How long does it take to reinvent?
Wednesday, October 17, 2007 by Dave Winer.
Technologist build on what came before. 
At least this is how I understand it.
I was struck by two posts that I came across this morning, both sad in a way, because they show how much is lost when engineers decide to reinvent.
First, a mail list post four years ago that argued that yet another piece of a our blogging architecture needed to be reinvented. It's sad because four years later it turns out that the old architecture was fine. It was just the last bit in a huge bonfire of reinvention. They said I stood in the way, but I didn't. The Roadmap of RSS 2.0 fully provided for what they wanted to do. I argued with the IETF too, privately, that they should start with RSS 2.0, so that as much of Atom would be compatible, just reinvent the parts that absolutely needed (in their humble opinon) to be reinvented. For some reason never made clear to me, they decided to throw out all our work. (Which they couldn't make stick, predictably.)
Then I came across this post by Paul Kedrosky, a very smart man, imho, who is looking for a river of news aggregator for his own feeds. Well, that was the first aggregator, the one in Radio UserLand (which was derived from My.UserLand, which goes back to 1999, the dawn of RSS). It was the aggregator that started the market. It's why everyone supports OPML, for example, because they wanted to make it easy for our users to convert to their products. I imagine that most developers working in this area don't know that's why they support OPML for feed lists.
I wish people had copied not only the file formats of our app, but also learned the lessons we learned about the user interface. Someday someone is going to declare a breakthrough in RSS apps, and it will work the way Radio 8 did in early 2002. I will say oy, probably to myself, softly, because I'm really starting not to care.
What's done is done. I'm not objecting to what happened, but I wanted to make a note that next time around, when someone wants to throw out what works and start over, tell them I said it takes a lot longer to reinvent than it does to invent. And it only seems like fun in the beginning, then it becomes a lot of hard work and in the end you don't accomplish anything, in fact you make everything harder, because you have to support both the old way and the new way, and that's more than twice as much work just saying fuck it and leaving well enough alone.
If it ain't broke don't fix it.
Anything that can go wrong will go wrong.
Be liberal in what you accept and conservative in what you send.
Every time someone comes up with one of those witty slogans, they paid for the lesson with a ton of wasted time and heartache. What will the next slogan be? There surely will be one, because we never learn not to reinvent.
