Monthly Archives: January 2009

LCA2009 Day 5 — Friday

Friday’s keynote was fantastic — Simon Phipps (who retained his job at Sun Microsystems) spoke of the Third Wave of Free Software, which was an observation that Free Software (specifically not Open Source) is becoming a sensible business proposition. It’s about time that it did. It was refreshing to see a large corporate’s view of the world of Free Software largely agreeing with my own. Talks after morning tea were Tridge’s talk on his automatic cluster testing framework (pretty cool), followed by Conrad Parker’s talk on Ogg Chopping, which despite the name, was actually a 50-minute rant about why Haskell is cool — I’m sold (I think), but somewhat confused about the talk — really, I have no idea what happened. I strongly urge you to watch the video (when it becomes available) in order to figure out what happened for yourself.

Lunchtime was the Great Unbeardening — Linus Torvalds (who was roped into the act at the auction on Wednesday) shaved Bdale Garbee’s beard — the result? Disturbing. Really disturbing, but all in the name of charity. The #lca tag on twitter was displayed on the projector screen, so live audience responses were shown as the shaving continued, including one Maclabbian pointing out the relative weirdness of the event; photos were up on Flickr well before the end of the event, and Southern Cross News came to film the event (focusing on the shavee, and not the mysterious Finn doing the shaving…). 2009 -- Day 5 2009 -- Day 5

Matthew Garrett’s talk on Power Management that works was great: nothing too technical, but an excellent discussion of the user interface issues surrounding power management. Matthew’s talk was unique in that his talk covered everything in his abstract — this includes answering the question “will we ever get to beer island?” — the answer? Yes, provided you’re in Texas. Following was Geek My Ride, presented by Jonathan Oxer and Flame — this was a pretty cool demo talk, showing how the two of them have modded their cars to include some pretty cool stuff, including in-dash diagnostics, MP3 playback, and remote ignition (wow cool!).

The final talk of the conference was Bdale talking about rockets, which as usual were pretty cool. Lightning talks concluded the conference as they did in 2008 — nice to see them becoming an LCA tradition, I will definitely aim to present at least one next year.


So that’s it. LCA is over for another year, and will reconvene in Wellington, New Zealand for 2010 — I’ve never been to New Zealand, and am really looking forward to going there next year. The 2009 Conference was excellent, the talks were well-presented, and the organisation of the conference was such that it appeared from the outside that the everything ran well (I’ve been told that that was certainly not the case). The conference allowed us to show off Hobart to the technical world, which is an opportunity that does not present itself regularly — I’m glad that we got that opportunity, and I think most delegates this year will agree that it was an opportunity that was well-received, and resulted in an excellent conference for all involved.

Also thanks to Adam Harvey and Monica Wood for helping out at UpDNS — you certainly made my job in organising it a lot easier; Linux Australia for having faith in the organisers ability to put on the conference — I hope your investment in the Tasmanian Free Software Community pays off.

LCA2009 Day 4 — Thursday

The opening Keynote on Thursday was a discussion of the Wikimedia/Wikia project, which was overall not too bad. The highlight of the talk was the relaying the Parable of the Vegan, which was quite hilarious. Sadly, I don’t think the talk was quite as good as it could have been — too much time was spent teaching the purpose of Wikipedia and the structure of the wiki community, which I think was generally common knowledge amongst the audience. Once questions were asked, it became generally more interesting.

After morning tea were the absolutely fantastic Perl talks of Paul Fenwick, the first was “The Art of Klingon Programming”, which made a hugely insightful analogy between the Perl programming language and Klingon culture, and used this to inspire his talk about libautodie, a library that makes Perl behave sanely in the face of supposedly fatal errors. It’s pretty damn cool, and cleans up one of my least favourite things about Perl (though quite a few still remain), Paul’s second talk of the day was on new features in Perl 5.10, which were interesting. Perl 5.10 has added a swtich-alike block, which I think is a model that other languages should adopt — instead of the C-style ‘break-or-fallthrough’ method (which introduces many stupid bugs for newbies), Perl adopts the ‘continue’ keyword to allow a fallthrough, or no statement to break — this is pretty damn clever, if I do say so myself.

This was followed by a talk entitled 7 Things Lawyers Don’t Understand About Software — delivered by a UTAS Law PhD student, who presented some very interesting arguments about the likeness of software and mathematics, and related this to the unpatentability of mathematics. His research appears interesting and I urge you to check it out.

After Lunch was Donna’s The Joy of Inkscape tutorial, the point of which was to have people tinker with inscape for two hours, with occasional supervision from experts. Sadly, the room (which holds 40 with tables) overflowed, and hence the tutorial didn’t appear to function entirely as planned. I don’t think this affected the ability of people with seats to enjoy the tutorial though, which is nice. 2009 -- Days 3&4

Following Afternoon tea, I relaxed for a while (indeed skipping a talk), and finished up at Hugh Blemings’ talk on learning Free Software Hacking from Clever People — this talk was a disappointment — from casual observation, people in the audience provided more useful input than the speaker, and the speaker was mostly inaudible (partially due to his tone of voice not agreeing with the room acoustics).

Thursday Evening involved me and various TUCS people running the Unprofessional Delegates Networking Session at the Uni Bar — this was a massive success for us (we turned a profit!!!) and we were happy to provide an opportunity for the non-professional delegates to socialise whilst the professionals enjoyed their pissup at a brewery. Never underestimate the power of meat and quanitity burger to attract and feed people. I should direct many thanks to the business staff at the TUU and the Uni Bar, who opened their facilities at 3 days notice, for what was effectively a break-even prospect. Hopefully TUCS can do more work with them in the future. 2009 -- UpDNS
(Photo by Adam Harvey)

LCA2009 Day 3 — Wednesday

Tom Limoncelli opened the conference proper with his keynote on “The Scarcity Mindset vs. The Abundance Mindset”, which was an interesting and insightful talk on how the modern abundance of computer hardware, coupled with open source software can help sysadmins make better use of their resources. I must congratulate the organisers of the conference for selecting a sysadmin talk as the opening keynote: LCA has suffered from a lack of sysadmin talks (which the miniconf has fixed to a certain extent), and having such a keynote open the conference is an excellent investment in furthering sysadmin content at future LCAs. That said, the talk was very sysadmin-specific, and was therefore not directed well at the majority of delegates (discussing things like implementing better tech support policies in the workplace). For those more willing to look at the bigger picture (i.e. by factoring out the direct application to sysadmining, the abundance mindset is certainly something that can be used to better support open source development. 2009 -- Days 3&4
(Photo by Thomas Karpiniec)

This, for me, was followed by Keith Packard and Carl Worth doing a double-feature talk on recent developments in X, followed by a demonstration of the Linux graphics pipeline, including some discussion of how graphics drivers can be improved to allow better rendering from (say) Cairo and OpenGL. I went to the Django tutorial after lunch, which was reasonably interesting, though I stopped actively participating about halfway through. It was interesting to see how Django works, and how some of its choices were made as far as design was concerned — in particular how there are many features designed for journalists, since Django was developed in-house by Journalists in Kansas.

Post-afternoon-tea, I went to Jonathan Corbet’s talk on joining the Kernel development process, which was a departure from his usual “Kernel Report” talk — he explained the significance of the various trees of kernel development, and explained how to work with subsystem maintainers in order to ensure that your driver became well-maintained into the future. This talk was significantly enhanced by Linus Torvalds being in the room, helping answer questions and providing further input into the presentation as it presented: this was probably the closest he got to presenting for the entire week. Following this was Martin Krafft’s talk on the project, or how distribution package maintainers can collaborate with their counterparts at other distros using distributed VCSes such as Git. He presented a good discussion of his own workflows, as well as discussion of the suitability of various tools for the purpose of cross-distro collaboration. 2009 -- Days 3&4

The real highlight was Wednesday night’s penguin dinner, which featured the most bizarre auction that I’ve ever seen, which ended up with a consortium of Kernel Hackers and Collabora paying AUD$10,500 for a print of a photo, and a beard. This has already been discussed adequately elsewhere, so I shalln’t bother.

LCA2009 Day 2 — Tuesday

Tuesday I spent mostly at the Free as in Freedom miniconf, where I saw Arthur Sale’s talk on Open Access journals (where he outlined the changes that need to occur in the research publishing industry in order to support research in the online age), Jeff Waugh’s talk “We are the Translators”, which drew some enlightening parallels between Gutenberg, early Protestants (who translated the bible into modern languages, much to the disgust of the Catholic church) and the modern Free Software movement. 2009 -- Day 2

The final talk of the day was presented by Rusty, which essentially consisted of a fantastic 25-minute rant against modern IP law. The talk was passionate, interesting, and featured an interpretive dance about the dangers of software patents. I think that just about sums the talk up. Here I also met Paul Fenwick (developer of Autodie, the library that makes perl behave sanely in the face of errors); to my horror, many of my friends, who were sitting in the same general area as me hadn’t seen Paul’s lightning talk from 2008, so I took the opportunity to show them — it’s still as fresh and witty as it was last year and if you haven’t seen it I urge you to again. At the conclusion of FIAF, we played Freedom Bingo, which although running for the first time ever, went pretty well — I (as the last person to win a prize) secured a copy of Girl Talk’s album, which I’ll listen to sometime in the nearish future. 2009 -- Day 2 2009 -- Day 2

Outside the conference proper, I went with some fellow student (and one professional) delegates to Da Angelo’s in Battery Point. Needless to say, it went down a treat (I was thanking my luck that we managed to get a table there at such short notice) — everyone was really happy with their meal, and other than that, the company was excellent, and was all-in-all a pretty good night.

LCA2009 Day 1 — Monday

So I’m sitting on a plane travelling out to Melbourne on Saturday (just in case I don’t get to post this until later), yesterday was the Google Party, and instead of relaxing at the Open Day, I’m getting shipped off for ACM practice (meh). So I guess I should make some attempt to wrap up the week given that I’ve made no attempt to blog this week amidst the rabid photo-taking that occurred this week…

Following is a 5-part set of posts about what I did at LCA, and the talks that I remembered going to — it’s mostly a late braindump, hopefully I remember everything of importance. Also, I don’t link to any of the talks’ video because it’s not up yet… I’ll blog with my best-of videos link once it’s available.


I spent Monday mostly in the Linux Kernel Miniconf, highlights included the opening talk (Parallel Programming: Is it hard, and if so, why? by Paul McKenney), which outlined some parallel programming models and a discussion of how parallel programming environments can be improved in the future. His talk was eclipsed by an analogy involving toilet paper — I’m not entirely sure what he was trying to get at, but he did challenge the audience to re-rolling the unravelled roll, so Rusty took up the challenge… 2009 -- Day 1

At lunch, I took some of the Sydneysiders to various computer stores in order to track down a wireless access point that would run OpenWRT. This proved to be a mistake, as they spent upwards of 30 minutes arguing over which model to buy. Never again! Luckily I managed to actually track down some lunch at La Bella Pizza in town.

Matthew Garrett’s talk “How I learned to stop worrying and love ACPI” (the Dr Strangelove reference clearly lost on some people) outlined some of the problems he’s encountered whilst getting ACPI support into the kernel. One of the main issues is the requisite Turing-complete language that ACPI must implement. I’ll let Xzibit explain: 2009 -- Day 1

Instead of seeing Rusty’s talk on Large CPUmasks, I went to Bdale’s talk on collaboration with large corporations, which was, as is usual with Bdale, quite insightful and enlightening. The conclusion of the day was the “Ask a Kernel Hacker” Panel at the Kernel Miniconf, which was interesting.

So, about that Wireless for the apartment? I took some intrepid guests to our local Officeworks store, where they picked up a Linksys WRT54G2 rapidly. I’m really glad that that one sorted itself out quickly, as it let me get onto the conference network rapidly, and actually upload my photos for the day.

LCA2009 Day 0

Yesterday, being the opening day for registration at 2009, I decided to attend the Newcomers’ session, in which Rusty Russell and Donna Benjamin detail the history of the conference to conference newbies. This was followed by drinks and nibbles at one of the local bars (the Metz, unfortunately, and became very loud before too long).

Last year’s Newbie’s session was excellent, I met many interesting people from other projects, and this didn’t happen to the same extent as it did (though I spent a fair amount of time chatting with Rusty about various things), though a few friends from Sydney made their way down independently, and I met another NCSSer (a tutor who tutored the returners at his first NCSS), so it was good to catch up with them.

The evening was spent watching the the aforementioned students from Sydney attempting to set up a wireless network. Who’d have imagined it was so difficult for people to three Uni Students to agree on which set of DHCP and DNS services to use?

No photos today sadly, hopefully I’ll be able to get some onto flickr this evening. Maybe.

New Camera (iei!)

So due to being fed up with my previous camera, I got myself a new Canon EOS 450D digital SLR camera last week, and needless to say, I’m really really quite impressed.

Snug Falls -- January 11, 2009

I’ve taken it out for a few test runs over the past week, and am really quite impressed with the results. To start out my lens collection, I’ve used some of my Dad’s old FD-mount lenses via use of an adaptor1 (including an f/3.3 200mm prime and an f/5.6 75-300mm zoom, for those of you who are at all interested), which has afforded some good opportunities for interesting photos:

UTAS Sundial

Tasman Bridge by Night

I’m really looking forward to LCA as an opportunity to give it a really thorough workout — I expect that I’ll take a ludicrous amount of photos as an opportunity to test the camera in more varied circumstances. More as they come!

[1]: Yes, I’m well aware of phographers’ disdain for using FD lenses on an EOS camera, currently I don’t care. When I get to the point where I need to, I will.

The March South

Ohgod, only 6 more sleeps until LCA2009 starts here in Hobart, and planet lca2009 has just been announced, so I suppose I’d better make a first post.

I’m really really excited (as Tom has already pointed out), and I hope all the excitement turns out to be justified (not that I have any doubt at all). There are some absolutely fantastic talks and tutorials lined up for every day of the week, which I’m really looking forward to.

Outside of the conference proper, I hope that you take as many opportunities as possible to take in Hobart’s scenery, including Mt Wellington (which is 30 mins drive to the summit from the city, and from which you can view the entire city) and Salamanca, amongst others.

I’m looking forward to meeting those of you that make the march south over the next week, and I hope you enjoy Hobart as much as I do.

Happy Arbitrarily-celebrated Public Holiday!

I notice that the most significant number on my clock has incremented (as it tends to do once every 365ish days), and hence I feel obliged to point it out to you all: Happy new year!

2009 looks like it’ll be a really exciting year for me (for the first few months of it, anyway) — I’m looking forward to (in no particular (chronological) order):

  • Not going to Sydney to tutor at NCSS, instead, filling most of the rest of my summer break doing programming competition practice (exciting!!!?!)
  • 2009 in Hobart (and the associated bonus of finally getting friends from interstate to reciprocate visits I’ve paid). Only 18 more sleeps until the first day of miniconfs kicks off — I’m thoroughly excited!
  • Starting my final semester of undergrad study (not so much the overload that I’ll be undertaking in order to actually finish my degree :()
  • Easter in Germany!
  • Competing in the World Finals of the ACM ICPC, to be held in Stockholm towards the end of April

May your 2009 also be fun, exciting and productive!