W3C home > Mailing lists > Public > www-validator@w3.org > February 2002

Re: HTML validator and WebServices

From: Nick Kew <nick@webthing.com>
Date: Sat, 16 Feb 2002 14:24:46 +0000 (GMT)
To: Karl Dubost <karl@w3.org>
cc: <www-validator@w3.org>
Message-ID: <20020216135422.H15289-100000@fenris.webthing.com>
On Sat, 16 Feb 2002, Karl Dubost wrote:

> I'm sure it's already on tracks, but it would be very helpful for the
> community to hack a webservice for the HTML validator.
> A kind of ping HTML validator like the URL-minder but in a sense more
> immediate. For example, now, people must submit the URL in a form to
> verify if the site is valid.

This is one of the dividends of XML-based reporting.  Site Valet is
offering this service to anyone who wants it.  I'll also be offering
the WCAG and Section508 summary assessments (i.e. a single assessment
Pass/Probable Pass/Uncertain/Probable Fail/Fail).

A thought that has occurred to me is that these summary reports
- or indeed full-detail reports - could also in principle be fed
to the W3C annotation server.  Valid RDF is already supported via
the EARL option.

> The (we)blogger tools have implemented this ping [3]. So the user do
> not have to bother about going to a website to say "Hey, there's new
> dog food on my website".

I don't quite follow.  Are you suggesting the weblogs.com server will
run a validation service by "ping"ing a validator?  I can't see how
their existing interface would permit a user to automate this.

> The idea:
> ---------
> Having the same kind of service for the HTML validator and encourage
> people who are creating authoring tool to use the service and
> integrate it in their products. When someone edit with such a tool,
> he/she will have the possibility to ping for his new content, but
> also to know if his/her page is still valid at the submission time.
> There's no need to have the detail reports of HTML errors, but just
> something very simple which returns.
> 	"valid" or "not valid"
> as parameters when you send the URI of the page.

You need at least one more condition, which might be "can't tell" (e.g.
if the charset is unsupported by iconv).  There are a couple of other
conditions that lead Page Valet to decline to say Yes or No.

As a point of order, XMLRPC *is* a reinvention of a wheel that goes back
to RFC822, and only really has merit when the requests are a great deal
more complex than we are dealing with.

Nick Kew

Site Valet - the mark of Quality on the Web.
Received on Saturday, 16 February 2002 09:25:14 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 1 March 2016 14:17:32 UTC