W3C home > Mailing lists > Public > public-lod@w3.org > July 2009

Re: cool urls and redirects

From: Alan Ruttenberg <alanruttenberg@gmail.com>
Date: Sat, 25 Jul 2009 09:43:42 -0400
Message-ID: <29af5e2d0907250643l3b5e8740h55b201d62a9edce6@mail.gmail.com>
To: Eric Lease Morgan <eric_morgan@infomotions.com>
Cc: public-lod@w3.org
On Fri, Jul 24, 2009 at 9:34 AM, Eric Lease
Morgan<eric_morgan@infomotions.com> wrote:
> To what degree is it a best practice to implement "cool" URLs as redirects?

The essential thing is consistency. It is common practice to use
redirects from URIs that name resources. Often this is done so that it
will be possible, as is proposed in the Shared Name project, to move
hosting to a different server without having to notify people who have
made statements about the resource (or bookmarked them) and have them
adjust their content. In you case you are using a redirect from the
same domain, but you might consider using PURLs for you public facing
URLs.

If you do so, be consistent - make all RDF statements use the
preferred URLs, as well as all links in the HTML. For a dose of good
measure you might want to use rel=canonical in your html.

http://en.wikipedia.org/wiki/Functional_Requirements_for_Bibliographic_Records
http://googlewebmastercentral.blogspot.com/2009/02/specify-your-canonical.html
http://sharedname.org/
http://purl.org/

-Alan

>
> I manage a small database of electronic texts (approximately 20,000 items).
> Because the content is in a database I can create reams of reports against
> it. Browsable lists by title, author, and keyword/subject. I can index the
> lot with Solr/Lucene. I can create various versions of the content: 1) plain
> text, 2) simple HTML, 3) HTML complete with a handy-dandy floating palette
> supporting various services against the texts, 4) PDF. Etc. About 18 months
> ago I started creating RDF files representing the texts. For example, Thomas
> More's Utopia. [1]
>
> My next self-assigned project is to present my small database as a set of
> linked data. To this end I have written a mod_perl module that dereferences
> a URI. For example [2], given a key, the module negotiates content, returns
> a 303, and a URL to RDF or HTML representations.
>
> Now, my question is, should the URLs I return be the long ugly ones pointing
> to actual files on my file system, or can they be "cool" URLs that are
> actually redirects? In other words, instead of returning something like
> this:
>
>  http://infomotions.com/etexts/literature/english/1500-1599/more-utopia-221.rdf
>
> is it acceptable to return something like this and have it redirected to the
> URL above:
>
>  http://infomotions.com/etexts/rdf/more-utopia-221
>
> Similarly, when returning URLs pointing to an HTML representation, should I
> return A or something more like B:
>
>  A.
> http://infomotions.com/etexts/literature/english/1500-1599/more-utopia-221.htm
>  B. http://infomotions.com/etexts/html/more-utopia-221
>
> Once I figure out the answers to these questions, then I can start cleaning
> up my RDF files and including the "best" URIs in them.
>
>
> [1] Utopia -
> http://infomotions.com/etexts/literature/english/1500-1599/more-utopia-221.rdf
> [2] dereference - http://infomotions.com/etexts/id/more-utopia-221
>
> --
> Eric Lease Morgan
> Infomotions, Inc.
>
>
>
>
Received on Saturday, 25 July 2009 13:44:43 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 16:20:51 UTC