You can see the notes by visting the site and clicking on the archives. Unfortunately I can’t do permalinks very readily, so I might export and clean it up later.

The last reason I didn’t post my notes here is due to my blog being in need of spring cleaning. I need to switch to a cleaner, less busy theme, so might pick one up from WooThemes. I’ll be taking another look at that later tonight, as well as merging my categories.

In place of a day three report I offer a link to Ben Kane’s summary of the conference. It turns out that not only did Richard Stallman give the same talk as last year, he gave the same talk at the University of Toronto a few days later, and still hasn’t bothered to look up any of the figures that he prefixes with “and I think that..” You may as well just read Joey deVilla’s notes. While you’re at it, read how as a Microsoft evangelist he subsidized the Free Software Foundation. I guess the hundred bucks was worth the thousands of dollars in publicity. Clever move.

I started out the day with James Golick‘s tutorial on dynamic languages. Unfortunately he couldn’t get the projector working due to issues with his MacBook. Since it was a tutorial the projector was essential, so after a several attempts it was cancelled and people shuffled to the Facebook tutorial. In my web programming course I had already learned about the API so I took the opportunity to get lunch. I spent too much time out there and didn’t return in time for the next session.

I returned in time for Avi Bryant‘s talk. Quote of the presentation, paraphrased: “… and if you find Montréal cold, you haven’t been drinking enough. With brings me to career advice…” He taught us that bad programmers copy, good programmers steal. That is, take advantage of all of the research that we as computers science students have access to. Implement bleeding edge algorithms. And for God’s sake, don’t write it for developers, write it for users. He suggests that you read proggit and Lambda the Ultimate.

After that came Giles Bowkett‘s keynote. He’s a great speaker and kept the crowd laughing between bursts of profane insight. Unfortunately I couldn’t take notes, his 400 slides were pretty rapid fire, but it was recorded and will be posted. Until then you can watch his previous presentation at RubyFringe. The main thing I took away: Learn everything you can, even if it’s not directly useful it can expand your thinking. Don’t get bogged down in a large corporation if you don’t want to. Also, your small business is not your career, it’s part of your career.

Later on, we checked out La Banquise for poutine. I had a poutine kamikaze. The selection was amazing, they did not skimp on ingredients, and ingredients always consisted of far more than “gravy, fries, dairy product.” I had a forkful of cheese curd the size of a baby’s fist. The portions were gigantic, I couldn’t quite finish it. On the way home I was mystified by the selection of alcohol available in the local Mac’s.

I settled for the night splitting a twelve pack and trying to use the internet. Best Western only allows us 250mb per suite of four people, which as you can imagine does not go far in a room of software engineers. We’ve exceeded our quota every day, and last night we asked them to reset our quota. They did, sort of, by giving us a new wifi key with 100mb and telling us that it will not be renewed. We’re going to run out of wifi before the end of the today I expect. Very disappointing. The cost of transfer is so cheap, especially in Montréal compared to Toronto. I can’t believe Best Western is so stingy. They won’t even let us buy extra quota. They had downtime for most of the first day too. The average Motel 8 had better service. Maybe I ought to stay at one of them for my next conference.

The first presentation was a pep talk given by Leah Culver. She explained how software engineers can thrive under pressure with perseverance and creativity. She extolled the benefits of open source, explaining how useful it can be towards getting a job. She snuck in a jab at RoR at the last minute that got the audience grinning. Her answers in the questions period were at least as well received. Her formula for success is already publicly available and open sourced on github. She inspired a drinking game and we might have already seen its first playthrough. Towards the end of the question period, when asked how she handles all the attention she gets, she unwisely admitted in a room full of software engineers that she loves boys. Thank you Leah Culver, now we know how to be awesome! I hope that you manage to find some peace and can enjoy the rest of the conference!

I already posted my notes for the Radialpoint talk. The most important thing that I took away is that every software engineer ought to have a 1.0. That is, work on something new and unproven, and it’s best to do it early on when you can absorb risk much more easily. You won’t have a family to starve by working at a startup when you’re a fresh grad. I was surprised when I found out that Radialpoint used to be known as Zero Knowledge Systems after the cryptographic proof. I recalled Austin Hill’s blog and sure enough the two are related. Marty of Radialpoint was the first employee of the company that Austin Hill co-founded. Austin shared his endorsement.

The last talk that I attended was about the Lively kernel presented by Dan Ingalls. The website does a much better job of explaining what it is than I could this long after the presentation. After the talk I asked him my two questions. It turns out that there has been work towards a translation layer between Qt and the Lively kernel! Later on I’ll post pictures and dig through the mailing list for more details. Right now I’m too tired.

In addition to reading up on the Lively kernel, posting pictures and doing CSC 324 homework, tomorrow I intend to check out some of the recommended local cuisine. Good night.

The answer to that was yes. Dan explained that there was effort to build a layer for Qt and there was talk of building one for GTK. He wasn’t sure whether it was publcly available but suggested I sign up for the mailing list.

That answer made my next question partly moot. I asked why pick Lively over another web OS such as Eye OS. Dan explained he didn’t know about Eye OS in particular but offered the phrase “self sufficiency” and how the code could modify itself. Interesting. I didn’t see the examples he used in the demonstration, but it’s worth digging into.

Marty is the speaker. He’s from Radialpoint formerly known as Zero Knowledge Systems, was employee #1 and helped build up from startup to medium sized enterprise. Zero Knowledge based off the crypto proof.

Learning is essentially unlimited at the beginning of the career.

Innovator’s dilemma: the more successful a company is, the more weighed down it is by its own success. For example, Microsoft exists because IBM was too distracted with enterprise customers to worry about a personal OS.

Every engineer should have the opportunity to create a 1.0, create something original. Get working on that 1.0 as early into your career as possible.

Takes an awfully long time to go from an idea to a product.

Innovator requires determination.

Skunkworks departments: Big companies organize development into two departments. Maintaining a product that’s already into market. And creating a new product for products and concepts that are allowed to fail because they’re on the edge.

Or, go venture capitalist, it may be cheaper than having your own R&D department. The drawback is it’s really hard to integrate companies when you buy then, especially when you have little history of innovation and pick up a startup. Most talent will jump ship.

Or, manage a company differently at different times. This is like Radialpoint. See a saturation point for certain products, when it happens to core products, switch back to startup mode.

Cost of being on the edge: If it doesn’t bring in revenue at such and such a frequency, that arm will be cut off. This is what happens now during the recession. Doesn’t usually allow for internal transfers.

Microsoft is tied up and gives Google a crack at market leadership now. Take a look at Android versus Windows Mobile.

Strengths of recent graduate: Have an ability to absorb risk. Family won’t starve if you take a pay cut to create a startup. Much closer to the edge than you’ll ever be again. You _will_ be out of the loop eventually.

Questions!

If I wanted to be an intern at a security company, what are you looking for?
Not domain expertise. General knowledge about networking and security is fantastic. Read the Bruce Schnier books. Applied Crypto is very dense, but check out his other books. Principles of security are all in there. Design principles. Never give up on the technical questions. They never expect you to get an elegant solution, they want to see people work and create a hack, then refine it. Don’t worry so much about domain expertise at the intern level, it’s a general interest in software development. Ask yourself “Would I do it for fun or for free?”

Austin Hill co-founded Zero Knowledge Systems!