Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.javascript > #22051 > unrolled thread

online gravitational field line explorer...

Back to article view | Back to comp.lang.javascript

Contents

  online gravitational field line explorer... "Chris M. Thomasson" <no@spam.invalid> - 2013-11-18 20:17 -0800
    Re: online gravitational field line explorer... "Chris M. Thomasson" <no@spam.invalid> - 2013-11-18 23:21 -0800
    Re: online gravitational field line explorer... "Evertjan." <exjxw.hannivoort@interxnl.net> - 2013-11-19 09:36 +0000
      Re: online gravitational field line explorer... "Jukka K. Korpela" <jkorpela@cs.tut.fi> - 2013-11-19 12:10 +0200
        Re: online gravitational field line explorer... "Evertjan." <exjxw.hannivoort@interxnl.net> - 2013-11-19 11:01 +0000
          Re: online gravitational field line explorer... "Jukka K. Korpela" <jkorpela@cs.tut.fi> - 2013-11-19 13:33 +0200
            Re: online gravitational field line explorer... "Michael Haufe (TNO)" <tno@thenewobjective.com> - 2013-11-19 07:44 -0800
              Re: online gravitational field line explorer... Thomas 'PointedEars' Lahn <PointedEars@web.de> - 2013-11-19 16:53 +0100
              Re: online gravitational field line explorer... "Chris M. Thomasson" <no@spam.invalid> - 2013-11-19 12:03 -0800
        Re: online gravitational field line explorer... John Harris <niam@jghnorth.org.uk.invalid> - 2013-11-19 16:11 +0000
          Re: online gravitational field line explorer... "Jukka K. Korpela" <jkorpela@cs.tut.fi> - 2013-11-19 19:21 +0200
            Re: online gravitational field line explorer... John Harris <niam@jghnorth.org.uk.invalid> - 2013-11-20 15:48 +0000
              Re: online gravitational field line explorer... "Jukka K. Korpela" <jkorpela@cs.tut.fi> - 2013-11-20 19:15 +0200
                Re: online gravitational field line explorer... John Harris <niam@jghnorth.org.uk.invalid> - 2013-11-21 20:19 +0000
                  Re: online gravitational field line explorer... "Jukka K. Korpela" <jkorpela@cs.tut.fi> - 2013-11-21 22:52 +0200
                    Re: online gravitational field line explorer... John Harris <niam@jghnorth.org.uk.invalid> - 2013-11-22 17:33 +0000
                      Re: online gravitational field line explorer... "Jukka K. Korpela" <jkorpela@cs.tut.fi> - 2013-11-22 19:50 +0200
                        Re: online gravitational field line explorer... John Harris <niam@jghnorth.org.uk.invalid> - 2013-11-23 14:17 +0000
                      Dollar sign in identifier names (was: online gravitational field line explorer...) Christoph Michael Becker <cmbecker69@arcor.de> - 2013-11-22 19:49 +0100
                  Re: online gravitational field line explorer... Eric Bednarz <bednarz@fahr-zur-hoelle.org> - 2013-11-21 22:53 +0100
                    Re: online gravitational field line explorer... Thomas 'PointedEars' Lahn <PointedEars@web.de> - 2013-11-21 23:25 +0100
                    Re: online gravitational field line explorer... John Harris <niam@jghnorth.org.uk.invalid> - 2013-11-22 16:57 +0000
                      Re: online gravitational field line explorer... "Jukka K. Korpela" <jkorpela@cs.tut.fi> - 2013-11-22 19:12 +0200
                        Re: online gravitational field line explorer... Christoph Michael Becker <cmbecker69@arcor.de> - 2013-11-22 19:00 +0100
                          Re: online gravitational field line explorer... "Jukka K. Korpela" <jkorpela@cs.tut.fi> - 2013-11-22 21:04 +0200
                            Re: online gravitational field line explorer... Christoph Michael Becker <cmbecker69@arcor.de> - 2013-11-22 23:46 +0100
                              Re: online gravitational field line explorer... Eric Bednarz <bednarz@fahr-zur-hoelle.org> - 2013-11-23 01:21 +0100
                      Re: online gravitational field line explorer... Eric Bednarz <bednarz@fahr-zur-hoelle.org> - 2013-11-22 18:49 +0100
                        Re: online gravitational field line explorer... Thomas 'PointedEars' Lahn <PointedEars@web.de> - 2013-11-22 20:06 +0100
                          Re: online gravitational field line explorer... Eric Bednarz <bednarz@fahr-zur-hoelle.org> - 2013-11-22 22:49 +0100
                            Re: online gravitational field line explorer... Thomas 'PointedEars' Lahn <PointedEars@web.de> - 2013-11-22 23:52 +0100
                              Re: online gravitational field line explorer... Eric Bednarz <bednarz@fahr-zur-hoelle.org> - 2013-11-23 01:40 +0100
      Re: online gravitational field line explorer... "Chris M. Thomasson" <no@spam.invalid> - 2013-11-19 12:05 -0800
    Re: online gravitational field line explorer... Tiger Sharktopus <tigersharktopus@gmail.com> - 2013-11-19 15:54 -0800
      Re: online gravitational field line explorer... "Chris M. Thomasson" <no@spam.invalid> - 2013-11-21 06:17 -0800
    Re: online gravitational field line explorer... Dr J R Stockton <reply1300@merlyn.demon.co.uk.invalid> - 2013-11-20 19:47 +0000
      Re: online gravitational field line explorer... "Michael Haufe (TNO)" <tno@thenewobjective.com> - 2013-11-20 17:43 -0800
    Re: online gravitational field line explorer... "Chris M. Thomasson" <no@spam.invalid> - 2013-11-21 06:58 -0800
    Re: online gravitational field line explorer... "Chris M. Thomasson" <no@spam.invalid> - 2013-11-22 12:56 -0800
      Re: online gravitational field line explorer... Dr J R Stockton <reply1300@merlyn.demon.co.uk.invalid> - 2013-11-23 18:15 +0000
        Re: online gravitational field line explorer... "Chris M. Thomasson" <no@spam.invalid> - 2013-11-24 13:38 -0800
          Re: online gravitational field line explorer... Dr J R Stockton <reply1300@merlyn.demon.co.uk.invalid> - 2013-11-25 20:47 +0000
            Re: online gravitational field line explorer... "Chris M. Thomasson" <no@spam.invalid> - 2013-11-25 19:51 -0800
              Re: online gravitational field line explorer... Dr J R Stockton <reply1300@merlyn.demon.co.uk.invalid> - 2013-11-27 20:36 +0000
    Re: online gravitational field line explorer... "Chris M. Thomasson" <no@spam.invalid> - 2013-11-29 20:56 -0800
    Re: online gravitational field line explorer... "Chris M. Thomasson" <no@spam.invalid> - 2013-12-01 19:28 -0800

Page 1 of 3  [1] 2 3  Next page →

#22051 — online gravitational field line explorer...

I have created a pre-alpha version of my online gravitational field line
explorer:

http://webpages.charter.net/appcore/fractal/field

This is hard-coded to six points. You can set the (x, y) coordinates and
their mass. Click the `Render' button to draw the damn thing. You can
also alter the number of lines per "positive mass" point and iterations...

I was wondering what you all think of it?

FWIW, I am also going to integrate this with my online circle intersection
fractal explorer:

http://webpages.charter.net/appcore/fractal/ttr

So than one can plot a gravity field on its fractal points. Also, I am
going to add support for equipotential field lines and flexible numbers
of mass points...



Enjoy!

;^)

[toc] | [next] | [standalone]

#22053

> "Chris M. Thomasson"  wrote in message 
> news:l6eopc$2qr$1@speranza.aioe.org...

> I have created a pre-alpha version of my online gravitational field line
> explorer:

> http://webpages.charter.net/appcore/fractal/field

> This is hard-coded to six points. You can set the (x, y) coordinates and
> their mass. Click the `Render' button to draw the damn thing. You can
> also alter the number of lines per "positive mass" point and iterations...

> I was wondering what you all think of it?

> [...]

FWIW, I have updated the code to plot the equipotential field.

:^)


BTW, here is a sample rendering:

http://webpages.charter.net/appcore/fractal/field/samples/equipgfield.png 

[toc] | [prev] | [next] | [standalone]

#22057

Chris M. Thomasson wrote on 19 nov 2013 in comp.lang.javascript:

> I have created a pre-alpha version of my online gravitational field line
> explorer:
> 
> http://webpages.charter.net/appcore/fractal/field

Why?

<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01//EN"
"http://www.w3.org/TR/html4/strict.dtd">
<html xmlns="http://www.w3.org/1999/xhtml" lang="en">

and

<canvas id="TTRCanvas">
Your browser does not support the HTML5 canvas tag.
</canvas>

in stead of:

<!DOCTYPE html>
<html>

[toc] | [prev] | [next] | [standalone]

#22058

2013-11-19 11:36, Evertjan. wrote:

> Chris M. Thomasson wrote on 19 nov 2013 in comp.lang.javascript:
>
>> I have created a pre-alpha version of my online gravitational field line
>> explorer:
>>
>> http://webpages.charter.net/appcore/fractal/field
>
> Why?
>
> <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01//EN"
> "http://www.w3.org/TR/html4/strict.dtd">
> <html xmlns="http://www.w3.org/1999/xhtml" lang="en">
>
> and
>
> <canvas id="TTRCanvas">
> Your browser does not support the HTML5 canvas tag.
> </canvas>
>
> in stead of:
>
> <!DOCTYPE html>
> <html>

Why would it matter? I suppose you mean just the DOCTYPE string, instead 
of suggesting that <!DOCTYPE html> would mean the canvas element 
unnecessary.

In some versions of IE that support <canvas>, Quirks Mode switches the 
support off. But both DOCTYPEs above trigger Standards Mode.

And in markup validation, you can select HTML5 mode in the validator's 
user interface, overriding the effect of DOCTYPE.

Of course new pages are usually best written with <!DOCTYPE html>, and 
with <html lang=...> rather than some odd mess of HTML and XHTML. But 
anyone who wants to be educated in such matters, which are off-topic 
here, can consult suitable HTML resources. Please avoid making the 
impression that you are imitating Lahn the troll.

-- 
Yucca, http://www.cs.tut.fi/~jkorpela/

[toc] | [prev] | [next] | [standalone]

#22059

Jukka K. Korpela wrote on 19 nov 2013 in comp.lang.javascript:

>> and
>>
>> <canvas id="TTRCanvas">
>> Your browser does not support the HTML5 canvas tag.
>> </canvas>
>>
>> in stead of:
>>
>> <!DOCTYPE html>
>> <html>
> 
> Why would it matter? I suppose you mean just the DOCTYPE string, 
instead 
> of suggesting that <!DOCTYPE html> would mean the canvas element 
> unnecessary.

It would matter, 
if you [general you] specify that HTML5 support is necessary,
for the required running of the Javascript script.

> Please avoid making the 
> impression that you are imitating Lahn the troll.

Off topic, Jukka.

I do not "make" such impression, this
impression is what you seem to deduct in your mind.

So better ask yourself just as politely
to refrain from such confessed painful deduction.

[toc] | [prev] | [next] | [standalone]

#22061

2013-11-19 13:01, Evertjan. wrote:

> It would matter,
> if you [general you] specify that HTML5 support is necessary,
> for the required running of the Javascript script.

As a statement about relevance of <!DOCTYPE html> to such support, this 
would be on-topic, but the content of the statement is not true. Or can 
you cite any evidence?

In HTML5 CR, there isn't even any hint of any possible effect of using 
"legacy DOCTYPE" vs. <!DOCTYPE html>. A browser that would treat them 
differently would be non-conforming in a pointless way.

-- 
Yucca, http://www.cs.tut.fi/~jkorpela/

[toc] | [prev] | [next] | [standalone]

#22064

On Tuesday, November 19, 2013 5:33:26 AM UTC-6, Jukka K. Korpela wrote:
> 2013-11-19 13:01, Evertjan. wrote:
> 
> > It would matter,
> > if you [general you] specify that HTML5 support is necessary,
> > for the required running of the Javascript script.
> 
> As a statement about relevance of <!DOCTYPE html> to such support, this 
> would be on-topic, but the content of the statement is not true. Or can 
> you cite any evidence?


IIUC, I believe the criticism is that Chris used an HTML4 doctype while explicitly requesting HTML5 support in his <canvas/> fallback text

[toc] | [prev] | [next] | [standalone]

#22065

Michael Haufe (TNO) wrote:

> On Tuesday, November 19, 2013 5:33:26 AM UTC-6, Jukka K. Korpela wrote:
>> 2013-11-19 13:01, Evertjan. wrote:
>> > It would matter,
>> > if you [general you] specify that HTML5 support is necessary,
>> > for the required running of the Javascript script.
>> 
>> As a statement about relevance of <!DOCTYPE html> to such support, this
>> would be on-topic, but the content of the statement is not true. Or can
>> you cite any evidence?
> 
> IIUC, I believe the criticism is that Chris used an HTML4 doctype while
> explicitly requesting HTML5 support in his <canvas/> fallback text

Not only that.  From what was quoted here, he completely mixed up the HTML 
4.01 Strict DOCTYPE declaration, the XML namespace declaration for *XHTML*, 
and the *HTML5* “canvas” element.  Markup one can rely on in one's program 
looks quite different, and “Evertjan.” has shown how; however, the markup 
has to be validated afterwards because Valid HTML5 (as per CR at least) 
follows own rules, and not obeying those *again* results in an unreliable 
markup base.

-- 
PointedEars

Twitter: @PointedEars2
Please do not Cc: me. / Bitte keine Kopien per E-Mail.

[toc] | [prev] | [next] | [standalone]

#22074

> "Michael Haufe (TNO)"  wrote in message 
> news:c013acce-eb4c-4830-a165-564c083fe86c@googlegroups.com...

> On Tuesday, November 19, 2013 5:33:26 AM UTC-6, Jukka K. Korpela wrote:
> > 2013-11-19 13:01, Evertjan. wrote:
> >
> > > It would matter,
> > > if you [general you] specify that HTML5 support is necessary,
> > > for the required running of the Javascript script.
> >
> > As a statement about relevance of <!DOCTYPE html> to such support, this
> > would be on-topic, but the content of the statement is not true. Or can
> > you cite any evidence?


> IIUC, I believe the criticism is that Chris used an HTML4 doctype while 
> explicitly
> requesting HTML5 support in his <canvas/> fallback text

Exactly right. I will correct the issue as soon as I can. Thank god the 
problem is not in the
JavaScript!

;^) 

[toc] | [prev] | [next] | [standalone]

#22066

On Tue, 19 Nov 2013 12:10:18 +0200, "Jukka K. Korpela"
<jkorpela@cs.tut.fi> wrote:

>2013-11-19 11:36, Evertjan. wrote:
>
>> Chris M. Thomasson wrote on 19 nov 2013 in comp.lang.javascript:
>>
>>> I have created a pre-alpha version of my online gravitational field line
>>> explorer:
>>>
>>> http://webpages.charter.net/appcore/fractal/field
>>
>> Why?
>>
>> <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01//EN"
>> "http://www.w3.org/TR/html4/strict.dtd">
>> <html xmlns="http://www.w3.org/1999/xhtml" lang="en">
>>
>> and
>>
>> <canvas id="TTRCanvas">
>> Your browser does not support the HTML5 canvas tag.
>> </canvas>
>>
>> in stead of:
>>
>> <!DOCTYPE html>
>> <html>
>
>Why would it matter? I suppose you mean just the DOCTYPE string, instead 
>of suggesting that <!DOCTYPE html> would mean the canvas element 
>unnecessary.
  <snip>

Why would it matter? Because if you give the browser misleading
information about what kind of document it is you can't be sure what
kind of document object model will be built. You then don't know if
your DOM code is going to work, and it's your fault, not the
browser's.

Relying on implementation quirks to make your code work is the work of
a cowboy, not of a professional.

  John

[toc] | [prev] | [next] | [standalone]

#22068

2013-11-19 18:11, John Harris wrote:

> Why would it matter? Because if you give the browser misleading
> information about what kind of document it is you can't be sure what
> kind of document object model will be built.

As I wrote in another reply, HTML5 CR itself says that <!DOCTYPE html> 
(often called "HTML5 doctype", which is thus partly misleading) has no 
effect apart from being (along with HTML 4.01 doctype) one of the 
DOCTYPEs that trigger standards (non-quirks) mode.

Given the lack of all evidence supporting the cargo cult belief about 
needing <!DOCTYPE html> as opposite to valid legacy doctypes, it is 
astonishing how widespread it is. But as we know, people do not actually 
check specs/drafts, so no wonder they have missed the "legacy doctype" 
part. Well, that's not so serious, really. But it becomes somewhat 
absurd when people with such beliefs start giving off-topic lectures 
based on them, almost in Lahn style.

-- 
Yucca, http://www.cs.tut.fi/~jkorpela/

[toc] | [prev] | [next] | [standalone]

#22081

On Tue, 19 Nov 2013 19:21:08 +0200, "Jukka K. Korpela"
<jkorpela@cs.tut.fi> wrote:

>2013-11-19 18:11, John Harris wrote:
>
>> Why would it matter? Because if you give the browser misleading
>> information about what kind of document it is you can't be sure what
>> kind of document object model will be built.
>
>As I wrote in another reply, HTML5 CR itself says that <!DOCTYPE html> 
>(often called "HTML5 doctype", which is thus partly misleading) has no 
>effect apart from being (along with HTML 4.01 doctype) one of the 
>DOCTYPEs that trigger standards (non-quirks) mode.
>
>Given the lack of all evidence supporting the cargo cult belief about 
>needing <!DOCTYPE html> as opposite to valid legacy doctypes, it is 
>astonishing how widespread it is. But as we know, people do not actually 
>check specs/drafts, so no wonder they have missed the "legacy doctype" 
>part. Well, that's not so serious, really. But it becomes somewhat 
>absurd when people with such beliefs start giving off-topic lectures 
>based on them, almost in Lahn style.

How on earth do you deduce any of that from the HTML5 Candidate
Recommendation 6 August 2013, which says :

  <quote>
A DOCTYPE is a required preamble.

DOCTYPEs are required for legacy reasons. When omitted, browsers tend
to use a different rendering mode that is incompatible with some
specifications. Including the DOCTYPE in a document ensures that the
browser makes a best-effort attempt at following the relevant
specifications.
  </quote>

In particular, note the first and last sentences.

  John

[toc] | [prev] | [next] | [standalone]

#22082

2013-11-20 17:48, John Harris wrote:

> How on earth do you deduce any of that from the HTML5 Candidate
> Recommendation 6 August 2013

By actually reading what it says, in the cited section and otherwise. 
And by taking some look at reality, just for confirmation.

> A DOCTYPE is a required preamble.

That's a formal statement. It means that a document without a DOCTYPE is 
not conforming. And it refers to DOCTYPEs in general, as described in 
the CR - incuding legacy DOCTYPEs.

Please let me remind you that the dispute was over using <!DOCTYPE html> 
vs. legacy DOCTYPEs. Nothing in what you quote deals with that issue.

> DOCTYPEs are required for legacy reasons. When omitted, browsers tend
> to use a different rendering mode that is incompatible with some
> specifications. Including the DOCTYPE in a document ensures that the
> browser makes a best-effort attempt at following the relevant
> specifications.

That's just a half-formalese attempt at saying that in the absence of a 
DOCTYPE (either <!DOCTYPE html> or a conforming legacy DOCTYPE), 
browsers will (or may) fall into Quirks Mode.

Quirks Mode is for the most of it a matter of (mis)interpreting CSS, 
with some impact in HTML and JavaScript interpretation, too. The most 
important JavaScript impact is probably that in Quirks Mode, Firefox 
does not support document.all at all (thereby violating HTML5 CR, but 
the spec might get retrofitted to this reality at some point).

Now, to return to what was really being discussed, can you present any 
evidence about the effect of using a legacy DOCTYPE vs. using <!DOCTYPE 
html>, in *any* way (CSS, HTML, JavaScript, whatever)? In fact, to 
really address the issue here, the evidence should be about support to 
the <canvas> element. But first, *any* evidence?

-- 
Yucca, http://www.cs.tut.fi/~jkorpela/

[toc] | [prev] | [next] | [standalone]

#22104

On Wed, 20 Nov 2013 19:15:16 +0200, "Jukka K. Korpela"
<jkorpela@cs.tut.fi> wrote:

>2013-11-20 17:48, John Harris wrote:
>
>> How on earth do you deduce any of that from the HTML5 Candidate
>> Recommendation 6 August 2013
>
>By actually reading what it says, in the cited section and otherwise. 
>And by taking some look at reality, just for confirmation.
>
>> A DOCTYPE is a required preamble.
>
>That's a formal statement. It means that a document without a DOCTYPE is 
>not conforming. And it refers to DOCTYPEs in general, as described in 
>the CR - incuding legacy DOCTYPEs.

I disagree. The about-to-be standard for HTML5 says that every HTML5
document has a doctype at the beginning, otherwise it is *not* an
HTML5 document. It also says that the doctype pseudo element must be 
  <!DOCTYPE html>
or
  <!DOCTYPE html SYSTEM "about:legacy-compat">
with some letters case-insensitive and a choice of ' or ".
Even then, the second case is only allowed if the document is being
written by an out-of-date, "legacy", program.


>Please let me remind you that the dispute was over using <!DOCTYPE html> 
>vs. legacy DOCTYPEs. Nothing in what you quote deals with that issue.

The legacy DOCTYPEs, by which I assume you mean SGML-like document
type declarations, say that the document is one of HTML 1 to 4.01. In
these document types canvas elements are not a requirement. The
program shouldn't rely on them being implemented in a customer's
browser.

  <snipped : Dealt with by the above>

>Now, to return to what was really being discussed, can you present any 
>evidence about the effect of using a legacy DOCTYPE vs. using <!DOCTYPE 
>html>, in *any* way (CSS, HTML, JavaScript, whatever)? 

This is irrelevant to the discussion, but if you go to IE5 or
thereabouts and do 'save as .mht file' for a page with a doc type of
HTML 4.01 IE changed the doc type to HTML 4 as it didn't know about
the later HTML version.


>In fact, to 
>really address the issue here, the evidence should be about support to 
>the <canvas> element. But first, *any* evidence?

I think you'll find that NN4 didn't know what <canvas> meant.

  John

[toc] | [prev] | [next] | [standalone]

#22105

2013-11-21 22:19, John Harris wrote:

> The about-to-be standard for HTML5 says that every HTML5
> document has a doctype at the beginning,

That formal statement was not under dispute.

> Even then, the second case is only allowed if the document is being
> written by an out-of-date, "legacy", program.

You are just making this up. HTML5 CR clearly says that an obsolete 
doctype string is conforming (though not recommended). And this is just 
a formality. The real question was whether it actually matters which of 
the conforming doctypes you use. You are not saying anything about this, 
which is quite understandable. It is less understandable why you 
continue babbling about doctypes as if they ever meant in reality what 
they mean in SGML (naturally misunderstanding completely what they mean 
in SGML).

>> Now, to return to what was really being discussed, can you present any
>> evidence about the effect of using a legacy DOCTYPE vs. using <!DOCTYPE
>> html>, in *any* way (CSS, HTML, JavaScript, whatever)?
>
> This is irrelevant to the discussion,

That was the very question being discussed.

> but if you go to IE5 or
> thereabouts and do 'save as .mht file' for a page with a doc type of
> HTML 4.01 IE changed the doc type to HTML 4 as it didn't know about
> the later HTML version.

Now *that* is completely irrelevant to the question. Or you are saying 
that IE5 made some decisions differently depending on the presence of 
<!doctype html> vs. legacy doctypes?

>> In fact, to
>> really address the issue here, the evidence should be about support to
>> the <canvas> element. But first, *any* evidence?
>
> I think you'll find that NN4 didn't know what <canvas> meant.

So you are saying that NN4 made some decisions differently depending on 
the presence of <!doctype html> vs. legacy doctypes?

Or are you just not getting at all what the question is about? Let me 
remind you: it was the practical effects of using <!doctype html> vs. 
legacy doctypes. (There was a theoretical question mentioned too, but it 
was already solved: they are conforming.)

-- 
Yucca, http://www.cs.tut.fi/~jkorpela/

[toc] | [prev] | [next] | [standalone]

#22138

On Thu, 21 Nov 2013 22:52:15 +0200, "Jukka K. Korpela"
<jkorpela@cs.tut.fi> wrote:

>2013-11-21 22:19, John Harris wrote:
>
>> The about-to-be standard for HTML5 says that every HTML5
>> document has a doctype at the beginning,
>
>That formal statement was not under dispute.

It's not 'formal', whatever you mean by that, it's a specification.
Without a suitable doctype it's not an HTML5 document.


>> Even then, the second case is only allowed if the document is being
>> written by an out-of-date, "legacy", program.
>
>You are just making this up. 

Are you sure you are reading the W3C Candidate Recommendation 6 August
2013 ? 


>HTML5 CR clearly says that an obsolete 
>doctype string is conforming (though not recommended). And this is just 
>a formality. 

In the W3C copy of the CR it's a very specific "DOCTYPE legacy
string". It is not one that can be confused with doctypes for earlier
HTMLs.

And it's not "just a formality", it bloody-well specifies.


>The real question was whether it actually matters which of 
>the conforming doctypes you use. You are not saying anything about this, 
>which is quite understandable. 

I've already said that either specified doctype can be used, but like
the $ identifier the legacy one is restricted to machinery.


>It is less understandable why you 
>continue babbling about doctypes as if they ever meant in reality what 
>they mean in SGML (naturally misunderstanding completely what they mean 
>in SGML).

All right, I'll say it one more time. If you don't include a doctype
you can't say that your DOM-accessing code should work. And you
shouldn't be surprised if it doesn't work.


>>> Now, to return to what was really being discussed, can you present any
>>> evidence about the effect of using a legacy DOCTYPE vs. using <!DOCTYPE
>>> html>, in *any* way (CSS, HTML, JavaScript, whatever)?
>>
>> This is irrelevant to the discussion,
>
>That was the very question being discussed.
>
>> but if you go to IE5 or
>> thereabouts and do 'save as .mht file' for a page with a doc type of
>> HTML 4.01 IE changed the doc type to HTML 4 as it didn't know about
>> the later HTML version.
>
>Now *that* is completely irrelevant to the question. Or you are saying 
>that IE5 made some decisions differently depending on the presence of 
><!doctype html> vs. legacy doctypes?

No, I'm saying the value of the doctype made a difference, an example
you asked for.


>>> In fact, to
>>> really address the issue here, the evidence should be about support to
>>> the <canvas> element. But first, *any* evidence?
>>
>> I think you'll find that NN4 didn't know what <canvas> meant.
>
>So you are saying that NN4 made some decisions differently depending on 
>the presence of <!doctype html> vs. legacy doctypes?

No, I'm saying that without a doctype you can't pretend that canvas
should work.


>Or are you just not getting at all what the question is about? Let me 
>remind you: it was the practical effects of using <!doctype html> vs. 
>legacy doctypes. (There was a theoretical question mentioned too, but it 
>was already solved: they are conforming.)

The question is all about whether cowboy programming is a good idea,
or not.

  John

[toc] | [prev] | [next] | [standalone]

#22141

2013-11-22 19:33, John Harris wrote:

> it's a specification.

Really? Did you miss this: "This is a draft document and may be updated, 
replaced or obsoleted by other documents at any time. It is 
inappropriate to cite this document as other than work in progress."?

> I've already said that either specified doctype can be used, but like
> the $ identifier the legacy one is restricted to machinery.

That does not parse. Well, it parses, but it makes absolutely no sense.

> All right, I'll say it one more time. If you don't include a doctype
> you can't say that your DOM-accessing code should work.

You still haven't given anything even remotely resembling evidence for 
that, formal or practical.

>> Now *that* is completely irrelevant to the question. Or you are saying
>> that IE5 made some decisions differently depending on the presence of
>> <!doctype html> vs. legacy doctypes?
>
> No, I'm saying the value of the doctype made a difference, an example
> you asked for.

This was not about <!DOCTYPE I just made this up>. What I asked for was 
very specific: conforming legacy doctypes vs. <!DOCTYPE html>. That is, 
the topic of the discussion.

> The question is all about whether cowboy programming is a good idea,
> or not.

That might be your question, but not the one discussed here. Have you 
considered stopping making foolish remarks if you have absolutely no 
evidence for the claims you have made about the topic of the discussion?

-- 
Yucca, http://www.cs.tut.fi/~jkorpela/

[toc] | [prev] | [next] | [standalone]

#22162

On Fri, 22 Nov 2013 19:50:24 +0200, "Jukka K. Korpela"
<jkorpela@cs.tut.fi> wrote:


  <snip>

As our understanding of the same English text is so different it's
clear that there's no point in continuing this discussion.

  John

[toc] | [prev] | [next] | [standalone]

#22143 — Dollar sign in identifier names (was: online gravitational field line explorer...)

John Harris wrote:

> I've already said that either specified doctype can be used, but like
> the $ identifier the legacy one is restricted to machinery.

While the third edition of ECMA-262[1] specified in section 7.6

  "The dollar sign is intended for use only in mechanically generated
   code."

this statement was removed in the 5.1 Edition[2].  One may guess for
what reason. ;)

[1]
<http://www.ecma-international.org/publications/files/ECMA-ST-ARCH/ECMA-262,%203rd%20edition,%20December%201999.pdf>
[2] <http://www.ecma-international.org/ecma-262/5.1/Ecma-262.pdf>

-- 
Christoph M. Becker

[toc] | [prev] | [next] | [standalone]

#22107

John Harris <niam@jghnorth.org.uk.invalid> writes:

> The legacy DOCTYPEs, by which I assume you mean SGML-like document
> type declarations, say that the document is one of HTML 1 to 4.01.

I don't know what you know about SGML, but I remember that you know
OOP. Please write this 500.000 times:
SGML document type declarations have no "is a" semantics. TIA.

What's more, there is no "HTML 1".

[toc] | [prev] | [next] | [standalone]

Page 1 of 3  [1] 2 3  Next page →

Back to top | Article view | comp.lang.javascript

csiph-web