A few days with the JAPHs has reminded me of a few things:
- I’m behind. I need to do a better job of keeping up with Perl things.
- Perl has a unique culture and community that is both interesting and valuable to what it produces.
- I’m so glad that the people I usually hang out with do not use a certain expletive as their favorite adjective.
Okay, so let’s review.
Sunday
I traveled to Columbus, Ohio on Sunday. My travel was mostly uneventful, other than an hour delay in Dallas to replace the brakes on the plane. Because of that, I didn’t get to have dinner with Terri’s cousins, just her uncle. However, I enjoyed a delicious meatloaf and lemonade with Keith at the Cap City Diner (warning the web site plays music). I arrived at The Blackwell that evening, which is apparently named after a professor at OSU who has been convicted of insider trading.
My first impression of Ohio State campus is that they must have had a beautiful football stadium at one time, but now appears to be just a modern monstrosity with a beautiful arch on the end facing The Blackwell. I suppose this stadium will soon be featuring regular visits from our northern neighbors in Nebraska… I’m not an OSU fan, but I hope OSU beats NU every time they play here.
While traveling and late into Sunday evening, I finished up my notes for my first talk.
Monday
I met my colleagues in the lobby and we walked together over to the Ohio Union, where the conference was being held. The Union is knew and was a very nice place to conference, particular one priced at the value that YAPC is—though, my conference registration was paid for by the low-low price of giving two talks, which is great since I was just talking about two subjects I like to talk to others about anyway.
The keynote on Monday was given by Jesse Vincent, who was the Perl5 v12 pumpking, which has been released and is well into it’s maintenance cycle now. For those that may not know, this marks the second major release of Perl in the last 3 years and is part of a new effort to consistently make a major release of the language every year. This is good news for at least a couple reasons.
- Perl language developing will stay fresher with new features coming regularly.
- It will help dispel the message of the naysayers and show that Perl is Alive.
After that, I went to a couple talks on Catalyst that I didn’t pay much attention to. This had far less to do with the quality of the talks and more to do with me thinking about my talk.
I gave my first talk at 10:00 am on Telecommuting. I will post the audio/video when Krishna gets it online. I may post the slides too, but there’s really not much to show in them. Most of it was me telling a few stories from note cards. I feel like I was a bit dull, but I got some good questions at the end, which is usually a good sign. As I said on Twitter, I hope it was useful.
After breathing a sigh of relief I spent most of the rest of the day in talks and working on a project for work. I attended Util’s talk on Amazon EC2. I learned that EC2 hasn’t changed significantly since I used it. It’s still the same, though there are more instance configuration options, Windows instances, and different storage engines. This is mostly just an extension of what was there when I was last using it a couple years ago.
We then grabbed some lunch from The Flying Pizza, which was cheap and greasy and pretty good.
After lunch, I ended up in Moose for Managers, which wasn’t a stand-out talk in my mind since I can’t remember anything about it. However, I was working on something related to work at the time, so that’s probably just me not paying close enough attention. I stayed in the Grant Street Room to listen to Paul Fenwick’s talk on Awesome things you’ve missed in Perl. Paul is a great speaker and showed a number of things I’ve put on my todo list to investigate.
I ended the regular talks for the day by attending Dave Rolsky’s talk on Fey and Fey::ORM, which I’m mildly curious about since I once tried to write an ORM and because Fey seems like something which could be useful for something at work. By the way, my attempt at an ORM was awful and Dave actually wrote a scathing review of it on CPAN Ratings, which is part of the reason it has been disappeared from CPAN…I think that may have been my first CPAN module. Oh well. No hard feelings.
Finally, there were the lightning talks. They were enjoyable, as usual, but I’m not sure there are any that stand out in my mind from Monday.
After that, my fellow Grant Streeters headed out to dinner. We went to a local, cheap Chinese place, which was pretty decent. I stayed up late working after that, though after all the talks and travel I was pretty spent, so I got done far less than I wanted.
Tuesday
I ended up sleeping in a bit and missed the first round of talks and caught only the last part of Michael Schwern’s talk on the coding cycle. I stayed in the Grant Street room to see Gabor Szabo’s talk on Padre, which I’ve long been interested in, but still haven’t looked at much. If you want to learn Perl, using this editor sounds like the best way to get started.
Then it was lunch again, at Chipotle this time. Mmmm… Chipotle.
I attended scrottie’s talk on running Perl in a cheap distributed computing system (i.e., a pile of cheap computers working together to form a supercomputer of sorts). This was followed by Nick Perez giving a talk on how he’s building apps using POE and Plack.
I moved downstairs to another Schwern talk, this time on perl5i, which is a pretty interesting project. He’s basically taken to pulling in his favorite language extensions and piled them into a single module, perl5i. I’m still thinking about how it works and wondering if I like it or not. I certainly like most of the modules he’s brought in and how it works. I’m just not certain I like the mo and mc meta-accessors… still considering it.
After that talk, I stepped out for a while and worked a bit and chatted with some folks in the lobby of the Union. I met a lot of great folks while here, too many to mention all of them.
Then, I attended the keynote, which was given by Stevan Little. This talk explored the history of Perl and how we got to Modern Perl today. Using some concepts by Larry Wall, he wanted to show how Perl had filled a void that once existed in Unix. That is, it used to be if you wanted manipulexity (lots of deep functionality) you wrote something in C. If you needed something with whipuptitude (built quickly and easily) you used shell script. Perl was an attempt to fill in the difference so you could have both manipulexity and whipuptitude. Modern Perl is the outcome and successful implementation of this.
This is also a good opportunity to explain that while Perl has long had the slogan TMTOWTDI (There’s More Than One Way To Do It) from Larry Wall, Modern Perl makes it longer with TMTOWTDI BSCINABTE. This is pronounced Tim Toady Bicarbonate, which acronymizes, “There’s More Than One Way To Do It But Sometimes Consistency Is Not A Bad Thing Either.” This is an extension to Perl thinking that basically adds a little more consistency to programming (via Moose and related thinking) without actually losing any flexibility. (In fact, Modern Perl is really much more flexible in many ways, it just adds a common foundation that makes that flexibility a little more solid and less re-inventive.)
After this came the second round of lightning talks. There were again several good ones, but I think the talk that outshined them all was the one by Makoto Nozaki titled “How I mastered English with Perl.” It was great. I hope the video of it is posted so I can watch it again.
After this, we had the banquet, where I sat with one of my coworkers, Dieter, some of the folks from Best Practical (Shawn Moore, Kevin Falcone, and Jesse Vincent) and got to meet a couple others, Elliot and Ingy. I had a pretty decent time learning a bit about others, talking with Elliot and Jonathan Rockway in the food line, and made a smallish bid on some coasters during the auction. Pathetic, I know, but I did by a new Perl T-shirt since my Perl polo I bought at OSCON is starting to wear.
After that, I went with some folks and ended up talking about work with my coworkers at a local bar before heading back to my room where I stayed up late inventing a new talk to replace the one I wasn’t able to get together in time.
Wednesday
I started the morning by attending another Schwern talk on gitPAN, which is a really cool way of mining information about CPAN using github. This was followed by a talk by Tatsuhiko Miyagawa on cpanminus, which I had only heard rumors of up until this point. Now that I’ve tried it out, I’m not sure why I haven’t been using it for months now. It’s a very nice (simple!) CPAN client that handles most of what you need when you just want to install a module to try it out.
I then gave my second and final talk on Form::Factory, which showed a bit on how to extend Moose and how I like to build forms, but I don’t think it’s earth shattering. I would like it to be, but my time to hack is constrained by so many things.
I followed that with mst’s talk on The Troll, the God, and the Mountain, which was a very unusual introduction to DBIx::Data::Store. The punchline is that the Troll, Thog and his friend Al, are renamed and made great by the god v10 (“Voton,” i.e., Perl 8 v10) to become “Orthog-an-Al Persisten-T roles.” That’s about all I remember, to be honest. I’ve been hearing a few rumors about this project via Rob Kinyon (I think), though, so I’m keeping an eye open for when it hits CPAN.
I had one more lunch at YAPC, with a bunch of folks I didn’t know and Rob. I confess to have forgotten all names at this lunch, but they were good company for lunch.
Next, I attended Cool Perl 6 today by Patrick Michaud, who is the lead on Rakudo, which is the engine for Perl 6. I’m really thinking Perl 6 should be renamed to something else because it’s like Perl, but it’s not Perl. I like to watch these talks mostly for fun, though, things are getting to the point now that you can write Perl 6 applications, which is tempting to try.
I then attended a really excellent talk on autodie, titled “The Art of Klingon Programming” by Paul Fenwick. The talk featured Paul in a TNG uniform talking about his time as an exchange officer in the Klingon empire where he was testing the new universal translator. The joke is this. If you’ve programmed very much Perl, you know a common idiom is to write “open or die.” The reason is that if the file doesn’t open it returns a false value, so the short circuit “or” operator allows you to skip the die on success or run the die on failure. This is a good way to ask for blood wine in Klingon, since trying to ask nicely won’t get you any. The autodie module comes along to make this kind of talk automatic (making open die without the “or die” on the end).
Prior to the final keynote by mst, I attended two more Moose talks, one by doy and the other by perigrin. Both had way more information than I was really able to consume and the major thing that I took away was that I need to look into the new declarative syntax and immerse myself more deeply into the Moose type system, which I’ve really barely touched.
The final keynote was given by mst, which was another summary of the history of Perl as well as a general summary of some of the good things that have happened in the world of Perl in the past year.
After that, I went over to the Blue Nile Ethiopian Restaurant with the other speakers (and hangers on) and experienced my first taste of Ethiopian food and had one more social event. I got to speak a bit with chromatic, whom had edited the articles O’Reilly published to OnLamp. I got to chat with Ricardo Signes, Dave Rolsky, Rob’s wife, Randall Schwartz, and several others. It was a good time.
Finally, it was time to pack and head home.
In Summary
Perl has a really interesting culture. Between the libraries on CPAN, the dedication of many developers to the language and the community, and the way the leaders in the community interact with it, Perl has a lot going for it. It almost seems like there is a bit of religious fervor to it, though, that’s partially just the concentration of people getting together to talk about Perl. Most of these folks work in other languages too and have important concerns outside of Perl.
On my list of things I need to look into from the conference:
- Perlbrew: A nice way of installing Perl VMs for development and testing.
- local::lib: A nice way of locating and using CPAN modules installed to your home directory.
- cpanminus: A tool that simplifies installing modules from CPAN.
- Bread::Board: A framework for inversion of control, which is something I want to do better with in Form::Factory
- CPAN Explorer: A visualization of CPAN.
- Reflex: A new syntax for doing POE-like things with Moose.
There are others, but I don’t want the list to get too long.
Anyway, that’s my summary of everything I have to say about YAPC::NA 2010. Lord willing, I hope to attend again.
Cheers.
