Had it failed, the XP high priests would probably say, “oh, but they didn’t do it right — they didn’t write the user stories on index cards” or something to that effect. But since it succeeded, they’d probably say it’s all due to the use of XP.

Yes, that’s what I meant about not falling for hype.

Just to be fair, it probably wouldn’t be too hard to find someone willing to blame XP if the project failed. But since it succeeded, they’d say it must have succeeded in spite of XP.

And that would be the reverse hype.

The truth is usually more nuanced than that. But people are not big fans of doing proper project retrospectives, and I suspect that part of it is that people get all too comfortable with the personal conclusions that they’ve jumped to and rationalized.

Anyway, if we project the “you didn’t do it right” into another context, it doesn’t necessarily seem so outrageous. XP enthusiasts believe the practices work together, like a well-oiled machine. If you told a bicyclist that you tried to ride a bike, but it fell over, would you not expect to hear something along the lines of “you didn’t do it right”? That’s the state of mind you’re dealing with.

you compare XP to “crossing railroad tracks in front of an oncoming train”. I can argument that this is a bad idea, because the train will hit you and the force will probably kill you in one way or the other.

If you think of XP, what is the equivalent to the “hitting experience”? I’d be interested, because at the moment I’m trying to cross the railroad tracks ;-)

L.E.,
I agree that the name “Extreme” is probably a bad choice from today’s perspective. I was repulsed by it at first, too. Perhaps it wasn’t such a bad idea in the last century?

OK. I don’t actually recall hearing any of them call it “XP”. But I know they pair programmed in an open workspace, had 1-week iterations, close contact with a customer, and when I was brought in on a contract for a different project, I saw their iteration velocity table on a whiteboard in their room, covering about a year of development. I didn’t see any index cards.

Had it failed, the XP high priests would probably say, “oh, but they didn’t do it right — they didn’t write the user stories on index cards” or something to that effect. But since it succeeded, they’d probably say it’s all due to the use of XP.

Anyway, I agree. Good developers are not big on labels. Especially “extreme”, which is a terrible misnomer. I also believe that it usually works out better to attract developers with agility than impose agility on an existing set of developers who may or may not be interested.

So… better teams of developers prefer to use XP? Hmm, interesting. :)

I’m only half-kidding here. Most of the very good developers I know aren’t very big on labels, and don’t generally call what they do XP (I personally don’t like the “Extreme” label, and pick and choose which practices to apply based on my current context), but they will freely admit they use a lot of the XP practices, and get a lot of value from them. I sincerely believe one of the biggest benefits of running a project in an agile way is that it’s a great way to attract a bunch of great developers. Just do your best to filter out the bandwagon-jumpers and buzzword-compliant drones.

If something strikes me as an obviously bad idea, but other people think it’s a good idea, I try to find out why. I’m curious, and I have a healthy level of self-doubt and skepticism. For what it’s worth, the XP projects I personally know about have all been successes, at least one of them spectacularly so, in particular because they succeeded after two other non-XP teams failed. (Appropriate skepticism: it could have been just because it was a better team of developers.)

So who says a cult can’t deliver good software? Being in a cult would be really annoying, I think, but at the end of the day I just want to deliver good software. If it’s a cargo cult, of course, that’s a different story. You have to be careful about that.

I’m reminded of the grizzled veteran mainframe programmer who told me in ‘92 that OO was just a fad, and the whole thing was going to blow over in a few years. Every fad generates its naysayers, and zealots, books, conferences, hype… But similarly, every fad has some kernel of truth. I see it as my job as a professional to get to the bottom of it, to cut through the hype, not to fall for it hook, line, and sinker, or to make the equal and opposite mistake and dismiss it out of hand.

The long term success of XP will probably depend a lot on whether it really works. To get the answer you can try it out for yourself. Apart from that only time can tell if XP will be a movement that changes the way we implement software systems or yet another hype.

I’m the guy who posted that link. In response to the paragraph quoted above, I’ll say that time and the marketplace have already rendered their judgements on XP. It generated a lot of hype for a few years after Beck’s book was published in 1999, but these days you don’t hear much about it anymore. I don’t have empirical numbers, but I’ve been in the industry longer than XP’s been around, I know lots of developers who work for lots of different companies, and not a single one is working in an XP shop.

Moreover, if something strikes me as an obviously bad idea (like crossing railroad tracks in front of an oncoming train) I don’t feel the need to test the idea to make sure it’s not a good one.