Archives for: October 2006

Oct 19, 2006 : Wikipedia!

I can't be held responsible for any time wasting that may ensue.

Lunch talk today covered various topics stemming from Wikipedia articles. Here's some good ones:

  1. 1,000,000,000; the largest number in wikipedia
  2. List of wikipedia's numbers
  3. Transnistria a region of Moldova that has claimed independence from Moldova. It's independence is not recognized by the UN, but it is by Moldova. See also Unrecognized states with de facto control over their territory
  4. Lesser known Generic TLDs such as .cat (for Catalan language & culture sites) and .int (for international organizations)
Technorati tags:

Oct 08, 2006 : 3 years

Cherith & I have been dating for three years now. We've come a long way :)

Oct 06, 2006 : URI structure

Non-nerdy abstract: some folks like to do all sorts of crazy things with what you simply know as a URL or website address. We abstract them to be names with no content, or internationalize them, etc, etc and call them URIs, IRIs, XRIs, etc. This is about the semantic structure of a URI.

A while ago there was some discussion started by Norm Walsh on the use of URIs using schemes other than http: . One direction the discussion took was that specifications for other schemes provided a human-readable description of how the URI was to be constructed. For example, the URN:ISBN specification has a "Declaration of syntactic structure" explaining the four parts it contains. Using the http: scheme for all our URIs means (at best) we'd have non-normative descriptions of the URI structure.

Today, however, I came across Marc Hadley's URI Templating post, which includes a link to a draft specification for URI Templates. This would mean machine-readable URI structure declarations (better than human-readable spec documents, on many levels). I think it's a great start.

The trouble still lies in how to discover a URI Template, given a URI. For URIs used in web apps, a WADL document you already have could reference the template. And I bet WSDL 2.0 can do something with URI syntax, since it supports descriptions of REST APIs. But URIs can be used in so many other contexts (even offline) where discovery would be probably impossible. URI Templates may be provided by a 3rd party, so even having the template located at some special URL (e.g. robots.txt) or referenced in a URL document (e.g. favicon) would not necessarily be ideal.

Technorati tags:

<< Previous Page

<< Previous Page

<  October 2006  >
Mon Tue Wed Thu Fri Sat Sun
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
30 31          




Syndicate this blog XML

powered by