BoostCon 2009 Trip Report 1

Okay, so it’s been a while.  Re-entry from BoostCon hasn’t been quite as smooth as I’d hoped, and I probably have a few other excuses if I dig for them.  Anyway, I know some of you are waiting for a BoostCon trip report, and I’ll be trying to push out one installment every day.  Here’s the first.

Saturday, May 2, 3:30pm: On the flight in from Denver, I have the seat with the most legroom of all. It’s all the way in back, in the middle, in a row of five seats straight across the whole aisle, just like where the bad kids used to sit on the bus. The only “E” seat on the whole plane, so I have the entire aisle for my long limbs as long as nobody needs to walk around.

The de Havilland Dash 8-200 is just a regional commuter plane, but with two engines hanging beneath a top-mounted wing it gives the impression from the outside of being a big, lumbering beast. The sound of the props on the inside manages to keep up that impression despite the cramped quarters. We’re grinding through thick, low clouds on our way to Sardy Field, the airport serving Aspen, Colorado. I can’t see a thing out there. Man, flying on instruments in the mountains. I’m glad that’s not my job!

Suddenly we drop out of the low clouds and land, all at once. The forecast has been threatening rain or even snow for days, so I’m a little concerned about what kind of weather we’ll have for the conference. This doesn’t look too promising, and it’s chilly. The second I step off the plane and take a breath, though, it’s like finding my place again in the world. That familiar thin Aspen air brings a smile to my face, and I stop worrying.

See full-sized image

The approach to Aspen on a clear day

Sardy Field is a tiny airport. There are no “jetways.” Your plane stops and you climb out and down the steps, wait for a baggage handler to bring over any gate-checked bags (in my case, a guitar), and walk across the tarmac to the door. Today an unanxious “crowd” of four or five people is gathered around the baggage carousel, waiting for suitcases. Somehow I ended up with a big one again. It always seems like I need to pack extra stuff for this conference.

The conference hotel (the Aspen Meadows Resort) runs a shuttle service, but I can’t remember whether I asked them to pick me up or not, so I wait around outside the airport for a little while. Ah, well, a cab pulls up and I’m tired of waiting, so off we go.

Last year’s room at the Meadows was a little ways off from the main building, with its back facing onto the Roaring Fork River. This year I’m back in the first building, with its back facing onto the actual meadow. As usual, the accomodations are amazing, especially if you’re a fan of mid-century modern design. Every “room” is actually more like an apartment, with huge windows on either end, a small kitchen in the middle and a big living area in back where you can entertain guests without seating everyone in your bedroom. Since I have to run the conference I spend my evening settling in, putting my stuff away, and taking it easy.

SD Desolate; Boost/C++0x BoF Packed!

It’s easy to see the economy is hurting; everything here at SD West 2009 is unusually quiet.  However, last the Boost and C++0x birds-of-a-feather session we held last night was standing-room-only.  I don’t know what to make of this, other than that people think this stuff really matters.

Update: by Friday, people were looking a little burnt, but the attendance and enthusiasm were high in my sessions throughout the week.  Thursday’s talk on the C++0x support for Generic Programming (Concepts) was another standing-room-only presentation.

BoostCon 2009 is Coming…

After two years of targeting my BoostCon proposals at the things I thought other people would want to see and do, I decided to take a cue from Scott Meyers. He’s always reminding me that when I run a session, I can have it be about the things I’m interested in. The logic, I guess, is that if it’s worth doing, others will be interested too.

So this year, instead of thinking about what I can teach people, I decided I wanted to see what I could learn, and get stuff done for Boost. Doug Gregor and I decided to submit two session proposals, “Boost++0x #1: hands-on rvalue references,” and “Boost++0x #2: hands-on decltype, variadic templates, advanced SFINAE.” Just like they sound, we’re going to be doing some coding!

The basic gist is to have a very short tutorial review of one or more new C++0x feature followed by 1/2 day of work exploiting these features in a branch of Booost, in groups of 2-3, using the features in GCC >= 4.3. There are two goals:

  1. Give BoostCon attendees (me included!) a chance to play with C++0x features in real-world code
  2. Get some useful C++0x features into the Boost Subversion Repository so that Boost is better prepared when C++0x compilers become more commonplace

I’m also really excited to learn more about CMake and make some progress on integrating it and CTest into our processes. Brad King from Kitware is going to be there and has committed to making the changes in their tools that Boost needs.