David Håsäther | 1 Aug 10:21 2005
Picon

Re: This page is not Valid ! - no results


On 2005-07-31 14:23, Nigel Ellis wrote:

> The validator then displays the message:
> 
> Below are the results of attempting to parse this document with an SGML 
> parser.
> 
> The only problem is there are no results. The page I am trying to 
> validate is 'http://www.mentalis.co.uk/dwid4/9/tv/2_test.htm'.

Your doctype declaration is erroneous. Use

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">

and then start fixing the errors :-)

--

-- 
David Håsäther

Risto Kaartinen | 3 Aug 05:44 2005

[VE][73] Error Message Feedback

I just wanted to know what is wrong whit this kind of start:
 
 
 
<!DOCTYPE HTML SYSTEM>
<head>
 <meta http-equiv="content-type" content="text/html; charset=iso-8859-1">
        <meta name="resource-type" content="document">
        <meta name="robots" content="index, follow">
        <meta name="keywords" content="Moduuli-Net Moduuli-Networks Garrett Risto Kaartinen">
        <meta name="distribution" content="global">
        <meta name="rating" content="general">
        <meta name="copyright" content="Risto Kaartinen">
        <meta name="author" content="Risto Kaartinen">
        <meta name="language" content="finnish">
        <meta name="doc-type" content="public">
        <meta name="doc-class" content="completed">
        <meta name="doc-rights" content="copywritten work">
 
 
<link href="tyyli.css" rel="stylesheet" type="text/css">
 <link rel="shortcut icon" href="http://moduuli.ma.cx/favicon.ico">
</head>
 
 
 
Validator says:
 

Line 22, column 6: end tag for "HEAD" which is not finished

</head>

Most likely, You nested tags and closed them in the wrong order. For example <p><em>...</p> is not acceptable, as <em> must be closed before <p>. Acceptable nesting is: <p><em>...</em></p>

Another possibility is that you used an element (e.g. 'ul') which requires a child element (e.g. 'li') that you did not include. Hence the parent element is "not finished", not complete.

 

And i dont understand what is wrong with that..? Thank you for helping! (only if you answer :)

Olivier Thereaux | 3 Aug 08:21 2005
Picon

Re: [VE][73] Error Message Feedback


Hello Risto,

On Wed, Aug 03, 2005, Risto Kaartinen wrote:
> I just wanted to know what is wrong whit this kind of start:

As the validator says: [[ Another possibility is that you used an
element (e.g. 'ul') which requires a child element (e.g. 'li') that
you did not include. Hence the parent element is "not finished", not
complete. ]]

Your <head> is missing a <title>.

Also, it would probably be a good idea to use a different Doctype
declaration than:
> <!DOCTYPE HTML SYSTEM>
See http://www.w3.org/QA/2002/04/valid-dtd-list.html for examples of
recommended doctype declarations.

Regards,
--

-- 
olivier

Jukka K. Korpela | 3 Aug 09:49 2005
Picon
Picon

Re: [VE][73] Error Message Feedback


On Wed, 3 Aug 2005, Olivier Thereaux wrote:

> Your <head> is missing a <title>.

We have often seen suggestions that this common error should be reported 
more directly - especially since it is a practically significant error
(see e.g. "<title>: the most important element of a quality Web page", 
http://www.w3.org/QA/Tips/good-titles ).

Unfortunately it seems that there is no way to achieve that without major 
changes in the validator's structure. At present it can only signal that a 
required subelement is missing, not what subelement is missing.

As a practical method, many people have adopted the habit of writing a 
<title> element right at the start of the <head> element. This may help 
against accidental omission of <title>, and it is a fairly logical 
order, too. Technically, the mutual order of subelements of <head>
is free; so you can choose an order that looks good to you.

But there's a catch. Sometimes the character encoding of a document can 
only be deduced (by a browser) from the information in a <meta> tag.
This is unfortunate, but for such situations, it is probably best to write 
an element like
<meta http-equiv="Content-Type" content="text/html;charset=iso-8859-1">
as the first subelement of <meta>, and the <title> element next. The 
reason is that your <title> element might, now or in a revised version, 
contain non-ASCII characters, so things work better if the browser knows 
the encoding at that point.

--

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

Marc "TROJjER" Kirkwood | 4 Aug 00:23 2005
Picon

Re: [VE][107] New Error Message Suggestion

I've been wondering about this...
 
attributeName="attributeName" would be annoying for me to have to repeatedly type out, and it just looks silly -- am I the only person to notice this?? Can't you just use something like checked="1" or checked="true"? This seems better to me; I hate the idea of unnecessary repetition, yet this way, the attribute still won't be empty -- just a bit shorter.
 
I hope it's not in the specification that you have to repeat the name of the attribute -- I see no reason why you should.
David Dorward | 4 Aug 00:30 2005
Picon

Re: [VE][107] New Error Message Suggestion


On Wed, Aug 03, 2005 at 11:23:32PM +0100, Marc TROJjER Kirkwood wrote:

>    attributeName="attributeName" would be annoying for me to have to
>    repeatedly type out, and it just looks silly

Do you /have/ to use XHTML?

>     -- am I the only person to notice this?? Can't you just use
>    something like checked="1" or checked="true"?

In HTML "checked" is not the attribute name, its the value. You write
'value' instead of 'name=value'. If XHTML 1.0 was designed from
scratch rather then being a port of HTML from SGML to XML, then it
might have been designed differently. The design makes sense when
taken in the context of the SGML features that were being used when it
was designed.

>    This seems better to me; I hate the idea of unnecessary
>    repetition, yet this way, the attribute still won't be empty --
>    just a bit shorter.  I hope it's not in the specification that
>    you have to repeat the name of the attribute

It is:

<!ATTLIST option
  %attrs;
  selected    (selected)     #IMPLIED

> -- I see no reason why you should.

Its legacy. See above.

--

-- 
David Dorward                                      http://dorward.me.uk

Rick F Martin | 3 Aug 21:46 2005
Picon

[VE][65] Error Message Feedback


Would be nice to supply info on to turn off OMITTAG NO.
Have been googling around and ca't figure it out.
Have a bunch of generated HTML files that are W3C compliant except for  <BR>s. Instead of editing them all to be </BR> would like to set OMITTAG YES.
Rick F. Martin
Web Operations Manager, Web Shared Service Center, PMS IT
Tel: 978 659-4743
E-Mail: rick.f.martin <at> philips.com
Olivier Thereaux | 4 Aug 06:05 2005
Picon

Re: [VE][65] Error Message Feedback


Hello Rick,

Thanks for your message.

On Wed, Aug 03, 2005, Rick F Martin wrote:
> Would be nice to supply info on to turn off OMITTAG NO. 

I think there is some kind of misundestanding here. Validation is
comparing a document against a formal grammar, in our case HTML or XML
DTDs. OMITTAG is just one rule in the DTDs, in other words it is part of
how the language you are validating against is constructed.

So there is no "turning off OMITTAG NO", unless you want to create
another (proprietary) language, and create a DTD for it that. I doubt
this is actually what you are trying to achieve.

> Have a bunch of generated HTML files that are W3C compliant 

strictly speaking, there is no such thing as "W3C compliant". I assume
you mean "valid against one of the W3C markup languages", although it's
not sure which. 

> except for <BR>s. Instead of editing them all to be </BR> would like
> to set OMITTAG YES.

HTML 4.01 would allow you to use <BR>, as it has OMITTAG YES. For XHTML
document types, indeed, you must use <br /> (lower case...)

Hope this helps.
--

-- 
olivier

olivier Thereaux | 5 Aug 02:28 2005
Picon

Re: [ANN] Beta test of the W3C Markup Validator (0.7.0 beta 2)


On Jul 28, 2005, at 16:40, Olivier Thereaux wrote:
> We are starting today a new Beta test period for the W3C
> Markup Validator, version 0.7.0 (Beta 2).
>
>                http://validator.w3.org:8001/
> Note that this will be a short test period, mainly aiming to check
> changes made during and after the first beta test period (see below).
> Please send your feedback (see below for instructions) on this beta
> versions through Thursday, August 4th, 2005.

The beta test is now closed, and we will be releasing the validator 
v0.7.0 in the next few days. Of course, comments, suggestions and bug 
reports are always welcome, either on this list or on the bug tracking 
tool, any time, not just during beta tests.

Thanks,
--

-- 
olivier

Ville Skyttä | 5 Aug 11:49 2005
Picon
Picon

Re: [VE][73] Error Message Feedback


On Wed, 2005-08-03 at 10:49 +0300, Jukka K. Korpela wrote:
> On Wed, 3 Aug 2005, Olivier Thereaux wrote:
> 
> > Your <head> is missing a <title>.
> 
> We have often seen suggestions that this common error should be reported 
> more directly - especially since it is a practically significant error
> (see e.g. "<title>: the most important element of a quality Web page", 
> http://www.w3.org/QA/Tips/good-titles ).
> 
> Unfortunately it seems that there is no way to achieve that without major 
> changes in the validator's structure. At present it can only signal that a 
> required subelement is missing, not what subelement is missing.

Yep.  But this particular case is so frequently reported that the error
message has been updated in the upcoming release to say:

    Most likely, You nested tags and closed them in the wrong order. 
    For example <p><em>...</p> is not acceptable, as <em> must be closed
    before <p>. Acceptable nesting is: <p><em>...</em></p> 
    Another possibility is that you used an element which requires a
    child element that you did not include. Hence the parent element is
    "not finished", not complete. For instance, <head> generally
    requires a <title>, lists (ul, ol, dl) require list items (li, or
    dt, dd), and so on.


Gmane