W3C home > Mailing lists > Public > public-grddl-wg@w3.org > February 2007

Re: Some spec comments (to .xsl or not to .xsl?) (#issue-whichlangs, #issue-conformance-labels)

From: Dan Connolly <connolly@w3.org>
Date: Tue, 20 Feb 2007 15:30:39 -0600
Message-ID: <45DB687F.30001@w3.org>
To: Chimezie Ogbuji <ogbujic@bio.ri.ccf.org>
Cc: Danny Ayers <danny.ayers@gmail.com>, "McBride, Brian" <brian.mcbride@hp.com>, GRDDL Working Group <public-grddl-wg@w3.org>

Chimezie Ogbuji wrote:
> On Tue, 20 Feb 2007, Dan Connolly wrote:
>
>> I can't tell if you're serious or not. In case you are...
>
> I was..
>
>> The spec currently (1.226) says
>>
>> "A GRDDL-aware agent should... Find each transformation... Apply each 
>> transformation..."
>> so yes, it has a SHOULD-level obligation to apply the javascript; 
>> meanwhile, that bit of the spec
>> follows "Subject to ... local policy as expressed in its 
>> configuration ..." and it's pretty
>> clear to me that "I don't do javascript" is a reasonable policy, 
>> especially given
>> our position on #issue-whichlangs...
>
> Since the idea of a 'local policy' is pretty loose (theoretically, one 
> such policy could be to *not* apply *any* transforms at all) does it 
> render the conformance label moot - as it is currently written?
No.

Should means should; it means do it unless you have some good reason not 
to. If you
have a good reason to not apply any transformations, then don't. If you 
have a software
module that never applies any transformations and you claim it's a 
conforming GRDDL-aware agent,
I don't think whether that claim is true or not has much impact; the 
market will judge
it harshly in any case.

The conformance label is not so valuable to me that I would object to 
taking it back out...
... so long as somebody else did the editing this time. 1/2 ;-)

>   Consider that such a policy would result in 'no' GRDDL result(s) for 
> an IR which is *supposed* to have one or more.
Whether an IR has a GRDDL result is orthogonal to whether any given 
agent computes it.

>> "Developers of transformations should make available representations 
>> in widely-supported formats. XSLT version 1[XSLT1] 
>> <http://www.w3.org/2004/01/rdxh/spec#XSLT1> is the format most widely 
>> supported by GRDDL-aware agents as of this writing, though though 
>> XSLT2[XSLT2] <http://www.w3.org/2004/01/rdxh/spec#XSLT2> deployment 
>> is increasing. While technically Javascript, C, or virtually any 
>> other programming language may be used to express transformations for 
>> GRDDL, XSLT is specifically designed to express XML to XML 
>> transformations and has some good safety characteristics."
>> -- http://www.w3.org/2004/01/rdxh/spec#txforms
>
> This is a pretty clear health warning IMHO.  It is the conformance 
> mandate that confuses things, it seems, as they are applied within the 
> context of open-ended policies. 
Is your preference to take out the conformance label? I guess the "1. 
Find each transformation..." list
would go with it. I think that list is responsive to a number of 
comments that Dave Beckett has made,
and I'm reluctant to take it out.

Would you leave Jeremy's security stuff in? Or take it out?

-- 
Dan Connolly, W3C http://www.w3.org/People/Connolly/
Received on Tuesday, 20 February 2007 21:30:49 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:11:47 GMT