W3C home > Mailing lists > Public > www-validator@w3.org > April 2006

[VE][127] Error Message Feedback - Please be less vague and irrelevant

From: Matthew Paul Thomas <mpt@myrealbox.com>
Date: Sun, 30 Apr 2006 15:52:01 +1200
Message-Id: <f54cdb3baf756dca66e37016f68b41d9@myrealbox.com>
To: www-validator@w3.org

The W3C Validator is much more helpful than it was five years ago, and 
it's great that it can now validate HTTPS URLs. However, some details 
are still aggravating. For example, in a page of validation results I 
received this error:

     Error Line 147 column 29: required attribute "alt" not
     specified.
         <img src="/@@/user.gif">

     The attribute given above is required for an element that you've
     used, but you have omitted it.

The validator knows very well that "The attribute given above" is the 
alt attribute, and it knows very well that "an element" is the img 
element, so it needn't be so vague. What the validator *doesn't* know 
is whether it's "an element that you've used" or whether "you have 
omitted it": I might be validating someone else's page, or dealing with 
an obnoxious attribute-stripping CMS. Correct all those errors, and 
you'd end up with something like "The alt attribute is missing from 
this img element", but that would merely be a more helpful rewording of 
the error itself. Conclusion: reword the error message to include the 
element name, then drop the first sentence from the hint.

Unfortunately the rest of the hint is even worse:

                                    For instance, in most HTML and
     XHTML document types the "type" attribute is required on the
     "script" element and the "alt" attribute is required for the
     "img" element.

     Typical values for type are type="text/css" for <style> and
     type="text/javascript" for <script>.

The validator knows very well that the problem is with a missing alt= 
attribute, not with an incorrect type= attribute, so chatting away 
about "typical values for type" does little except make the validator 
seem buggy. It would be great to offer attribute-specific suggestions 
for this error, but until that is implemented, it would be better to 
stay silent than to deliberately discuss irrelevant things.

Cheers
-- 
Matthew Paul Thomas
http://mpt.net.nz/
Received on Sunday, 30 April 2006 12:50:45 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 25 April 2012 12:14:22 GMT