Re: [VE][325] Error Message Feedback

> > However, i think it would be better if the validator "ignores"
> > "&"-characters in URLs at all.
>
> Why? Just becuase lots of people make an error, and that most user
> agents can recover from it, doesn't mean that the documents shouldn't
> be fixed (and fixing it /is/ trivial).

Why does the use of plain & in URLs need to be fixed? Can you give me one or 
more solid reasons why a browser should change an & that it knows is inside 
an URL? The browser is parsing the HTML. It knows it is inside the quoted 
part of the hypertext reference inside the <a> tag, or inside the action 
reference isde a <form> tag. It knows that a plain & is valid as an argument 
separator in an URL. And every browser I've ever used correctly handles URLs 
with such separators. An URL written into HTML code as part of an HTML 
element is not an URL 'encoded in HTML'.

An URL 'encoded in HTML', to use W3's own words:
  &lt;a href="http://www.w3.org/validator.php?arg1=this&amp;arg2=that"&gt;

An URL written into HTML code as part of an element:
  <a href="http://www.w3.org/validator.php?arg1=this&arg2=that">

One should reasonably expect that the former is intended to be rendered for 
display rather than being used as an URL. And one should reasonably expect 
that the page's author expects the URL to be used as she wrote it rather than 
being re-written by the browser before it can be used.

I expect the usual pedagogic "Because we wrote the spec, we know what is best, 
and we do not make mistakes" response and will be really surprised to receive 
anything else. I wouldn't be at all surprised to learn that the reason for W3 
eliminating most default HTML values is to ensure that the average person 
becomes incapable of writing her own HTML; after all, such wide-spread 
communication really should be reserved to the elite members of society, like 
W3.

I just won't your validator any more. It insists that such bare &s cause 
unrecoverable ERRORS in browsers, but your validator can't even be bothered 
to tell me that it implicitly closed an open tag, thus leaving me to spend an 
hour trying to figure out what it is complaining about. IT's like software I 
expect from Microsoft.

Neal Murphy

Received on Monday, 27 November 2006 05:13:34 UTC