On Thu, January 12, 2012 2:51 pm, raj kumar wrote:

> I don't understand why. A url that ends with a slash just redirects to the
> same url without a slash on the end:
>
>> curl -v http://openlibrary.org/books/OL18215289M/
> < Location: http://openlibrary.org/books/OL18215289M

But Ben was talking about URIs, not URLs. A URI is the name of an object, a
URL is the name of a place. All places are objects, so all URLs are also URIs,
but not all objects are places, so not all URIs are URLs. A URI may look like
a URL, but it isn't. The notion of using the same syntax for URIs and URLs is
one of the stupidest ideas ever envisioned, because (among other things) it
leads the the very kind of confusion you have fallen prey to.

Should all URIs end with a slash? I can't see that it matters. It's like
saying all names should end with a period. A name is a name, and a different
name is a different name.

In practice, is there software which consumes RDF and which consider trailing
slashes significant? I certainly don't know. But if there is, then some
standardization or naming convention is certainly necessary.

> When you request the url without a slash, you get a redirect to the canonical
> url for the edition:

This is perhaps true for open "library" but it is not generally true. Behavior
of a web server when it gets a URL without a slash is determined by the web
server configuration. It may be common, but it is not universal, and is not
governed by any standard.


_______________________________________________
Ol-tech mailing list
[email protected]
http://mail.archive.org/cgi-bin/mailman/listinfo/ol-tech
To unsubscribe from this mailing list, send email to 
[email protected]

Reply via email to