W3C home > Mailing lists > Public > www-archive@w3.org > August 2002

Re: log:uri and relative URIs

From: Tim Berners-Lee <timbl@w3.org>
Date: Tue, 13 Aug 2002 22:35:11 -0400
Cc: www-archive@w3.org
To: Dan Connolly <connolly@w3.org>
Message-Id: <74C0CA43-AF2E-11D6-A399-000393914268@w3.org>

Yes, I agree with you, that though it would in practice
be simpler to say

"1".os:argv^log:uri.log:semantics

rather than

"1".os:argv.os:baseRelative^log:uri.log:semantics

it is best to keep things crisp at the moment, so people
are clean in their thinking.

Tim

On Tuesday, August 13, 2002, at 12:07 AM, Dan Connolly wrote:

> reviewing recent changes, I see in
> the log:uri built-in:
>
>
> +       Note that relative URIs can be OK as the whole process
> +       has a base, which may be irrelevant.
>
> That's taken care of elsewhere in the code, no?
> i.e. a base URI is constructed on input, and
> then subtracted out on output.
>
> If I just write
> 	this log:forAll :X, :Y, :TXT.
> 	:something rdfs:label "../foo".
> 	{ :X rdfs:label :TXT.
>           :Y log:uri :TXT } log:implies { ... }
>
> in an N3 document, that shouldn't bind :Y,
> should it? the ../foo label for something
> could come from a different file. strings
> don't carry their base URI around with them.
>
> I guess I should write a test. But
> this seems plain to me.
>
> --
> Dan Connolly, W3C http://www.w3.org/People/Connolly/
>
>
Received on Tuesday, 13 August 2002 22:35:13 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 7 November 2012 14:17:22 GMT