I think that the 'architectural' answer to that question is that it
depends on what is being identified! :-)
If you are identifying 'an MP3 file', then it is OK to assign a URI with
the .mp3 suffix. If you are identifying 'an audio resource', then you
probably want to assign a non-technology-specific URI.
In general, you may want to use both.
Here's an example. We recently published a report about Second Life in
PDF format. In the future we might choose to publish it in other
formats (HTML, Word, whatever) but so far we have only published in as
PDF. We have assigned both a generic URI for the report
http://www.eduserv.org.uk/foundation/sl/uksnapshot072007
and a specific URI for the PDF version
http://www.eduserv.org.uk/upload/foundation/sl/uksnapshot072007/final.pd
f
and linked from one to the other. I guess this is pretty standard
practice?
However, one of the issues with this approach is which version people
cite (link to). We would prefer them to bookmark and link to the
generic resource - but in practice, they typically bookmark and link to
the PDF version. This potentially has a negative impact on Google-juice
and so on.
An alternative approach is to assign a single non-technology-specific
URI and then use HTTP content negotiation to serve the correct format -
but this is non-trivial to set up.
Andy
--
Head of Development, Eduserv Foundation
http://www.eduserv.org.uk/foundation/
http://efoundations.typepad.com/
[log in to unmask]
+44 (0)1225 474319
> -----Original Message-----
> From: Managing an institutional web site
> [mailto:[log in to unmask]] On Behalf Of K Fearon
> Sent: 26 July 2007 12:43
> To: [log in to unmask]
> Subject: Re: Value of file extensions in URLs
>
> Would this include all filename extensions or just server
> technology? I use .pdf, .mp3, etc in the url to judge whether
> or not to download large files or file types my home machine
> can't handle.
>
> Kriss
>
> --------------------------------------------------------------
> -----------
> Web Coordinator HB30 Tel: (01904) 434682
> Fax: 434685
> University of York, UK 9-5.15, Mon-Fri
> http://www.york.ac.uk/weboffice/
>
> York welcomes the BA Festival of Science 2007, 9-15 September
> www.york.ac.uk/ba
>
> On Thu, 26 Jul 2007, Jim Higson wrote:
>
> > Trying to open up a debate: Recently I've been considering
> whether we
> > should drop file extensions from URLs on our web site. For example:
> >
> > www.aber.ac.uk/foo/bar.php
> >
> > would become just:
> >
> > www.aber.ac.uk/foo/bar
> >
> > Ignoring for now how this would be implemented, do you
> believe having
> > the ".php" on the end of the file gives any useful
> information to the
> > user?
> >
> > My argument for dropping the ".php" is the user doesn't
> really care if
> > we are serving from a PHP file (and if they do care, they probably
> > know how to look in the http headers). By using the
> extension in the
> > URL we are exposing an implementation detail that isn't
> really relevant.
> >
> > However, maybe users familiar with the Windows filesystem
> look at URLs
> > with an extension as a "file page" and URLs without as a "folder
> > page", thereby using the URL to discern a page's place in the
> > hierarchy. The majority of institutional web sites have the
> extension
> > in the URL - is this by design, or a side effect of implementation?
> >
> > Do you think this distinction between "folders" and "files" really
> > exists in user's minds on the web, or are there just pages
> that form a
> > tree-like hierarchy delimited by slashes?
> >
> > Do you think hiding implementation details like which programming
> > language we are using really makes things simpler for end
> users, or do
> > they not really care about the URLs anyway?
> >
> >
>
|