Category Archives: Web

links for 2007-01-26

Everybody will be their own information architect

Tim O’Reilly has a post about short-form content. Nick Carr commented it in his blog.  One way to view it is it’s an application of the most fundamental technique in computer science: divide-and-conquer.  End users will be responsible for combing them, based on their own preference and interest. Everybody will be their own information architect! 

links for 2007-01-23

Problems with RDF

RDF, a W3C standard which is supposed to be the foundation for semantic web, is not obtaining universal acceptance as HTML did.  Frederick Giasson analyzed some of reasons in Wikipedia concepts to help managing Semantic Web described subjects.  He thinks RDF “is so beautiful and powerful is also a huge problem and probably the bigger weakness of the semantic web.”To me, granularity is another issue.  We already have XML.  If we use RDF and XML to define the concepts in RSS and OPML, people will find XML is much simpler.  RDF will give us some benefits but it’s not a big deal in these cases. Anyway, at this time, people need standard or agreement.  RDF might be more useful if people want to exchange information dynamically.

From data structure point of view, it’s hard for people to grasp graph.  Tree is a kind of ok.  List is much easier.


We’ll see non-programmers will be able to code.  Structured data is the first step towards this direction.  Surely we need a programming model for non-programmers.  So far, all programming models: procedual, OO, SQL, Lisp, scripting etc are not designed for non-programmers.

Some quotes:

“The next step is user created apps and user built automation” – Rod Boothby, “The Executable Web” 

“Hyperlinks connect the pages of the web. Automation will connect the applications of the web.” – John McMahon, CEO of Extentech

Also see: Read, Write, Execute

Yahoo vs. Google and Directory vs. Search

While Yahoo moved away from Directory oriented, Google has a Directory now. 🙂

Programming on the web: copy and past

Copy and past is the easiest way of “programming”.  In many cases, you can specify you requirements and get the code. Then copy and past to your blog’s template.  For example, MuseStorm allows you to create a RSS widget by just inputing RSS URL.   See more on A Widget Storm is Coming.

Web 1.0, 2.0 and 3.0

Web 1.0: for the user!
Web 2.0: for the user, by the user!
Web 3.0: for the user, by the user, of the user!

See more NUWeb: A project for Web 3.0

Programming Model

As we observed, web is becoming more and more programmable.  See number of APIs available today on

However, to my knowledge, we don’t have a universally accepted programming model for writing programs on the web.  Mash-up, Copy-Past-Modify and Drag and Drop are techniques rather than programming models.  OO (Object Oriented) is the most popular programming model.   But I don’t see it’s quite fit for programming on the web, even for programmers.  Surely it’s very challenging for people without computer sceince backgroud to learn OO programming.

We’ll see which model will emerge.  Maybe AI (Artifical Intelliegnce) research will help us in this case.

MyBlogLog: Know Your Visitors

Now MyBlogLog is acquired by Yahoo.  One of the most impressive features of MyBlogLog is to allow you see who visited your blog, if you and your visitors are member of MyBlogLog.  It reflects a trend of the web: many people want to identify themselves!