OSCON 2007: Friday

OSCON Friday is much like Wednesday and Thursday, but it's a half day: keynotes, two sessions, a final keynote, and a final schmoozing session. The Friday keynotes were pretty good. The first was from Philip Rosedale of SecondLife. He gave a demonstration, which was my first time to actually see Second Life. It looked pretty cool, and he displayed some data graphs from inside the virtual word. He said that lots of people use it for telemeetings. They're beginning to offer voice features, so people will probably start using it for teleconferencing. Some of the components of Second Life are open source, and Rosedale was encouraging the community to contribute to the project.

Next was Jimmy Wales of wikia, which hopes to fulfill Wales' ambitious goal of giving everyone free access to the entire sum of human knowledge. Wales also announced wikia's acquisition of grub, Web-crawling software which lives on unused computer cycles (sort of like SETI@home, but less pointless).

Simon Wardley then spoke about information technology commoditisation. I believe his point was that open source means allowing people not to have to spend all their time building entire infrastructures from scratch, because someone else has hopefully already done it for you and has open-sourced the code. And Simon evidently really likes ducks.

Nat Torkington, OSCON program chairperson, gave a farewell keynote. He's stepping down from OSCON organization duties (and perhaps leaving O'Reilly altogether--not sure). A (paraphrased) quote from his talk was "ignore the noise and make some signal" (participate in and contribute to a cause about which you feel passionate).

The last keynote was "Pimp My Garbage" by an electronics engineer named James Larsson. Larsson like to make interesting things out of computer equipment that other people want to throw away. His creations have to be seen to be fully appreciated, and he took the notion of building a better mousetrap to a hilarious conclusion.

The first session I attended was "Subversion Worst Practices" [sic] by Ben Collins-Sussman and Brian W. Fitzpatrick, a couple of Google software engineers who moonlight as subversion developers. They gave a good list of things to avoid when implementing version control, like storing ISO images in a repository, skipping out on backups, and manually editing repository files.

I was wanting to go to "Testing With Selenium" as my second session, but it was cancelled (selenium is an in-browser unit testing framework, which can be a good way of automated javascript testing). So I went down the hall to "Running Your Programming Language on Parrot", and was quite promptly confused. shrug

So I had some time to kill Friday afternoon. I decided to go to the Chinese Garden and the Japanese Garden. I didn't much care for the Chinese Garden, which was pretty small. It was $7 and 30 minutes of my life I'll never get back. It was OK, but nothing special. It didn't seem very well-kept: the water was pretty murky, and cobwebs were everywhere.

But the Japanese Garden was awesome. It was $8, it's in Washington Park (western Portland), and it was really beautiful and peaceful. I spent a couple of hours there, and took several pictures (I posted them to flickr). I highly recommend this to anyone with a couple of hours to spend in Portland.

Here are a couple of my favorite pictures:

OSCON 2007: Thursday

Enjoyed about half the keynotes too much this morning. The first one was cool: Ben Fry of processing.org showed some data visualization animations that have to be seen to be believed. And Steve Yegge of google gave an irreverently hilarious discourse on branding, and pointed out that branding is something that open source really lacks.

My first session was "Vim for PHP Programmers" by Anerei Zmievski (although the content was not very PHP-specific). This was an excellent talk with lots of neat Vim tricks. It was way too much material for a 45-minute presentation, but he made his slides available, so I'll be able to go through them later. As a Vim noob, I'm quite keen on learning tricks like easily wiping out the text between paired curly braces (or angle brackets or parentheses).

In a couple of the sessions I've attended this week, including today's "Untangling the Web: Dealing with Legacy PHP Code" by Clinton Nixon, people have expressed significant disdain for templating engines in PHP. I've used smarty for a couple of years and really like it. So I guess I don't really understand this sentiment. shrug

"Cache That" by Gopal Vijayaraghavan talked about APC, an opcode and data cache for PHP (Gopal is the lead maintainer of APC). This was pretty interesting, and he talked about some subleties of using APC.

My favorite talk of the day was "PHP Addons for Fun and Profit". (I didn't get the speaker's name, but she said that her blog is coffee.geek.nz). This talk was incredibly informative, with lots of good stuff about memcache and eaccelerator. I'm looking forward to giving memcache a try.

OSCON 2007: Wednesday

Wednesday and Thursday of OSCON are made up of 40-minute sessions after a morning of keynotes. This morning's keynotes started with Tim O'Reilly himself. He talked about how successful Web sites these days are more about user data than they are about the software driving the sites. He listed flickr and del.icio.us as examples, and he listed a few new sites, as well:

1. freebase: "a structured, searchable, writeable and editable database" of just about anything/everything
2. dabble DB: looks like an impressive online spreadsheet service
3. open ads: "the web's largest ad-space community"
4. hadoop (an Apache project): "a framework for running applications on large clusters of commodity hardware"

Tim also mentioned that StumbleUpon had been purchased by eBay (I didn't know that).

Next were a couple of guys from Intel: James Reinders (a suit) and Dirk Hohndel (looked like he belonged on a snowboard in the Swiss Alps). They talked about the new Intel Threading Building Blocks, a C++ parallelism library (you'd use this library to write software to take advantage of a multi-core architecture). Intel has open-sourced this technology, and there's even an O'Reilly book about it. Hohndel also mentioned moblin.org: Linux for Intel-based devices.

Next up was Simon Peyton-Jones, a researcher who talked about concurrent programming. He said that one approach to addressing the challenges of concurrent programming is to wrap code in a database-like transaction (with a transaction log) to acheive atomicity (like a journaling filesystem, I guess). At that point my eyes started glazing over, and I became hypnotized by his bright red sweater with a picture of a red-eyed treefrog on the chest. (It was really a very interesting keynote, and he's a good speaker--I just didn't understand a whole lot of it.)

Then Tim O'Reilly interviewed Mark Shuttleworth of Canonical. As I'm not a passenger on the Ubuntu bandwagon, and was still recovering from the amphibious atomicity assault, I went looking for coffee or something.

During the course of the day, I attended a handful of sessions, including a couple about nagios (a host- and service-monitoring system) and APD (for profiling PHP code). I enjoyed David Verba's "Practical Design For Web Developers", a discussion of user-centered design (whose 'further reading' bibliography included several interesting-looking books). Perrin Harkins' "Care and Feeding of Large Web Applications" was also pretty cool: he talked about the challenges of devoloping, maintaining, and distributing an enormous Perl-based codebase (the Arcos CMS/CRM).

But my favorite of the day (and maybe of the whole conference) was Joseph Smarr's High Performance JavaScript. I highly recommend the slides from his talk about the development process of the Plaxo online calendar and address book synchronizer.

OSCON 2007: Tuesday

This morning's tutorial was "PHP and MySQL Best Practices" by Luke Welling and Laura Thomson. I've heard her speak previously at OSCON, and she's a good speaker. And this morning's talk was also good, but it turned out to be mostly review for me, so maybe I should have signed up for something else. But I did pick up a few interesting things that I'll want to research further:

• test_more (which is like Perl's Test::More) for unit testing
• PEAR's quickforms2 for form processing
• PDO might be a good alternative to the PEAR DB in PHP5, but I should probably consider just using the mysqli interface, since I don't really need to be writing my applications to support multiple backend DB engines
• xdebug looks interesting for profiling
• if I need to develop in a load-balanced environment, I'll need to move away from storing session data in flat files on the Web server--encrypting the session data and putting it in a browser cookie might be an interesting alternative to storing sessions in a DB
• I need to look at memchached again
• mysqlperformanceblog.com looks worth following

Anyway, they had a couple of copies of their book, PHP and MySQL Web Development (3rd ed.), to give away. I asked a couple of questions during the talk, so I scored a free book. Woo-hoo!

The afternoon session was "How To Be a Better Programmer" by Michael Schwern. I've been to Michael's talks before, and he's a bit different (and opinionated). He appeared to have three non-OSCON groupies today, which was a little odd. The first half of his talk turned out to be mostly about soft skills (like getting along with people). While those are certainly useful and important skills, I guess it's not really what I was looking for. Oh, well.

It picked up a bit after the break, and Michael talked about things like version control (specifically SVN and SVK) and unit testing.

A couple of the more interesting (to me) points from Michael's talk:

• he suggested learning a radically different programming language, and then bringing that language's best features back to your 'native' language
• hiveminder.com looks interesting for task management

Then there was the evening event.

Seems like Tuesday night is typically a fairly well-organized affair, with several speakers and a pretty interesting program. This year, not so much. Instead of setting up the ballroom with chairs lined up seminar-style (as in years previous), the chairs were arranged around lots of tables, and we were made to go from table to table meeting people. Like speed dating.

There was a musical guest who, to his credit, sang and played guitar fairly well. But he sang about man pages. Yes, man pages. It was supposed to be funny, but no one was really laughing.

There was the annual open source award ceremony, in which PJ of Groklaw (who, disappointingly, was not present) and Paul Vixie (of BIND and cron fame) received some well-deserved recognition. But lots of people were talking, rather than paying attention, so it was difficult to hear the people on stage.

And there were no other speakers. So the evening event was pretty lame. I left early. At least they served booze, which took the edge off, a bit.

OSCON 2007: Monday

Monday and Tuesday of OSCON are made up of tutorials during the daytime (there are also some evening events). A tutorial is an in-depth presentation of a specific topic, and it lasts 3-to-4 hours.

This morning I went to "Advanced Vim Scripting", for which the speaker was Steve Oualline (it was originally going to be Damian Conway, but they changed speakers a couple of weeks ago--dunno why). It was a good talk. I'm sort of new to vim (I've used joe for about ten years). Here are a few lines I've added to my .vimrc file today, as a result of the tutorial:

iab #p #/usr/bin/perl -w
iab teh the
map \1 :s/^/#/<CR>
map \2 :s/^#//<CR>
map \3 :%!perltidy<CR>
map \4 G:r!svn diff<CR>:sp<CR>gg<C-W><C-W>gg

The first two lines define a couple of abbreviations: if I type #p, it'll be replaced by #/usr/bin/perl -w. And I think I probably misspell the more often that I spell it correctly, so the second one will probably help a lot.

The first three map items allow me to do some tricks I've previously talked about doing in joe. The first two comment or uncomment a region (determined in visual mode), and the third runs perltidy on the whole file. The fourth is useful if you have vim in your VISUAL or SVNEDITOR envariables. When you type svn commit, svn opens vim so that you can record a commit message. If you type \4, the macro sends inserts the output of svn diff at the end of the file, splits the screen, sends each half to the top of the file, and leaves the cursor in the lower half. So you're ready to page down in the lower half to review the changes while recording the commit message in the upper half.

The afternoon tutorial was "Linux Performance Monitoring" with Darren Hoch. This was very interesting to me, as it covered lots of diagnostic techniques for troubleshooting server performance problems. Darren covered vmstat, iostat, mpstat, ps and a few other programs. He pointed out that it's a good idea to run these things when the system is quiet and when it's under load, and to archive this information. This gives good baseline data. In fact, I'm thinking about setting up some cron jobs to do this sort of thing on a regular basis, and maybe doing some trend analysis.

Darren also talked about iptraf, and this may be a good way to track the amount of traffic coming in and going out on various tcp and upd ports. I'd previously tried a very complicated process involving user-defined iptables chains and parsing the output of iptables -L -v. But I may be able to get comparable data by running iptraf for a few minutes at the top of every hour or something.

2007 OSCON: hotel Internet fee

So I asked at OSCON registration about the hotel Internet fee. I talked to an incredibly hot woman with a nosering, and she said that, alas, it was no mistake.

She said that Portland had recently launched citywide wireless, but that the hotel was blocking it. Charming. Don't know if I need to complain to the hotel, the conference, or both, but I think someone probably needs a wee bit of wrath.

Well, my work (which is paying for this trip) will probably pay for the Internet fee, as I'll likely do some work this week. So I guess I'm still sticking it to 'the man', but I sort of feel like I'm sticking it to the wrong man. shrug

And I stopped by the Ubuntu Live party looking for the Linux Action Show guys, but didn't see them. Oh, well.

OSCON 2007: arrival

Left home this morning to fly to Portland, OR, for the 2007 OSCON. This is my fourth year to attend, and I always have a really good time at this conference.

So far, I'm a little disappointed in the hotel (the Doubletree). In past years, they've waived the $10/day in-room Internet fee, but they say they're not doing that this year. I'm hoping that's a mistake. I'll ask about it when I go check in at the conference in about an hour.

The guys from the Linux Action Show are having a get-together tonight (they're here for Ubuntu Live), and I was planning to join them. Unfortunately, they've picked a place which looks a bit too far to walk (I didn't bother renting a car), so I don't know if I'll do that or not. Looks like they'll be at the Fun, Food, and Drink event, so maybe I'll try crashing that.