Exactly this. While I'm not a huge fan of the "Bazaar" as a development model, I'm even more terrified of the "Cathedral" and/or anybody who has an idea that th

Exactly this. While I'm not a huge fan of the "Bazaar" as a development model, I'm even more terrified of the "Cathedral" and/or anybody who has an idea that there's one-true-way to do things (and it's always theirs and only theirs). Behind that scary door is the war of architects and their egos, splintering half-completed standards, and nightmares from the past like CORBA etc -- maybe-decent on paper, but good luck ever finding an open and usable and actual quality, working implementation. What we need is to develop consensus on how to do things, and that, unfortunately, is messy, takes time, and involves people compromising all over the place. And I'm not convinced that committees are the right instrument for that either. The essay's slagging of the .com era is funny to me as well; because I lived through that era and was absolutely one of the messy green hackers/devs that he is complaining about. Funny thing is though that things shipped, still, and it led to a lot of people being able to make web pages, write code, and get themselves out on the Internet who hadn't had that privilege before. And it wasn't the quality of the software that caused the .com implosion after, it was the quality of the business plans. And I also saw what came after, the kind of reaction to the .com chaos; it was all J2EE and EJBs and big enterprise application servers, and I never worked in a single shop where that actually led to anything quality and saw more than one where a team that was doing Perl+CGI+Apache ate itself by rewriting everything "the right way." A few years later, the trend swung back the other way; Ruby on Rails, web2.0 etc. etc. Software development is messy. That's just how it is.