There’s been a pretty fascinating discussion going on in the blogosphere recently about the benefits and hazards of centralization on the Internet.  Three main hot button topics have come and have ignited the discussion:  TechMeme, TinyURL, and FaceBook.

If you’re not familiar with TechMeme, it’s a site that aggregates blog posts around a particular topic (meme) and creates “conversations” around them.  It’s pretty popular among the Web 2.0 crowd and many people—myself included—tend to use it as a tech newspaper.  It’s really handy in that regard, because you can see when there’s a popular topic that’s getting a lot of coverage without having to subscribe to thousands of feeds.

On the other hand, the way TechMeme aggregates the posts is something of a mystery.  It’s unclear whose blogs are on its “reading” list (my blog is part of the mix, and I’m not exactly sure how it got there), and they seem to be giving a lot more traffic to the big corporate blogs such as CNN, CNET, and TechCrunch lately. People are complaining that this is killing genuinely new and innovative memes off the site, and I tend to agree.  It used to be an incredibly interesting site to read because you’d see all kinds of offbeat and obscure ideas that you would otherwise never have heard of.  Recently it’s been more along the lines of technology news commentary, which is a real shame.  I would almost rather see the site broken up into an “A List” and “B List” site, I’d probably pay much more attention to the “B List” version.  But the point is that the one site having the power to decide which blogs are important, and thereby which memes are important, seems to be hurting the diversity and egalitarianism that made blogs interesting in the first place.  It’s a single point of failure in the brainstorming business.

And then there’s TinyURL.  Most people are probably familiar with it—but if you’re not it’s just a service that will take a long, obtuse URL and shorten it into a URL of about 15 characters.  It’s become more popular recently thanks to the micro-blogging phenomenon using sites like Twitter which let users post extremely small blog posts.  Usually the posts are limited to about 200 characters, so if you want to embed a link in your Twitter post, TinyURL is a handy little tool to help save space.

The downside, as with TechMeme, is that TinyURL is a single point of failure.  It went down the other day, and all of a sudden many, many links stopped working.  It poses the question of whether using any single point on the Internet as a conduit for so much traffic is actually a good idea, even if it makes life a little easier.  (I also have questions about where the Google juice for TinyURL’s go, to TinyURL itself or the destination site.)

And then there’s privacy—the big 500-lb gorilla hiding in the corner.  People are starting to have concerns about what Facebook and other social networks are doing with their data.  I’ve been blowing this horn for so long I’m starting to get tired, and the only way around this roadblock is to actually OWN OUR OWN DATA (duh).  IMO trying to fix the companies themselves is just a band-aid--just like tamed lion, even the most trustworthy companies can turn on you at any time ("Don't be evil" comes to mind here).  Of course, if you willingly give a company your data, they will use it to make the most money they can.  If you don't like it, you should rethink capitalism.  Don’t try to fix the companies, fix the freaking paradigm.  This is one area that needs decentralization in a bad way.

What I think is interesting is to watch this dance back and forth between centralization and decentralization.  For every two steps we make towards decentralization, we take a step backwards towards centralization.  What starts out decentralized ends up centralizing for a time in the name of progress and convenience, until the centralized systems are shattered into a million pieces as necessity and self-preservation demands it.  Technology innovation tends to happen in one spot (company, site, community, etc), and shortly afterwards that one spot becomes a centralized hub for all activity around that innovation.  Until it’s not.

This whole discussion leads to another fascinating topic, one that I think is going to be the defining trend in the next phase of Internet evolution:  personal Internet ownership.  It’s the ultimate level of decentralization, decentralizing the Internet so that it operates on the basis of each individual person, not relying on these central hubs we have today, like social networks and Old Media outlets.  Taking the Internet’s server network and turning it into a human network.