Friday, November 23, 2007

Centralization Bad for Internet

How could tinyurl, feedburner, and webcite be bad things? "Centralizing stuff is not good for the Internet," wrote Dave Winer recently. De-centralization is at the core of RSS and re-use of content. Centralization is for control freaks, bean counters, and moochers.

Most of us love tinyurl for what it provides, but did you ever consider what happens when it breaks, or what would happen if it disappeared for good? Steve Rubel addresses the idea recently in his post entitled "Could a Billion TinyURLs Go 404?" The possibility certainly exists. A better solution for a maintainable Internet would be using our own tinyurl services, rather than relying on a centralized service which could easily become overloaded or go extinct for other reasons.

So what about feedburner? A glitch in this service could render all subscriptions through it useless, with no notification to the users. RSS was created as an infinitely scalable method for distributing content. Feedburner works by funnelling traffic through a central site for the purpose of counting hits. It's very tempting to want to know who's reading, but using feedburner to do so makes us no better than the control happy paper publishers we so often criticize.

For kicks I checked the preferred subscription method to local newspaper columnist Joe McAdory's blog and the feed(s) from nearby newspaper The Montgomery Advertiser. Neither used feedburner. I was relieved to find that, and reminded of a familiar Confucious saying--"He who lives in glass house shouldn't throw stones."


Kevin Gamble said...

It may be bad, but the size of a site matters to Google. If the primary way people find you is through search then it's a catch-22. Decentralization might work for bloggers, but I'm not sure it'll work for "publishers".

I use Feedburner not because I can count subscribers. I don't really care about how many people are reading me. I do care what they are reading. I like to know if there is any interest in a topic. I've been quite surprised at the things that people read, and even more so by what they ignore. I have stopped writing on certain topics (so much) knowing that there is little interest. I wouldn't have known that without Feedburner.

I'm not particularly worried about Google going away. I think Feedburner will be with us for a while.

Greg said...

I wondered who that post might hit. I appreciate the comment, not only to hear the other perspective, but also so I can get an idea of who read the post. :)

I expect Google to be around a long while too. Doesn't mean they won't change feedburner to increase their branding, ease maintenance, or incorporate the feedburner system (software and hardware) with their searching system.

I have two primary issues with the "service." First is that it circumvents a design feature built into the protocol. That's a problem for me on pure principle. The second is that using such a "service" is exactly counter to the culture I'd like to promote. I'd like people to post because it is something important to them, not because it is what people want to hear.

This issue is not unlike the issues faced in mass media broadcasting. TV and Radio producers have long dealt with the inability to know exactly who was listening to what. They learned to deal with it by watching the indirect results, and we should too.

Kevin Gamble said...

You make a good point. It's easy enough to turn Feed Burner off. I'll have to give that some consideration.