I'm about to leave for a long weekend but before I do I wanted to put some brief follow up thoughts to the whole Identi.ca thing.  In doing so I wanted to make the point that I'm not actually trying to bash the program.  Anyone who goes to the effort of writing a program, putting up a server, and sharing it all with the community deserves praise.  But that doesn't mean we should ignore the flaws in that system. 

With that said, here are some things I didn't cover yesterday...

Wanting To Believe

Without quoting anyone specifically the general sentiment from those who were praising Identi.ca seemed to be...

"Its lacking features but I'm sure they'll be implemented soon"

The irony here is that an open source project is less likely to implement features quickly.  Say what you will about the monstrous amounts of money being pumped into startups the reality is that millions of dollars goes a long way towards implementing features quickly.  An Open Source project relies on the free time of the people involved which is a recipe for slow implementation. 

But people want to believe in this so they twist the facts to fit their desire.  It never ceases to amaze me how easily people can convince themselves that what they want is the likely outcome even when every logical factor is stacked against them. 

I love this spec so much that I'm just about to start reading it

One of the things that surprised me the most is that people were willing to jump right into supporting the openmicroblogging.org spec even though they obviously hadn't read it.  Had they done so it would have become clear that it doesn't account for a lot of the basic Twitter features.

In recent years people seem to have strayed from the concept of approved standards because the approval process takes so long.  But it pains me to see us having gone so far in the other direction that we are encouraging others to put up servers based on a protocol that's completely unvetted.

Happy Days Really Aren't Here Again

Even more shocking to me is the fact that people are giving this credit for things it doesn't solve.  If you want a decentralized Twitter this doesn't seem to solve your problem.  Sure, other people's servers might stay up, but you still won't be able to post to yours.  So if your problem is that you can't post during server outages this really doesn't solve that problem.  In fact, given Twitter has multiple servers while anyone implementing this will probably only have one, it might just make your problem worse. 

This product is great, just don't click on that.  Or that.  Or that. Or that...

The most interesting part of the reaction to Identi.ca is that half the time I tried to click on a link and I got a message asking me if I wanted to download a file with an unrecognized MIME type.  I'd think this was a configuration problem on my side if those same links didn't work the other half the time.

I can't be the only one who had this problem so my question is: Why did no one mention this?  I couldn't find it once in all the blog posts about the service. 

Again with the Open Source

This is a reiteration from yesterday's post but it really bugs me.  Open Source is great.  I personally learned a lot from people who made open source programs available.  But when talking about actually USING an open source program the fact that its open source is only valuable if there's a team of people dedicated to working on it. 

Despite what some advocates would have you believe 99.9999999999999% of programmers wouldn't go anywhere near open source projects because (a) they program all day and don't feel like doing it at night, (b) there's a huge disincentive at the outset because you have to learn someone else's code backward and forwards, and (c) coordinating efforts between other programmers who are doing something part time is a huge pain. 

The point about learning someone else's code is the most potent of these issues.  You can't just jump in and start coding on a big project.  Doing even the simplest thing requires you know every subsystem you'll be interacting with backwards and forwards.

So when ReadWriteWeb says...

Can it work, work it scale? At least it's open source so the development community doesn't have to play armchair quarterback for a black box like they are with Twitter.

Or Ewan Spence says...

I suspect that with an Open Source code base they’re going to get a lot of eyes looking over problems and tweaks

They're being naive.  Twitter's has experienced programmers whose full time job is to sort out its problems and look how far that's gotten them.  Finding code bugs in a large program that you aren't intimately familiar with is next to impossible.

As far as program functionality goes, the truth is that a good plug-in framework is way more useful than open source is. 

I Could Go On All Day

I don't want to belabor the point anymore but hopefully you see what I'm saying.  There are a lot of questions  here with seemingly no answers.  The blogosphere seems to think appending "Open" to the name of something instantly makes it either Golden or destined to be Golden but that really isn't the case at all.

Maybe this will take off and maybe they will implement every feature that Twitter has but until they do it seems premature to consider this a viable Twitter alternative. 

P.S.  To anyone in the U.S., enjoy your holiday and remember: No explosion, no matter how cool, is worth the loss of a limb.

Well ok, maybe a REALLY cool one but lets face it, you can't afford those kind of Fireworks.  If you could you'd also be able to afford someone to light them for you.