Thursday, April 8, 2010

Stepping on the Little Guy in Blogging

@holdenpage recently concocted a very interesting idea that would definitely be a revolutionary concept in the world of blogging. Instead of aggregating the most popular posts as they come out, why not look for the more interesting and meaningful posts that relate to the most popular posts, and aggregate those instead. In other words, if TechMeme comes out with the latest TechCrunch article, this new aggregator would wait a day or so, then pull in articles related to that original topic. The articles pulled in would be from personal blogs, not the top-tier bloggers, and would have more in-depth analysis on the subject. All I have to say is that it's about time somebody thought of the little guy.

The unknown blogs spread across the Internet, much like this one, are set apart from the top-tier blogs like TechCrunch. Why? Because there are two time-tested methods of creating good content: the first is to be the first person to publish about a topic, and the second is to offer a unique view on the topic. TechCrunch and other big blogs usually take the first method, publishing an article mere minutes, or even seconds, after the content is announced. The speed of service attracts visitors because people do not care for old news. Repeat content is boring, and users want to hear what is happening as it happens, then just be done with it. Because small bloggers cannot output at the speed of TechCrunch, they resort to the alternative method: giving a unique view on a topic. This means being a devil's advocate and taking a perspective the normal user would not expect. This brings in users because it makes for interesting articles, though it is unlikely this method will make you as popular as TechCrunch.

Getting back to the proposal. Holden wants to bring these personal blogs, the ones with unique viewpoints, and surface them in one location. In this case, that location is LayeredByte, a blog that seems to be inactive right now. If this idea pulls through, it will mean revolution in the blogging community. For the first time that I can think of, there will be a way to pull meaningful content about a subject and read about something from all sides, rather than just get the non-POV news. I don't know about you, but if I had the opportunity to read about a current event from all sides rather than just here an announcement about it, I would be able to formulate my opinion with a more accurate picture of the situation. Anyway, why this would be a revolution is because, if pulled off right, the aggregator could pull young and small bloggers into the community, something that will enrich the entire world of tech, and allow new perspectives to be introduced with more ease than ever before. Also, this would ease the media's control over what we read, because we would no longer just be reading the top-tier blogs' content.

There is only one problem with the idea: how? The reason we do not hear about personal blogs like this one is because they are hard to come by. So how do you come by them then if you intentionally want to find them? Just taking the latest meme and searching it will not help, because search engines put the big blogs first, and sifting through search results is inefficient as best. Furthermore, without a search engine, how can you be sure the content is related to what you're aggregating? Finally, how can you implement quality control, so that after all is done and the website's algorithm fishes out those personal blogs, the site can then look back and find ways to improve itself automatically. These are all difficult problems to solve.

There is only one real solution I can think of off-hand, and it might just be a pretty good solution. You start with a seed, let's say Holden Page himself, and then you branch off into the social network of that seed, analyzing every person that the seed follows (and maybe everybody who follows him). Much like a PageRank system, the algorithm then takes the number of followers of each branch from the seed, and ranks the network members according to popularity. From here, top-tier bloggers can easily be eliminated, as they will likely have a few thousand followers. From here, most personal blogs, such as this one, are either WordPress or Blogger-hosted, so comb through any branches under the cut-off point for blog-like sites (the cut-off point should probably be something like over five hundred followers). So by now, you have a list of blogs from the less-popular members of the seed's social network. Continue this process until you have a sizable list of blogs (at least a few hundred, if not thousands).

Now comes the fun part. So you have your list of less popular blogs taken from the social network of the seed. You need content. Use some type of internal search engine to comb through the RSS feeds of the sites to find relevant posts to the topic being aggregated. How it would work in the beginning is that the list of blogs are arranged according to the popularity calculated in the first step, and whatever blogs happen to be on top of the list get precedence. Finally, the website can generate feedback by asking for ratings (such as a five-star scale or something) to determine the quality of the post. If a blog's post consistently get low ratings, it will move down on the list of blogs, making it less likely to be picked for aggregation. So in the end, you bypass the top-tier blogs by cutting them off, then use a feedback system to teach the website what posts to display.

This system could probably be improved, so if anybody has any suggestions, feel free to send Holden an email, or post a comment on this post.

No comments:

Post a Comment