silly HTML

Fleeing from the madness of the http://groups.google.com jungle
Andy Dingley <dingbat.TakeThisOut@codesmiths.com> stumbled into news:alt.www.webmaster
and said:

>
> William Tasso wrote:
>
>> > Legend is an inline element.
>>
>> yes, yes, I know that
>>
>> Doesn't stop it being silly.
>
> Silly? Simple, certainly.
>
> The alternative would be to push "blockness" further down into the
> children of <form> and that would complicate their layout
> substantially.
>
> Also a <h*> element isn't a caption, it's a header. "Header" iin HTML
> terms is more to do with the logical structuring of document content,
> not just captioning. In these terms then headers are inherently block
> elements rather than inline. Conversely your intended use of one here
> (inside the <form>) is implicitly wrong because it has the behaviour of
> being a caption (arguably just presentational) but not that of being a
> header (in the HTML sense).

Oh yes - I know the spec and the arguments.

This is a long and (potentially) unweildy form that is the page content.
In those terms the structure would suit each fieldet having its own
structural blocklevel heading.

Sorted now anyway.

--
William Tasso

http://williamtasso.com/words/what-is-usenet.asp
 >> Stay informed about: silly HTML 

hug wrote:

> Yah, like it's not a full and accurate parser.

For all intents and purposes the W3C validator *is* a full and accurate
parser. There may be one or two little bugs in it, but you'd have to throw
some highly esoteric markup at it to trip one up.

> I once had it (the w3c one) tell me there was no "id" attribute in a
> particular tag, and the "id" it was pointing at was part of an argument
> list in a url.

Even without seeing the page, I can guess that the error was not saying
that there was no such *attribute* as 'id', but there was no such *entity*
as '&id;'. You almost certainly had, as part of a URL:

&id=blah

The validator was correct in pointing out this error. You must always
"escape" ampersands in URLs (indeed, anywhere in your HTML when not
already being used to indicate an entity) like so:

&id=blah

--
Toby A Inkster BSc (Hons) ARCS
Contact Me ~ http://tobyinkster.co.uk/contact
 >> Stay informed about: silly HTML 

Toby Inkster <usenet200611 RemoveThis @tobyinkster.co.uk> wrote:

>You must always
>"escape" ampersands in URLs (indeed, anywhere in your HTML when not
>already being used to indicate an entity) like so:
>
> &id=blah

Okay Toby; I won't argue the point, but I will ask you where this is
specified. To me, anything inside a quoted string in a tag has
already been "escaped" by virtue of being inside the quoted string, so
using html entity-names for those characters inside a quoted string
seems "wrong". I'd like to find the real deal, I think fixing it
everywhere would take me about 5-10 minutes, but I'm averse to fixing
what works to do something someone says is correct without having a
full understanding for why that's so. Thanks.

--
Legacy browsers never heard of emerging standards.
(contact via http://www.ren-prod-inc.com/hug_soft)
 >> Stay informed about: silly HTML 

hug wrote:

> Toby Inkster <usenet200611.RemoveThis@tobyinkster.co.uk> wrote:
>
> >You must always
> >"escape" ampersands in URLs (indeed, anywhere in your HTML when not
> >already being used to indicate an entity) like so:
> >
> > &id=blah
>
> Okay Toby; I won't argue the point, but I will ask you where this is
> specified. To me, anything inside a quoted string in a tag has
> already been "escaped" by virtue of being inside the quoted string, so
> using html entity-names for those characters inside a quoted string
> seems "wrong". I'd like to find the real deal, I think fixing it
> everywhere would take me about 5-10 minutes, but I'm averse to fixing
> what works to do something someone says is correct without having a
> full understanding for why that's so. Thanks.

Probably using the expression 'escaped ampersand' I was wrong,
but it is an accepted expression for writing the
ampersand character & as &
As you know HTML uses & when representing characters,
for example the character < as part of a string (not mark-up)
is represented as <
or the copyright character as ©
so it is easy to understand why a & character within a string
should be written as &

Also, inside an XML document the character & has to be written
as & otherwise there is a serious XML syntax error,
and this applies to XHTML documents.

Usually this error is shown and well explained in the W3C validator.
The validator's FAQ points to the following URL
for explanation of the ampersand in URL strings error
http://www.htmlhelp.com/tools/validator/problems.html#amp
 >> Stay informed about: silly HTML 

hug wrote:

> Okay Toby; I won't argue the point, but I will ask you where this is
> specified.

ISO 8879 (Standard for SGML), which is normatively referred to by the HTML
4.01 spec.

--
Toby A Inkster BSc (Hons) ARCS
Contact Me ~ http://tobyinkster.co.uk/contact
 >> Stay informed about: silly HTML 

Toby Inkster <usenet200611.TakeThisOut@tobyinkster.co.uk> wrote:

>hug wrote:
>
>> Okay Toby; I won't argue the point, but I will ask you where this is
>> specified.
>
>ISO 8879 (Standard for SGML), which is normatively referred to by the HTML
>4.01 spec.

Thanks Toby.

--
Legacy browsers never heard of emerging standards.
(contact via http://www.ren-prod-inc.com/hug_soft)
 >> Stay informed about: silly HTML 

Toby Inkster <usenet200611.RemoveThis@tobyinkster.co.uk> wrote:

>hug wrote:
>
>> Okay Toby; I won't argue the point, but I will ask you where this is
>> specified.
>
>ISO 8879 (Standard for SGML), which is normatively referred to by the HTML
>4.01 spec.

Huh. It seems everyplace where I find ISO 8879 wants you to buy a
hardcopy manual. Are you aware of a web location where its complete
text is published?

--
Legacy browsers never heard of emerging standards.
(contact via http://www.ren-prod-inc.com/hug_soft)
 >> Stay informed about: silly HTML 

Toby Inkster <usenet200611 DeleteThis @tobyinkster.co.uk> wrote:

>hug wrote:
>
>> Okay Toby; I won't argue the point, but I will ask you where this is
>> specified.
>
>ISO 8879 (Standard for SGML), which is normatively referred to by the HTML
>4.01 spec.

Huh.

I found RFC 3986, "Uniform Resource Identifier (URI): Generic Syntax".

It says in section 2.2 that the ampersand character is a reserved
character of the subset "sub-delims", and goes on to say "characters
in the reserved set are protected from normalization and are therefore
safe to be used by scheme-specific and producer-specific algorithms
for delimiting data subcomponents within a URI."

Based on this, it remains unclear to me why a "&" character within a
double-quoted URL should be translated into an html entity.

So what am I missing here?

--
Legacy browsers never heard of emerging standards.
(contact via http://www.ren-prod-inc.com/hug_soft)
 >> Stay informed about: silly HTML 

hug wrote:

> Cristina, I think between you and Toby it is clear enough that you two
> think & should be used inside a quoted string. Certainly
> ampersands should be coded as & within html documents, but it's
> within quoted strings that I'm asking about.

It seems that HTML rules still apply
even inside the URL quoted string of an a tag,
you are still using HTML mark-up.
If the & character confuses the W3C validator, then
it is very possible that it will confuse other bots as well,
better write it as &

If something is flagged as an error by the W3C validator,
and you can fix it, then why not fix it.
If your web pages show validated OK by the W3C validator,
at least it shows to the visitors of your site that you
took the trouble to present to them pages that
as far as you know are correct HTML.
 >> Stay informed about: silly HTML 

"cristina" <cristina.DeleteThis@asymptoticdesign.com> wrote:

>hug wrote:
>
>> Cristina, I think between you and Toby it is clear enough that you two
>> think & should be used inside a quoted string. Certainly
>> ampersands should be coded as & within html documents, but it's
>> within quoted strings that I'm asking about.
>
>It seems that HTML rules still apply
>even inside the URL quoted string of an a tag,
>you are still using HTML mark-up.

Please see my post to Toby; I remain unconvinced because I have read
the URI Syntax RFC and it appears to agree with my point of view.

>If the & character confuses the W3C validator, then
>it is very possible that it will confuse other bots as well,
>better write it as &

<shrug> Bots are less important to me than customers, I have a simple
priority set, (1) make the functionality solid, (2) make it render as
intended on all current browsers including legacy browsers still
commonly in use, (3) pretty-up stuff like validation issues. Actually
it's more complex than that with continuing development coming ahead
of the pretty-up stuff, and other things too, but the basic idea is
there.

>If something is flagged as an error by the W3C validator,
>and you can fix it, then why not fix it.

I understand that point of view, but frankly I am not going to risk
incorrect rendering on some browsers because the W3C validator throws
a hissy about what I believe to be correct syntax. Sure, I'll fix
things like missing tags etc, but I'm not jumping through hoops for
validation -or- for bots unless the hoop-jumping is actually worth my
development time.

>If your web pages show validated OK by the W3C validator,
>at least it shows to the visitors of your site that you
>took the trouble to present to them pages that
>as far as you know are correct HTML.

Well, it shows a few visiting webmasters that I jumped through the W3C
validator's hoops. It's vastly more important to me that the pages
render properly on the browsers currently in use than it is to impress
a few webmasters with my conformace to the standard-de-jour.

It's not like it's a big deal for me to change it, all I have to do is
insert a single line of PHP code and it's done; but until/unless I
know that will improve things without breaking anything, the validator
can go fish. I tested it this morning, both ways work on one browser
(Opera 9.02) but restored it to its previous operation; I'm not going
through a full regression suite for something that I personally
consider to be erroneous to begin with.

--
Legacy browsers never heard of emerging standards.
(contact via http://www.ren-prod-inc.com/hug_soft)
 >> Stay informed about: silly HTML 

hug wrote:

> "cristina" <cristina.RemoveThis@asymptoticdesign.com> wrote:
> Please see my post to Toby; I remain unconvinced because I have read
> the URI Syntax RFC and it appears to agree with my point of view.

I still think that the W3C validator is correct
about writing & as & when you want to present to
the user agent (browser or bot) a text & character,
and not encode a character (like © for example).
You are still using HTML mark-up inside the a tag,
even within the URL double quotes,
and this is the case for browsers as well as bots
(not to mention if you are using XHTML, and even you agreed with
me that in XHTML documents it is an XML syntax error)

Why don't you give an URL?
It will not change the way the & character is dealt with,
but it will be easier to discuss on a specific URL.

Thank you for thanking me in a previous posting.
 >> Stay informed about: silly HTML 

hug wrote:
> I found RFC 3986, "Uniform Resource Identifier (URI): Generic Syntax".

That's the syntax of URIs themselves, and has nothing to say about what
further transformations may be necessary in order to embed them within
the syntax of a language such as HTML, SGML, XML, or XHTML.

> It says in section 2.2 that the ampersand character is a reserved
> character of the subset "sub-delims", and goes on to say "characters
> in the reserved set are protected from normalization and are therefore
> safe to be used by scheme-specific and producer-specific algorithms
> for delimiting data subcomponents within a URI."

The character is OK for use (in the proper context) within a URI, but
the entire contents of the URI still needs to be properly encoded when
it is placed into an HTML document.

> Based on this, it remains unclear to me why a "&" character within a
> double-quoted URL should be translated into an html entity.

Because it's in an HTML document.

> So what am I missing here?

The HTML specs. The URI specs only specify what a valid URI is, not
whether it needs special encoding to be included or transmitted in some
particular context.

--
Dan
 >> Stay informed about: silly HTML 

William Tasso wrote:
> Fleeing from the madness of the . jungle
> hug <contact_info.TakeThisOut@sig_line.clickit> stumbled into news:alt.www.webmaster
> and said:
>
> > ...
> > the validator can go fish ...
>
> Well I heard it was clever, must also have the patience of a saint.

Give a validator a fish, and it will be fishy for a day. Teach a
validator to fish, and it will be fishy for a lifetime!

--
Dan
 >> Stay informed about: silly HTML