It is still nonstandard garbage and the support for it is iffy because it
is nonstandard garbage.

 

[Toby A Inkster]

I did read the manual that how i know the
<!--[if gte IE 5]><input type example><![endif]>
garbage a Javascript command hidden as a comment.

It is HTML. It is not Javascript.

 

[Eric B. Bednarz]

I did read the manual that how i know the

<!--[if gte IE 5]><input type example><![endif]

garbage a Javascript command hidden as a comment.

For starters, technically it *is* a comment (as opposed to e.g. the
cargo-cultish in(cl|tr)usion of '<!--' and '-->' in elements that
misfeature CDATA as a content model).

It would be interesting to know which manual you have read, in any case
you can read up on comment declarations at
<http://www.w3.org/MarkUp/SGML/productions.html#prod91>

Not only does M$' excuse for documentation of its homegrown conditional
featurism imply that no scripting is involved

<http://msdn.microsoft.com/workshop/author/dhtml/overview/ccomment_ovw.asp>
| Conditional comments offer certain advantages over scripted browser
| detection techniques

it also works in practice when active scripting, vbscript and whatnot
are disabled. So maybe you should reveal your source of information
respectively concern.

years ago Mac users had a bad joke:

Well, years ago Mac users were the pioneers in point and drool
interface pollution. I'm not sure whether that's a suitable environment
for light bulb jokes.

How many Microsoft programmers does it take to change a light bulb?

It is still nonstandard garbage

A couple of things you have been referring to as 'nonstandard garbage'
before are documented by an international *standard* (as opposed to w3c
*recommendations*) that is used as a normative reference by said
recommendation (and if you discuss subjects of the recommendation you
*are* familiar with all of those *normative* referenes, of course). Now
how many Macintoy users does it take to grok the consequences?

 

[Bruce Grubb]

I did read the manual that how i know the

<!--[if gte IE 5]><input type example><![endif]

garbage a Javascript command hidden as a comment.

For starters, technically it *is* a comment (as opposed to e.g. the
cargo-cultish in(cl|tr)usion of '<!--' and '-->' in elements that
misfeature CDATA as a content model).

It would be interesting to know which manual you have read, in any case
you can read up on comment declarations at
<http://www.w3.org/MarkUp/SGML/productions.html#prod91>

HTML Complete (ISBM 0-7821-2421-2467-4) pg 554

In fact this section is CALLED "Using HTML Comments to hide JavaScript Code"

"The comment tags cause the JavaScript statements to be treated as coments
by JavaScript-challanged browers. JavaScript-enabled-enabled browsers on
the the other hand, know enough to ignore the comments tags and process the
enclosed statements as JavaScript."

Not only does M$' excuse for documentation of its homegrown conditional
featurism imply that no scripting is involved

WRONG see "18.3.2 Hiding script data from user agents"
<http://tinyurl.com/2j28>

 

[Bruce Grubb]

I did read the manual that how i know the
<!--[if gte IE 5]><input type example><![endif]>
garbage a Javascript command hidden as a comment.

It is HTML. It is not Javascript.

WRONG

"Using HTML Comments to hide JavaScript Code

"The comment tags cause the JavaScript statements to be treated as comments
by JavaScript-challanged browers. JavaScript-enabled browsers on the other
hand, know enough to ignore the comments tags and process the enclosed
statements as JavaScript."

(HTML Complete (ISBM 0-7821-2421-2467-4) pg 554)


"Documents that do not specify default scripting language information and
that contain elements that specify an intrinsic event script are
incorrect." (18.2.2 Specifying the scripting language)
<http://tinyurl.com/2j28>

As I said before nonstandard garbage and I have TWO QUOTED sources to back
me up. The only refutiation allowable now is QUOTED soucres that say the
above is correct. In short you lose.

 

[Bruce Grubb]

Which has javascript based browser detection crap hidden as a comment:

Please explain where Javascript[tm] or even active scriting is used on
Toby's page, file a bug report if necessary or RTFM for a change.
It's a feature -- and if your UA supports it, just drop it, if

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<title>Example</title>
<body><!--[if gte IE 5]><input type example><![endif]--></body>

It is not a feature as it violates HTML 4.01's section on scripts
<http://tinyurl.com/34lbd>

Furthermore '18.3.2 Hiding script data from user agents' renders
Conditional Comments as NONSTANDARD garbage.

"As HTML does not rely on a specific scripting language, document authors
must explicitly tell user agents the language of each script. This may be
done either through a default declaration or a local declaration. [...]
Documents that do not specify default scripting language information and
that contain elements that specify an intrinsic event script are incorrect."

The above does not use the script element as required by 18.2.1, it does
not use the required type =content-type [CI] element, and there is no
character that comments out the ending --> which as all three examples in
18.3.2 imply is also required.

NONSTADARD scripting garbage. No wonder it fails. Come back with STANDARD
examples.

 

[Steve Pugh]

I did read the manual that how i know the

<!--[if gte IE 5]><input type example><![endif]

garbage a Javascript command hidden as a comment.

For starters, technically it *is* a comment (as opposed to e.g. the
cargo-cultish in(cl|tr)usion of '<!--' and '-->' in elements that
misfeature CDATA as a content model).

It would be interesting to know which manual you have read, in any case
you can read up on comment declarations at
<http://www.w3.org/MarkUp/SGML/productions.html#prod91>

HTML Complete (ISBM 0-7821-2421-2467-4) pg 554

In fact this section is CALLED "Using HTML Comments to hide JavaScript Code"

"The comment tags cause the JavaScript statements to be treated as coments
by JavaScript-challanged browers. JavaScript-enabled-enabled browsers on
the the other hand, know enough to ignore the comments tags and process the
enclosed statements as JavaScript."

That's presumably referring to the outdated cargo cult of putting
comments inside <script> elements to hide the contents from Netscape
1.

Nothing to do with IE's conditional statements that are not JavaScript
at all.

WRONG see "18.3.2 Hiding script data from user agents"
<http://tinyurl.com/2j28>

Jiang gives Chinese reformers little hope ?

Steve