Andrew Tridgell has long been something of an idol of mine, so it was with great anticipation that I took my seat at the Linux.conf.au keynote speech this morning.
If you donâ€™t know who Andrew â€œTridgeâ€ Tridgell is, or havenâ€™t at least heard of him, you can probably safely stop reading at this point. Suffice is to say, heâ€™s probably Canberraâ€™s greatest single contribution to the world of Free Open Source Software. He is in no small part responsible for Samba â€“ software that allows Linux machines to happily talk to Windows machines on a network (amongst a horde or other less well known but equally impressive contributions).
Manning Clarke Theatre 1 was basically full for this talk, with people lining the back wall. As has earlier been noted, the ratio of male to female attendees at this conference would, I would guess be something in the ratio of 50-80:1 and this audience was no different. Not nearly so many laptops out for this one though â€“ most happy to just sit and listen â€“ those of us with them were much more in the minority.
While stating at the outset that this would not be a â€œtechnical talkâ€ but more along the lines of â€œphilosophical musingsâ€, these things are all relative. Those in the audience who werenâ€™t software people probably got a whole lot less out of the presentation than the coders present (however given the type of people who turn up to these things, I think it was an excellently pitched talk).
The bulk of the talk was made up discussion about software engineering practices used in Samba, which was fascinating to those of us who spend our lives staring at code, but I wonâ€™t go into in any detail here â€“ Iâ€™m sure thereâ€™s a thousand other blogs on the net that will happily share his wisdom on that topic with those who are interested.
Amongst the other things touched upon was the fact that he never formally studied computer science (despite having taught it to me and many others). Damn him. Now I have to respect his abilities even more.
But the real thing a lot of us wanted to hear about was his reaction to the recent â€œattacksâ€ on him by fellow Open Source God Linus Torvalds regarding Bitkeeper (this has been covered in a couple of previous stories for which Iâ€™m too lazy to find the links). He has been conspicuously silent on the matter, allegedly at the behest of legal advisors, so everyone hoped this might be an opportunity to put his side.
Alas, direct comment on Linusâ€™ statements was not forthcoming. What he did have, however, was possibly even more interesting. For those who arenâ€™t aware, the specific bit of biffo between the two has been regarding Tridgeâ€™s â€œreverse engineeringâ€ (the reason for the quotes will become obvious) of the Bitkeeper protocol which was until very recently used as the main repository for the Linux kernel source code.
While most of the opinions Iâ€™ve read seems to side with Tridge, the split seems to be whether he could have done it without having access to the proprietary client (which might put him in somewhat more shady legal territory). Some say â€œhe must have used itâ€, others â€œTridge can reverse engineer anything â€“ heâ€™s a wizard!â€ (the latter is not a wholly unfounded view).
So what did he do? He â€œreverse engineeredâ€ Bitkeeper right in front of our eyes on the big screen. Or rather he didnâ€™t â€“ he got people in the audience to yell out what commands he should type in â€“ turns out that not only is it so trivially easy to do that anyone with even the most basic knowledge of network protocols could do it, but that the Bitkeeper server itself has a â€œhelpâ€ command that tells you exactly what you need to do as a client. I present now, an audience generated â€œhackâ€ of Bitkeeperâ€™s â€œproprietaryâ€ protocol:
echo colne | nc server.org 5000 > localfile.dat
Anyone whoâ€™s ever tinkered with Linux networking stuff will find that ludicrously trivial. It made the whole argument thatâ€™s been going on about Tridge acting irresponsibly, and going out of his way to reverse engineer the protocol look really really dumb.
The only other thing of note is that it was a full hour and twenty six minutes before the first Monty Python quote was yelled out from the audience. I was disappointed â€“ you can fit a lot of geeky quotes into that much time, and as a city we should have done better.