Re: [ANN] Beta test for the Markup Validator - 0.6.5beta3

Hi Frank, all.

Thanks for all these suggestions, thoughts and remarks.
My thoughts on these, inline.

On May 2, 2004, at 06:22, Frank Ellermann wrote:
> You might be able to add a lot of similar character encodings,
> it's like us-ascii + 8-bit characters (80h - FFh).

One constraint for supporting charsets is that they should be supported 
by iconv. But indeed, we may be able to expand our list of supported 
charsets.

> Are you absolutely sure that you want to mention bugzilla on
> the feedback page ?

With the current wording and the fact that bugzilla requires a (simple) 
registration, I guess it's fine. What trouble can you think of?

>   A news interface to this list on gmane.org
> is nice, maybe add it on the feedback page

Hmm, haven't been on usenet for ages I must confess. I thought gmane 
was mostly a read-only thing, was I wrong?

> Other observations:
> - A source listing without explicitly wanting this is a major
>   pain for modem users.  I have the source, I sent it, I don't
>   want the echo.

This is an observation many have made. Echo-ing the source when the 
error message pointed to the whole source made sense, now that source 
snippets are used, it may not make a lot of sense. Any advocate of 
keeping the current behaviour?


> - The "DOCTYPE override" info _can_ appear twice in the output
>   (not always, maybe only if there is a conflict (?))

Do you have a test case for that? Never seen such a behaviour.

> - There's no link to <http://validator.w3.org/images/vh20>
>   for a validated strict HTML 2.0 page

Need to look into that

> - the error message...
> | Sorry, I am unable to validate this document because its
> | content type is http://xyzzy.dnsalias.org/, which is not
> | currently supported by this service.
>   ...does not exactly match "start your httpd, stupid" ;-)

AFAIK "start your httpd, stupid" is usually "Can't connect to 
xyzzy.dnsalias.org:80 (Timeout)". test case for the (strange) message 
you're quoting?

> - the error message...
> | No Character Encoding Found! Falling back to UTF-8.
>   ...makes no sense for HTML 2.0 (the default is Latin-1).

The current mechanism is not perfect, but without knowing the encoding, 
finding the doctype is a matter of garbage-in garbage-out, and it's 
therefore not logical to adapt the mechanism in case of absence of 
charset to an hypothetical detection of the doctype.

> - the "skip navigation" and "results" links don't work with
>   legacy browsers.

Indeed, fixed (in CVS).

> - IMHO "skip navigation" and "results" qualify as "jump to",
>   but ",text" and ",tablin" are different documents.

Agreed. Suggested fix?

> - "help and FAQ" <http://jiggles.w3.org/servlet/docs/help.html>
>   is apparently broken.

Weird, where did you find that link?

> - I wanted to find an explanation for "parse tree", "no atrri-
>   butes", "validate error pages" (maybe you could link these
>   options to the relevant section of "help and FAQ", if that's
>   a compatible document with name= fragments - otherwise forget
>   it, no link is better than a broken link).

This is now done in CVS.

-- 
olivier