September 20, 2004

Pruning

How much information is too much of information? It takes me almost the entire day, when unhindered by meetings, to multitask and finish reading my RSS subscriptions. Last night, during one of those lack-of-sleep-driven mental ruminations, I was wondering how I could prune my subscriptions and reading lists and manage my time better.

Why this step is needed is because of two things. The first is redundancy. Too many people saying or quoting the same thing, without any new new perspective, leads to the waste of a significant amount of time. The second is the inability of technologists to counter the first problem. There is not a single RSS aggregator which will give me ranked results. This really sucks when you have to skim over thousands of words in a day to satisfy your need for feed.

Why would I want ranked results in my aggregator when there are websites like Popdex, Daypop or Technorati? The reason is pretty simple. These sites only crawl and rank popularly linked content. I want something that does the ranking only from the piece of the pie that I consider is important. I want a link or a button on my aggregator which, on clicking, will give me the day's most talked about link or topic.

Is it totally impossible? I do not think so. RSS and Atom feeds are XML files in the first place and almost every desktop aggregator caches the feeds locally. I have read, time and again, many a fan going on and on Xpath queries and/or Libxml2. What would it take to implement something similar on the desktop? Is it time, effort or just a pure lack of imagination?

Coming back to the matter of pruning, I have decided that the maximum age a feed will have to prove itself useful is about two weeks. If I am not compelled to read/check it again in that period of time, it gets the boot from the list. That leads to another problem. How do I synchronise my feeds with my Bloglines subscriptions? Anyone feeling particularly bookmarkletish, please?