Eric S. Raymond - Cathedral and the Bazaar

Eric Steven Raymond (born December 4, 1957), often referred toas ESR, is a computer programmer, author and open source softwareadvocate. His name became known within the hacker culture when hebecame the maintainer of the "Jargon File". After the 1997publication of "The Cathedral and the Bazaar", Raymond became, fora number of years, an unofficial spokesman of the open sourcemovement. (Source: Wikipedia)

Copyright 2000 Eric S. Raymond

Permission is granted to copy, distribute and/or modify thisdocument under the terms of the Open Publication License, version2.0.

Linux is subversive. Who would have thought even five years ago(1991) that a world-class operating system could coalesce as if bymagic out of part-time hacking by several thousand developersscattered all over the planet, connected only by the tenuousstrands of the Internet?

Certainly not I. By the time Linux swam onto my radar screen inearly 1993, I had already been involved in Unix and open-sourcedevelopment for ten years. I was one of the first GNU contributorsin the mid-1980s. I had released a good deal of open-sourcesoftware onto the net, developing or co-developing several programs(nethack, Emacs's VC and GUD modes, xlife, and others) that arestill in wide use today. I thought I knew how it was done.

Linux overturned much of what I thought I knew. I had beenpreaching the Unix gospel of small tools, rapid prototyping andevolutionary programming for years. But I also believed there was acertain critical complexity above which a more centralized, apriori approach was required. I believed that the most importantsoftware (operating systems and really large tools like the Emacsprogramming editor) needed to be built like cathedrals, carefullycrafted by individual wizards or small bands of mages working insplendid isolation, with no beta to be released before itstime.

Linus Torvalds's style of development-release early and often,delegate everything you can, be open to the point ofpromiscuity-came as a surprise. No quiet, reverentcathedral-building here-rather, the Linux community seemed toresemble a great babbling bazaar of differing agendas andapproaches (aptly symbolized by the Linux archive sites, who'd takesubmissions from anyone) out of which a coherent and stable systemcould seemingly emerge only by a succession of miracles.

The fact that this bazaar style seemed to work, and work well,came as a distinct shock. As I learned my way around, I worked hardnot just at individual projects, but also at trying to understandwhy the Linux world not only didn't fly apart in confusion butseemed to go from strength to strength at a speed barely imaginableto cathedral-builders.

By mid-1996 I thought I was beginning to understand. Chancehanded me a perfect way to test my theory, in the form of anopen-source project that I could consciously try to run in thebazaar style. So I did-and it was a significant success.

This is the story of that project. I'll use it to propose someaphorisms about effective open-source development. Not all of theseare things I first learned in the Linux world, but we'll see howthe Linux world gives them particular point. If I'm correct,they'll help you understand exactly what it is that makes the Linuxcommunity such a fountain of good software-and, perhaps, they willhelp you become more productive yourself.

Since 1993 I'd been running the technical side of a smallfree-access Internet service provider called Chester CountyInterLink (CCIL) in West Chester, Pennsylvania. I co-founded CCILand wrote our unique multiuser bulletin-board software-you cancheck it out by telnetting to locke.ccil.org. Today it supportsalmost three thousand users on thirty lines. The job allowed me24-hour-a-day access to the net through CCIL's 56K line-in fact,the job practically demanded it!

I had gotten quite used to instant Internet email. I foundhaving to periodically telnet over to locke to check my mailannoying. What I wanted was for my mail to be delivered on snark(my home system) so that I would be notified when it arrived andcould handle it using all my local tools.

The Internet's native mail forwarding protocol, SMTP (SimpleMail Transfer Protocol), wouldn't suit, because it works best whenmachines are connected full-time, while my personal machine isn'talways on the Internet, and doesn't have a static IP address. WhatI needed was a program that would reach out over my intermittentdialup connection and pull across my mail to be delivered locally.I knew such things existed, and that most of them used a simpleapplication protocol called POP (Post Office Protocol). POP is nowwidely supported by most common mail clients, but at the time, itwasn't built in to the mail reader I was using.

I needed a POP3 client. So I went out on the Internet and foundone. Actually, I found three or four. I used one of them for awhile, but it was missing what seemed an obvious feature, theability to hack the addresses on fetched mail so replies would workproperly.

The problem was this: suppose someone named 'joe' on locke sentme mail. If I fetched the mail to snark and then tried to reply toit, my mailer would cheerfully try to ship it to a nonexistent'joe' on snark. Hand-editing reply addresses to tack on<@ccil.org> quickly got to be a serious pain.

This was clearly something the computer ought to be doing forme. But none of the existing POP clients knew how! And this bringsus to the first lesson:

1. Every good work of software starts by scratching adeveloper's personal itch.

Perhaps this should have been obvious (it's long been proverbialthat "Necessity is the mother of invention") but too often softwaredevelopers spend their days grinding away for pay at programs theyneither need nor love. But not in the Linux world-which may explainwhy the average quality of software originated in the Linuxcommunity is so high.

So, did I immediately launch into a furious whirl of coding up abrand-new POP3 client to compete with the existing ones? Not onyour life! I looked carefully at the POP utilities I had in hand,asking myself "Which one is closest to what I want?" Because:

2. Good programmers know what to write. Great ones know what torewrite (and reuse).

While I don't claim to be a great programmer, I try to imitateone. An important trait of the great ones is constructive laziness.They know that you get an A not for effort but for results, andthat it's almost always easier to start from a good partialsolution than from nothing at all.

Linus Torvalds, for example, didn't actually try to write Linuxfrom scratch. Instead, he started by reusing code and ideas fromMinix, a tiny Unix-like operating system for PC clones. Eventuallyall the Minix code went away or was completely rewritten-but whileit was there, it provided scaffolding for the infant that wouldeventually become Linux.

In the same spirit, I went looking for an existing POP utilitythat was reasonably well coded, to use as a development base.