Sponsored Link •
A narrative account of my week at PyCon in Washington DC
This was my first time attending a Python conference. Many thanks to the Python Software Foundation, Stephan Deibel of Wing IDE, and all those who helped and contributed to get me there for the Docutils sprint and for the conference itself. It was a lot of fun, a lot was learned, and a lot was accomplished.
I kept an account of my week at PyCon DC 2004. Enjoy!
6:30am -- Left home after saying goodbye to my wife & kids. I'll miss them; hope they miss me too!
7am -- Arrived at Pierre Elliott Trudeau (formerly Dorval) International Airport. Got my boarding pass from a kiosk with my credit card. Neat. Checked one bag, but the new security requires me to carry it through US customs & immigration. Then, because my destination was Reagan National Airport, a security guard pawed through my checked bag and sent it through X-ray. Fine.
But when I got to the gate, I had to undergo another search with a hand-held metal detector, including my shoes. My carry-on, jacket, and wallet were searched too.
I feel safe now.
10:30am -- Arrived in DC after an uneventful flight. Took the metro in to Dupont Circle, but I should've gone one further -- long walk. In the metro, I kept hearing the PA system saying "George Clinton". I finally realized I was misunderstanding "doors closing" through a poor speaker system.
Checked in at the Washington International Student Center (WISC) by a Japanese guy named Tomo, with whom I had a nice conversation. WISC is a hostel, very no-frills. Each room has 3 sets of triple bunk beds, everything needs a paint job, but the bathrooms are clean. Can't beat $22/night, breakfast included (toast, jelly, and cheap "citrus punch").
12:30pm -- Made my bed, dropped off my stuff, and headed in to the Smithsonian museums. Tomo noted that pattern: whereas Americans usually wanted to see the White House first, Canadians tended toward the Smithsonian. I did the geekiest two: Air & Space and Natural History. Apollo capsule, moon lander, Saturn V engines -- very cool. Hope diamond, crystals and mineral formations -- amazingly beautiful.
I consider this an "advance scout" mission for the day when my whole family visits. That happened once before, when I went to France on business and had a free day in Paris. Saw all the major sights -- from the outside. Didn't have time to wait in line to get in.
I was a bit disappointed that the Reflecting Pool was empty. I wanted to do the "Forrest?" "Jenny?" "Forrest!" "Jenny!" scene or at least imagine it, but no luck.
The new World War II Memorial, around the Rainbow Pool, looks quite impressive. Almost ready to open.
The Lincoln Memorial was most impressive. I'd never read the entire Gettysburg Address, so it was moving to read it carved into the wall.
Next, I visited the Korean War Veterans Memorial -- very moving. Especially reading these words carved into the wall: "Freedom Is Not Free".
The Vietnam Veterans Memorial was particularly moving. I know no one who fought in Vietnam or lost a relative or friend there, but still I felt tears welling up. The names just go on and on. What a waste of life, on both sides. War has no winners, only losers, but the price must be paid from time to time.
Back to WISC, dropped off the Smithsonian gifts for the wife & kids. Then off to Best Buy to check out DVDs. I found a few that were good deals, considering the exchange rate. But only a few. Most had the same price point in US dollars as in Canadian, favoring the Canadian consumer.
10pm -- Back at WISC now. Had no dinner, but don't feel hungry. Had a Polish sausage between Smithsonian museums around 2:30.
I'll be heading for bed soon. I don't have an alarm clock or watch for the morning, but I should be able to wake up in time.
I noticed that DC is a jaywalking town, much like Montreal. Warms my heart.
Woke up around 6:20 without an alarm. Pretty good, considering that last night, somebody started a party nearby with loud, booming reggae/dance music. Earplugs to the rescue.
Showered, changed and ate -- hostel-provided toast & jelly and fake orange juice. Brushed my teeth and set off for George Washington University.
The city is very quiet at 7:20am on a Saturday. The walk took me about 35 minutes, over 3 kilometers (2 miles). It's warmer today; I could open my jacket and didn't really need my gloves or hat.
Arrived at the Cafritz Conference Center just before 8am. A few people were already there -- Michael McLay, Ted Leung, Brian Dorsey, and Bill Sconce. I helped set up the tables, extension cords, and network cables. People trickled in over the morning, becoming a good-size group. I met the (former) PythonLabs crew: Barry Warsaw, Fred Drake, Guido van Rossum, Jeremy Hylton, and Tim Peters.
The Docutils sprint began with Oliver Rutherfurd and Matt Gilbert. I did a walk-through of the architecture & the code, beginning with PEP 258's block diagram. Soon more people arrived though, including Aahz, Fred Drake, Bill Sconce, Steve Holden (our most esteemed conference organizer), and Ian Bicking. Oliver began working on his DocBook writer, Matt & Fred on MoinMoin integration, Steve & Bill on a DocPy/LaTeX writer, Aahz on a FrameMaker writer and generic stylesheet ideas, and Ian on the Python Source reader. Andrew Kuchling joined us in the afternoon to write a directive for the ABC music language. We transferred lots of knowledge and made some progress on the code.
Around 1pm lunch discussions began. Aahz forced the issue by standing up and saying something like, "Anybody who wants lunch stand up; let's go." We ate at a Thai place on Pennsylvania Avenue. Very nice. Then back to the sprint.
Although I own a laptop, I didn't bring it to the conference for several reasons. My wife is working on a project and needs it; I didn't want to have to worry about breaking it or having it stolen; and I figured as sprint coach I ought not to be concentrating on my own code, but on the sprinters'. That last reason turned out to be the most significant. My lack of computer meant that I was an accessible consultant to all. I'd recommend this strategy to future sprint coaches.
Sprinters started taking their leave in the afternoon. Ian and I were the last in our sprint room; we left at 6:30pm, asking the 15 or so people in the other room to have building management lock the doors.
Ian and I took a walk through the city, up to Adams-Morgan, where I'm staying. We had dinner at a Middle Eastern restaurant at the corner of 18th and Florida -- not bad at all, spicy & cheap. We walked through the Saturday evening crowds, but neither of us being clubbers, did not join in. I left him at the Woodley Park-Zoo metro station, with its long, looooooong escalators. I returned to WISC, taking the scening route, and found a good deal at the local Blockbuster store.
Back at WISC now, writing this up. Think I'll turn in early tonight.
Woke around 7:20. Loud music again last night, but earplugs did the trick. A brisk, windy morning, an a 30 minute walk to PyCon. On the way, I noticed a building being build at the corner of 21st and K (I think). Affixed was a sign: "Hitt Contracting". Amusing.
Docutils sprint: Ed Loper has arrived this morning, to work on Epydoc integration. Tracy Ruggles worked with him. Laura Creighton and Jacob Hallen joined us, but pressing business seems to have arisen. Reggie Dugard worked on a Table of Contents bug, mostly as a learning exercise. Matt Gilbert made good progress on MoinMoin integration, as did Aahz with his FrameMaker writer, Oliver Rutherfurd with his DocBook writer, and Steve Holden & Bill Sconce with DocPy/LaTeX.
For lunch, Matt, Tracy, Ed, Reggie, and I went to an Italian place across the street. Was good, and filling; became my main meal of the day.
I was the last one out, after asking the building manager to lock up our room. Bought a watch at CVS (the pharmacy, not the code repository) on the way back to the hostel, for the alarm.
The watch's alarm didn't wake me up, possibly because I was wearing it, and I was bundled up under the covers. I woke up a few minutes later anyway.
Was cold this morning, so I was glad for my sweater, scarf, hat & gloves. A lot more traffic on Monday morning than Saturday & Sunday.
Mike Orr joined the Docutils sprint for the morning; he and Reggie began working on an API for document fragments, particularly HTML. By evening, Reggie Dugard had made good progress, almost complete. Aahz seems to be making steady strides with his FrameMaker writer, and Matt Gilbert with MoinMoin integration. Ian Bicking was back with us after attending the Zope tutorial yesterday, but Oliver Rutherfurd is at work today. Ed Loper and Bill Sconce also continued hacking on DocPy/LaTeX; Steve Holden is now busy with conference matters.
For lunch, Reggie and I visited the conference building's food court, and impressive place with a dozen or so vendors, from Burger King and Taco Bell to Crepes and Chinese. We settled on Chinese; I had sweet & hot chicken on rice. We got big helpings; again I wasn't hungry for dinner. There were some kind of peppers in the food and I ate one by accident. Hot!
Met two fellow PyConeers at the hostel at breakfast: Tamer Fahmy of Austria, and Andrew Gaul of the USA. We walked down to GWU together. When I introduced myself, I was asked my area of interest. I answered "Docutils," and Andrew said something like, "Oh, right! You're famous!" In a very small circle, perhaps.
Today was the last day of the Docutils sprint. As on each of the previous days, I did an architecture & code walk-through and then hacking continued. In addition to hacking, we formed a human assembly line and stuffed tote bags. Most of the sprinters present registered and received their totes and T-shirts. As a PyCon volunteer, I received a second, black "Event Staff" shirt. Good thing, too; I'd planned it that way. Brought only 6 shirts for an 8-night stay.
Lloyd Kvam joined the Docutils sprint today, and worked on The Docutils Document Tree document. (How many times can you say "doc" in a single sentence?) Reggie completed the fragment/publish_parts API. Matt, Ed & Bill, Aahz, and Ian all made progress. It's been a fun, productive four days.
Laura Creighton offered some advice. She and Jacob Hallen hadn't been able to devote a lot of time to the sprint, but would have had there been a collection of smaller tasks (bug fixes, tests, docs, etc.) prepared beforehand. Valid point; I'll have to take that into account if and when I do another sprint.
Around 7pm, Laura, Jacob, Armin Rigo, and I headed out to Georgetown for dinner at Makoto, a tiny Japanese restaurant. Laura's treat. It turns out that back when she wrote me several times asking for reST help, it was regarding the PyPy spec draft she was writing, and she felt she owed me a meal. The meal was excellent, 8 or 10 courses (lost count): sashimi, sushi, yakimono, soup, on and on.
We took a bus back, and I transferred up to Adams-Morgan. Gotta get up early tomorrow -- manning the registration desk at 8am.
Woke up around 5am, tried to get another precious hour of sleep, but no luck. Met another PyCon-goer, Nestor Nissen of Germany, when I was on my way out of the bathroom after brushing my teeth, and he was on his way in. I was wearing my black staff PyCon shirt.
Arrived at PyCon at 7:15am, ran around setting up schedules and helping out wherever I could. I manned the registration desk from 8 to 9; met a lot of people. Bill Sconce had his stuffie python, Eric, on the registration table, so I made a name badge for him. Was wearing my name badge with "Docutils" plus a "FOR HIRE" badge, which got some response. Talked to a recruiter after Mitch Kapor's keynote address.
Anthony Baxter's talk, "Scripting Language" My Arse: Using Python for Voice over IP, was a lot of fun. Attended Andrew Koenig's talk & Andrew Kuchling's Quixote tutorial, both good.
Lunch was fine, met Thomas Wouters, Neil Norwitz, Neil Schemenauer, and others. Made a BOF session schedule board, then went to buy some socks at Gap. Caught Jacob Hallen's PyPy talk; interesting work.
Did a bit more sprinting, then held the Docutils Birds-of-a-Feather session at 8pm. About 20 people showed up. I described the results from the sprint, then opened the floor for discussion. Got some good ideas too.
Walked back to the hostel with Tamer Fahmy, talking about middle-eastern politics among other subjects. He worked on his slide show for Thursday's session; Nester arrived and we critiqued. My advice: fewer words, more talking, and show the pretty pictures as early as possible. We seem to have convinced him.
Met an Italian couple as I was getting ready for bed, who couldn't speak much English. Tried French and Japanese, but their "little English" was all we had in common. I bid them "buena notte".
Woke at 6am, again an hour before my alarm was set to go off. Must be eager. Arrived at PyCon early again, and had breakfast there. PyCon's sticky buns & muffins beat the hostel's toast & jelly any day.
The day began with Guido's keynote speech, which was very entertaining and light. A phrase Guido used struck a chord: "active champion". The reason most (if not all) proposals/PEPs/projects fail is for lack of an active champion. I was an active champion for reStructuredText and Docutils.
Next up were some PyRex talks by Paul Prescod. I wonder if his choice of "shrubbery" as example was deliberate. I got enough out of the talk to know that PyRex is cool, and if I ever need to code C extensions it may be the way to go. But I don't now, so my attention wandered.
Met a couple of people from last night's BoF, who had ideas for reST: a one-page reST cheat sheet, and YAML support, perhaps with flowing columns to save on vertical real estate.
Was joined by Vic Kelson for lunch, who had questions about a database-driven documentation system with boilerplate and custom sections. Feasible, for sure. He suggested a possible line for a business: open source industrial control data acquisition systems. Evidently the industry standarc, SCADA, sucks -- can't extract data from it easily.
After lunch, saw part of the new-style class tutorial, and the tail end of the Panda3D talk. They had a video of a virtual reality system that makes Happy Tree Friends seem tame. Next was Tamer Fahmy's High-Level 3D Graphics talk, which went well.
I chaired the last session of the day, 3 talks: Ed Loper's Epydoc: an API documentation generation tool, Jim Fulton's Literate unit testing: Unit Testing with Doctest, and Phil Pfeiffer's Twelve Thousand Test Cases and Counting.
Immediately afterward was the PSF member meeting. We had committee & treasurer reports, then voted on new Sponsor & Nominated members. Some discussions followed, about various aspects of communication. We need better, more focused communication (not necessarily more volume, though). Roundup's nosy lists seem like the might be the solution.
15 of the PSF people went to a nearby Chinese restaurant for dinner. The conversation was lively and mostly non-technical, at least at my end of the table. Back to the PyCon site for a late-night Py-dot-org BoF. I volunteered myself and Brett Cannon to add encoding support to ht2html. The group came to a tentative consensus about a website redesign, or at least a new design for a new portion of the website.
Back in Adams-Morgan, Thursday evening seems to be a lively night. Or it might just be the exceptionally warm weather. Hopefully no booming music tonight, but the earplugs will be on stand-by.
As I finished writing the above in the hostel's kitchen, two young women walked in, speaking Japanese. I shocked them by greeting them in fluid, if not fluent, Japanese. Turns out they're exchange students living in Lawrence, Kansas, visiting New York City and DC during spring break. We had a nice conversation, soon joined by Tamer and Nester. I left the unattached young'uns and headed off to bed.
Today's keynote address was by Bruce Eckel, of "Thinking in..." fame. Entertaining but a bit technical and monotonous. Anna Ravenscroft's Distutils intro talk was very good -- ought to be repeated next time.
After lunch, I discussed some Docutils & code metadata issues with Terence Way & Brad Clement, then with Eric Jones. The plenary session concluded the conference, and I went off for a bit of sight-seeing.
At the beginning of the week, Saturday/Sunday, I was wearing a sweater, leather jacket, scarf, hat, and gloves. By Thursday, the morning required a light jacket but the evenings were warm enough for short-sleeved shirts. Spring has come to DC!
Walked past the south lawn of the White House, where a large crowd of tourists were gathered taking pictures. Yawn. Stopped by the Smithsonian's Natural History museum again to see if I could find a relative of Eric the python, but no luck. They had cobras and rattlesnakes, but no pythons.
Walked down 14th street to the Tidal Basin, which was surrounded by cherry trees, beginning to bloom. The Jefferson Memorial was very impressive, solemn, and peaceful. Continuing on, the FDR Memorial was another worthwhile site to visit, although I went through it accidentally (it wasn't on my free tourist map) and the wrong way, reverse chronologically. Its fountains reminded me of the "emergence" scene near the end of Logan's Run, but the FDR Memorial was built in the '90s. Perhaps the inspiration was the other way around?
Heading back north, I passed by the conference site so I popped in to see if anyone was still there. Spoke with Jeff Kowalczyk again, who was in the amphitheater taking advantage of the wired internet access. The Twisted people were still going strong. Some people were out on the terrace enjoying the warm weather, including Mike Fletcher, Thomas Wouters, Anthony Baxter, Sean Reifschneider, and Sylvia Candelaria de Ram. Their Thai food arrived, but I had other plans: dessert at TGIFriday's next door. My wife and I had gone to TGIFriday's in Honolulu once and had a great time. Excuse me, my brownie delight has arrived.
It was raining this morning. Packed up and left the hostel at 9am, took a bus down to a metro station then on to Pentagon City. I'm not sure, but I think the "Fashion Center" mall there may have been the mall in Minority Report, and possibly also in No Way Out. Bought a "PFC" -- portable folding cart -- for my bags because I was getting tired and sore carrying them around. A couple more wee gifties for the kids. Visited the Bang & Olufsen store, and fantasized.
On the opposite side of the street -- reachable via a metro underpass -- was another mall, not so upscale. Browsed around a Borders book/CD/DVD/coffee store, and a Best Buy. Then it was time to get to the airport, 2 stops away.
Got my boarding pass, and two last-minute gifts: a nice bookmark for my wife, and a black & yellow "jungle carpet python" plushie, for my kids. And for me, too, I guess -- I'll name it "Monty" and it'll be a cousin to Eric, the conference mascot brought by Bill Sconce. A Cinnabon later, and it was time to get to the gate. Spent my last 90 US cents on some mints -- no currency to carry home.
On takeoff I had a good view of the Pentagon, and I see the need for so much security. The airport is literally a stone's throw away, and a last-minute course correction would be (and was) deadly. I wonder if security is the reason why, on both arrival in and departure from DC, the plane wasn't brought all the way to the jetway. Passengers had to go down a flight of stairs to the tarmac and back up.
As I write this on the plane, I look back on the last week as an unmitigated success. I met loads of people I'd only ever exchanged email with before, and plenty others new to me. It's amazing the places my software has gotten to -- rumored to be in use at NASA, the NSA, the OSAF, and I'm sure at many other companies and organizations large and small. I guess Docutils and reStructuredText have hit a sweet spot.
Warms my heart.
I'm looking for work, as a programmer, sys admin, consultant, or related position. My resume can be found on my web site, http://starship.python.net/~goodger.
Have an opinion? Be the first to post a comment about this weblog entry.
|David Goodger has been using Python since 1998, and began working on reStructuredText and Docutils in 2000. A proud Canadian, he lived in Japan for 7 years, where a stint at a document processing company in Tokyo began his love/hate relationship with structured markup. David is a Python Enhancement Proposal (PEP) Editor and a member of the Python Software Foundation. He currently lives outside of Montreal, Quebec, with his Japanese wife and their two children.|