The central ideas behind the emergence of the PubSubWeb are as follows. Tools like wikis and weblogs are making writing easier. Search engines like Google make finding information and websites easier. Yet, repeated, periodic reading of an information source triggered by when it is updated remains hard. To get the incremental content on a website, one needs to not only remember that site but also suffer through while a lot of seemingly irrelevant content is downloaded around the content we really want.
What is needed is a mechanism for microcontent from the sites (or people or databases) we want to be delivered to us in near real-time. Ideally, we should be able to do this with the tools that we already have, specifically the email client and the browser. There is no need to add to the complexity of downloading and learning yet another application.
Another way to think of the PubSubWeb is as an EventWeb. Each update of the content (publishing) is akin to the occurrence of an event. What is needed is for us to be able to (a) subscribe to the event stream and (b) receive notification and details of the event as and when it happens. From a publishers point of view, there may also be a need to restrict access to who can subscribe to the event stream.
What we think of as microcontent or events is only limited by our imagination: a breaking news story, a commentary published by one of our favourite writers, an intimation of a speech in the town we live, the birthday of a friend, a stock quote, the latest scores of a cricket match, an alert about a flights arrival or departure, a snippet about an SME wanting to buy something we are selling, an alert about the most recent credit card transactions, or perhaps an update on the new leads that have come in to the sales department.
Of course, we are able to get access to most of this information and in some cases, even receive the appropriate notifications today. But it isnt easy. Because, so far, the tools that we have been using have been focused on one-way flow of information. Our ability to customise what we see and focus only on the incremental information has been extremely limited. In most cases, we have to go seek out the information source and pull in the information we need. In most cases, push would have worked far better. But so far, there hasnt been a standardised way to make this happen.
Tomorrow: RSS Revolution
TECH TALK PubSubWeb+T