Re: Public beta test of the W3C Markup Validator

At 12:07 02/10/25 +0200, Chris Lilley wrote:
>On Friday, October 25, 2002, 10:17:32 AM, Martin wrote:

> >>One workaround would be to look at the filename of the uploaded file
> >>and then treat this as the validator server would treat such a named
> >>file if it were serving it ....
>
>MD> I would be willing to close an eye or two for stuff sent in,
>MD> and e.g. for file upload ignore the strong us-ascii default or
>MD> so (which as far as I understand would mean that even with text/xml,
>MD> the file would then be validated using the DTD it gives,...), but
>MD> I'd rather not go into the business of sniffing on the server side.
>
>Huh?
>
>In what way is this 'sniffing' - rather, it is to avoid using the
>dubious results of the browser sniffing.

Sniffing on content and sniffing on extensions are not that different.
What I'm saying is that if we want to avoid sniffing, then we should
do that altogether, not replace one way with another way that looks
better for the moment.


>And bythe way, the need to 'turn a blind eye' when doing file
>transfers and server-less processing is merely one example why the
>*brain dead stupid* rule of forcing to ascii an XML file that has a
>perfectly good encoding declaration right there in the file is
>*actively harmful* to interoperability, reliable processing, and the
>use of languages other than English on the Web.

Mileages vary on this point. If you want this to be changed,
you need to rewrite RFC 3023 and convince the IETF that the
new version is better. The strict ascii default for text/foo+xml
was a rather explicit requirement from the IETF.

Anyway, we have a tradition on the validator list to leave
discussion of the specifications themselves to other lists.
So let's concentrate on how to make the validator work best
within the constraints that we have.

Regards,   Martin.

Received on Friday, 25 October 2002 08:25:08 UTC