The Evolving web

I have been quite excited about all the buzz in the blogosphere about and I have spent several hours reading opinions about the future of the Internet. While there is a lot of hype, (and hope and excitement) one thing is certain : the web is evolving as a platform. It is getting easier for computers to understand and talk to other computers and this is causing our interaction with information to get vastly more functional and useful.

Web 1.0

In the good old days we had a pretty simple web. People wrote stuff and put them up on static webpages and other people read them. They sent each other emails and Instant Messages. Being able to access so much information and so many people with such ease was simply amazing. With search engines like Yahoo and Google, we (literally) had this information at our fingertips. Then came websites like Amazon, NY Times, BBC News, IMDB and Wikipedia, and suddenly the web was more database driven and much more dynamic. This allowed us access to news, information, entertainment and products that were more suited to our individual tastes and myriad needs. It gave us access to The Long Tail and made the web infinitely more useful to us.

Whats next?

The Semantic Web

Jeff Bezos described Web 2.0 as a way to make the web more usable for computers rather than people. When I first heard that, I thought “What!! Why would anyone want to do that?” Well, computers are excellent at storing and retrieving vast amounts information, however they do not understand the information.

For example, you cannot (yet) tell a computer program to find out where the next Jethro Tull concert is and if it is somewhere in the 48 contiguous US states, to automatically buy you a concert ticket, an airline ticket and book a hotel room. It took me a couple of minutes to look for the official Jethro Tull website and find the tour dates. Then I would have to go to orbitz or some other website and buy the cheapest (because ticketmaster would have gouged me badly on the concert ticket), airline ticket and hotel room to Lowell, Massachusetts. Of course, I would ideally like the computer program to realize that they are also playing in Chicago in November and tell me that it might be a better idea (i.e. cheaper) to wait. However that might be asking for a bit too much. Computers don’t know how to look at a web page and understand what it says. In fact there are some who say that it will never be possible for computers to develop Artificial Intelligence. The Semantic web is not about AI, it is about making information easier for computers to understand.

Imagine you are buying a box of cereal at the grocery store. You know it is a cereal box because that is written on the box. The computer cannot tell that it is a cereal box, even though it says cereal in funky colourful letters, and it has that fuzzy animal on it and pictures of the cereal. However, at the checkout line, the computer knows that it is a cereal box because it scans the bar code on the box. This bar code is correlated with other information in a database and the computer knows how much it costs, how many boxes are left in the store, how many are bought every day and whether it should automatically place an order for more. Suddenly the computer understands cereal boxes and can be made much more functional.

Analogous to this, we need to give webpages codes so that computers can understand them. One of the first things I did while writing this post was to give it an innocuous tag. If you look at the html source of this page you will see in the first line of this post Web2.0 has been linked using
< a href="http://technorati.com/tag/Web2.0" rel="tag"> Web2.0</a>
The rel="tag" is a way of saying that this post has been tagged with the keyword Web2.0. If you search at Technorati.com for Web2.0 you get all posts that have been tagged Web2.0. You can tag several things in a webpage or a weblog post. Why do things need to be tagged? Surely google could have found this post using the same keyword. (Never mind that this site would show up after several hundred results.) But google could have also found this site while looking for database. Since the post has been tagged, a computer can be programmed to find its tags and then it knows this post is about Web2.0 and not databases. This is tagging in its infancy and it doesn’t seem very different from the infamous Meta keywords. However it is like meta keywords on steroids for everything. I tag my bookmarks. People have been tagging everything from news stories, and images to bookmarks and websites. Computers now have a rudimentary understanding of what things mean and can put together a page like this together http://www.technorati.com/tag/italy which includes blog posts, images, and bookmarks tagged with the keyword Italy. Of course this does not always work well when we have words with two different meanings. However, the system is evolving and certain aspects of it will die out and others will survive and grow stronger.

Most homepages typically say things like “My name is …”, “I study at …”, “My phone number is…”, “Some interesting links are…”, etc. FOAF is a way to describe people and relationships to computers. It stands for Friend of a Friend. Once you have a FOAF document attached to a webpage, a computer knows what your name is, who your friends are, what your phone number is, etc.. It can also automatically explore groups of people and relationships within them and what common interests they might have, etc.. Here is a link to sample of a FOAF document. It is written using RDF and XML which is basically syntax and grammar for computers to understand things. Naturally you would get a computer to generate one for you.

Even this blog is available in a computer-friendly format. For example if you look at http://dcostanet.net/wordpress/feed/rss2 you will see this blog in a format that a computer can understand. This document is called an RSS feed and it is written using RDF and XML and attached to my blog. The computer understands several things about this blog including the content of the posts, the headlines, who the author is, when the posts were published, etc… The last one is my favourite, because I have 30 or so links of RSS feeds of various news websites and blogs and every morning I have instructed my browser to show me all the posts that were published in the last 24 hours. I do not need to check any of these sites regularly. Every morning I get, on a single webpage, all the articles published in the last 24 hours from those 30 websites. The computer knows and understands a little bit about the content of these pages and is able to provide a very useful service.

Click here to read the 2nd part of this article

Links

4 Responses to “The Evolving web”

  1. hkaul Says:

    waiting for part 2 :-)

  2. haris Says:

    Very nice article. Waiting for the sequel too.

  3. Test Blog » Blog Archive » The Evolving Web part 2 Says:

    [...] from del.icio.us

    The Evolving Web part 2

    Click here to read the first part of this article. Web Applications Many websites like Amazo [...]

  4. Bob Wyman Says:

    For a really exciting demonstration of Web 2.0 ideas, take a look at what Joe Reger has recently done! See:

    http://bobwyman.pubsub.com/main/2005/09/joe_reger_shows.html

    bob wyman

Leave a Reply