Leon is making a great presentation on how to make programs for parrot. Larry, Damian, Chip, Allison Randal and Dan Sugalski are here too. Dan is answering questions when Leon has to draw a blank.

We are being told that it will be really easy to use Parrot as your VM if you are writing a dynamic language. And it'll be really easy to extend parrot. Exceptions are almost, but not quite, finished. We have a parrot debugger, pdb. Or you can trace with ./parrot -t run.pbc. Or you can use gdb to debug the parrot internals. Or print statements, "like the rest of us". Things still needs to be added: complete symbol tables, classes and objects, Async IO, more documentation.

"Keep up to date by reading perl6-internals or the Perl 6 summary".

Leon finished in just 30 minutes; great. Questions now ... Dan goes up to answer. :)

In other news then I'm glad I didn't go to the NetTopBox talk after reading Aaron's transcript and Leon's comments.

I can't give any details on the Mailing List Judo talk; except to say that it was really really funny and that if you missed it, you missed it.

The Conference Presentation Judo was really funny too; and useful! If you give any kind of (technical) conference presentation, please read it. Dominus tells about how to make a tutorial teach people thing and not be boring. Did I mention it was really funny?

Dan Sugalski tells about the State of Parrot. It's really cool; he tells about how flexible it is; the languages it already supports. Parrot already supports quite a few different platforms.

I knew that; but Parrot seems like it's in pretty good shape. Dan is a really entertaining speaker.

Parrot can count from 0 to 100M in ~5 seconds; other languages (Ruby (+200), Perl (~160), Python (~120)) takes more than 100 seconds. :-)

Next is mjd's Mailing List Judo and Conference Presentation Judo.

I got myself out of bed at ~9.15 to take a quick shower and make it down for Milton's keynote at 9.30. I was here at 9.35 and the previous keynoters are still on. They blow. I don't know what they talk about, but it's really boring.

Nathan is introducing Milton know. "He is going to talk about open source, you know blah blah blah and Lord of the Rings [applause]".

Milton goes on. "How many here saw the first movie". Lots of hands up. "Okay, so how many didn't see it?" A few hands goes up.

After that he showed the new trailer for LotR II. Talk about crowd pleaser!

He tells about technical things. Each frame is 12MB. They have 20TB storage online on NetApps. Almost 300 workstations (2/3 Linux, 1/3 SGI) for the modelers and designers. They have ~1300 CPUs in the render farm.

Now he is showing lots of examples of evolving the movie from normal shots to shots with creatures. Before that he showed how "forced perspective" works to make some guys look really really tiny while having other actors be normal size.

Each CG generated creature has several GB of textures which are accessed by hundreds of machines at a time; so one of their big problems is actually read performance.

Now he is showing how they made the balrog and the fire simulation which is really hard. The balrog and the fire was getting so much data that the Linux boxes were (almost) crashing.

He shows digital doubles. We get told about what software they use and that it's for Linux (so they can use it).

Question from the audience about a scene he showed: "so this is all digital". "Yeah, this is pretty much all digital. All of it except for the shot in the end, which is a miniature".

Hehe, this is cool. Peter is doing virtual reality camera work. He walks around in a big empty room with goggles on and a camera in hand to "film" in a virtual world.

Their gcc optimized Linux version of some of the software is 3 times faster than it was on Irix. Noone listened to that though because he showed a funny scene while talking about it. He says everyone in the industry has moved from Irix to Linux in the last few years.

He shows a bunch of different tools called things like "Elf-_Sword_Agent-PrePro v. Massive-38". Very cool.

open source in vfx

perl is the primary tool. [applause] Tcl/tk they use for GUIs. mysql. apache/php. java. python.

They are all about quality; so if they get better quality buying something they do so, but he prefers open source. He thanks CPAN module authors, says it has been very important for the movie. Neato.

Someone asks if they could give financial support for open source and he tells that they are a really small company actually. Most people are just contracted for the project; and if there's no new work from Peter after they are done with LotR Weta (the company) will actually cease to exist...

Very cool keynote! People didn't even IRC much, that's how cool it was.

All busy in the sunlight
The flecks did float and dance,
And I was tumbled up with them
In formless circumstance.
        -- Leonard Cohen

This conference rocks. Way too much fun stuff to do. Way too many cool people to meet. Moving from my carefullly selected group of friend in Denmark, it takes a long time to pick up bright people who are fun to hang out with (you know who you are). The Open Source Convention is like an instant boost to the brain of fun bright people. I'm just back from Nathan's p5p party in his suite. Lots of fun people, lots of drinks, lots of laughs.

BoFs

After the last talk I stayed in the room talking to Perrin, meeting Eric Hammond and Christopher Fry whom ironically is on the same floor as I am when I go to do stuff at CitySearch (although he works on the Ticketmaster part). We talked about various ways to manage session data across a large cluster or multiple clusters. Just before this was the last time I saw Chris and Kevin by the way. I wonder where they went and if they made sure to get some of the free beer later. :-D

The room we were in was used for the p5ee bof, so I just stayed there after Leon came in. It ended up just being lots of ranting about managers who gets sold on Java even when perl would be a better choice. There was not really much consensus or energy behind P5EE at all, so it seems a bit dead. It would have been nice to have Stephen Adkins and Matt Sergeant in the BOF, but Matt was elsewhere and Stephen not at the conference. Stephen has done pretty much all the work and Matt has a bunch of insight on this. Oh well.

Then I went to the weblog BOF which was fun. Low tech. Nice. People talked about why they liked the different tools. There was talk about trends in the community and things we would like to see. The last people got an explanation of TrackBacks. I need to remember to ask Rael about the log aggregator for Mac OS X that he talked about. What was the name again?

Ben and Mena Trott (of Movable Type fame) was there, but somehow I didn't really get to say hello to them. What's up with that?! Someone else said hello afterwards and while talking to him Mena and Ben disappeared. It was cute, they seemed a bit shy in there. They've made the coolest thing, they are very likeable and friendly people; I really wish I'll get a chance to say hello properly before the conference is over. They make cool stuff, I'd like to praise them some more in person.

After changing to shoes (feet in sandals got really cold as the air condition made the rooms colder and colder), I went back to the same room for mjd's BOF. I wanted to go to the Perl Foundation BoF too, gaah. At mjd's BOF we talked about random things and in particular debuggers and the perl debugger. I'd like to make a simple command line interface to the regexp debugger he made. Maybe that could be a start of getting someone else to make a cool free GUI thing. My bloodsugar got really really low, so I headed towards the bar and ordered whatever Tim Bunce had. He was there with Graham Barr, Sarathy, Gisle Aas and one more whom I in my semi drunken state right now can't remember who was. Gaah.

p5p party

After a few rounds of free beer courtesy the Aaronson Consulting group, Stonehenge and Dyndns (I think it was), we headed up to Nathans perl5-porters party. On the way out of the bar for a last drink I met Ben Hockenhull. After talking to him frequently because he does the dns for some of our perl domains it was nice to say hello. I had no idea that he was he. Someone sent me a caption for a photo where he was on. "Ah, that's him!"

Anyway, up to the p5p party I went. Lots of fun people there; too many to enumerate. Robert and I took a bunch of photos with my camera and I got to play with Nathans DV camera. Poor Nathan if he tries to edit the ~45 minutes people recorded. Or lucky Nathan as some of it is probably funny.

I finally got to meet Jeffrey Friedl, author of the Regular Expression Book. (Go buy it already; it's out in a second edition). Somehow he was a lot younger than I thought. I'm not sure what I expected, but maybe I think he's too smart *and* nice to not be really old. :-D

I also talked to Milton Ngan, operations manager at Weta where they work on the Lord of the Rings movies. It was very interesting to hear about how the problems are similar and different when your ~1000 servers are rendering movies (compared to have your server farm serve some kind of webfoo). He seemed like a really nice guy; if saleries were only higher in New Zealand I would hurry to explain to Viridiana the benefits of living there and try applying for a job. :-D

Ingy told a bit more about his roadtrip in his car. It really makes me want to not play with computers and travel around for a bit. Maybe I would bring my powerbook. Just to download the photos from my camera. And maybe I'd bring the list of dialup numbers from Speakeasy. Just to check a bit of mail and use CVS. And do a bit of work... Just a bit.

Rael Dornfest told about his dream job (which he has). It sounds like he has a bunch of fun. Lots of it is just staying on top of what's cool. I wonder if I could do that. I'm afraid I'd dig myself down in some of the coolness too much instead of staying up where you can see more of what's going on.

Jason May was there (and at the P5EE BOF). I'd like to talk to him some more too, but I don't even remember when I met him before and where he works. Maybe at last years p5p party where he worked at eBay's onllne payment thing. Hmn... Could be. Must try to find him again.

Oh, did I mention that the multimedia repository from this TPC is getting pretty cool? Check it out.

Okay, very late now. Sleepytime .... ZzZzZz....

Worked on getting pictures online for way too long last night, so I missed our lunch hookup with Brian Behlendorf and DirkX (too tired to remember anything). Aaargh. I even set it up, so I felt bad. But I guess it's okay as for me it was mostly just getting the other Perl Documentation guys to drink the ASF Kool-Aid.

Chip, Jon Orwant, Jarkko and Elaine were still in the lunch tent when I came over. Very entertaining as always.

I wanted to go to the NetTopBox talk; Orwant said that Carl is cool. I got distracted though, so I didn't make it. Aaron Swartz took notes. Excellent. Thanks Aaron!

I saw a bit of Stas talk on mod_perl 2.0. He had way too many slides (surprise surprise ;-) ) but it was good. He sent me the chapter from the book it's based on. Later I met Linda Mui (finally) who told that their book is going into formal tech review soon, very cool.

Then I saw most of Dirk's "writing apache 2.0 modules" talk. It seemed pretty useful, but I knew most of it. So I moved to catched the last bits of Allison Randals talk; it seemed really cool. People were suggesting that she should do it next year in a bigger room. It was fun. Larry, Damian, Chip and everyone were there commenting and cracking jokes.

We had a #perl photoshoot with Julian. Lots of fun. Julian liked it too. Quote: My ego has never been so stroked!
.

Shortly after I took photos of mjd laying on the floor with the octopus and a GNU t-shirt. Tim Bunce helped us make it look authentic.

Now I'm in Perrin Harkin's "shared data with mod_perl" talk. He spend forever getting to the pretty graphs. Useful stuff, but kinda information overload for my attentionspan. All the benchmarks, and that he looked at pretty much all the standard modules to do this is really cool though.

BerkeleyDB; Tie::TextDir and IPC::MM seems to be the fastest "shared data" modules. Another interesting, but known, point is that shared memory isn't as fast as you would think and the file system is faster than you think. At ValueClick Graham and I once made a "shared storage thing" that was all simple flat files in a hashed directory structure. That was by far the fastest we could get, and we had tried quite a few different things. (Including a very sophisticated BerkeleyDB thing with Semaphore locks).

Now he is telling about all the problems with the test. As I said in my talk: Remember kids, always do your own benchmarks.

Perrin's 45 minutes went by really fast. Good sign; I usually get easily bored in talks.

It seems like the 802.11 network is much better anywhere not close to the speakers lounge thing.... Hmmmn...

There are about three thousand BoF's tonight I want to attend. Help!
DBI Bof; p5ee bof; Writing for Ora bof; Weblog bof; Foundation Bof; Aron Wall's Self-Modifying & OS Card Game; Something Interesting Dominus Is Working On. And then there is of course the Free Beer^W^WLAN Party thing.

Yay, I have OSCON Photos online now. It took umpteen hours editing them and making the gallery software work for perl.org (it does some funny proxying). So, now it's 5.30am and I'll certainly be tired tomorrow... aaargh. Oh well. Enjoy!

There is still some editing to be done for "day 3" and a quite a few more photos that could use some cropping and other adjustments.

I don't have the pictures from the activestate party up yet... Later, later.

As I wrote earlier, I had a hard time falling asleep last night and it was really bright outside before I did. So I got up just in time to sorta, but not really catch lunch. Great!

Afterwards I sat in the speakers registration room and worked a bit on my slides. Nathan asked if I could give my CVS talk tomorrow at 2.30, but I can't because by then I'll be in the middle of the mod_perl talk. So I handed over the pod file and tools to make it into slides to Robert Spier and he is going to give it.

Jim Winstead was there, which was fun. Well, lots of people were there, but most I had talked to already. Justin Erenkrantz and Aaron Bannert walked by just as I couldn't get the new mod_cache in apache 2.0 to work. I wanted to demo it a bit in my talk. They have not been involved in it, so they told me to write to Ian Holsman.

Christopher Solomon from ValueClick arrived in the afternoon and came by and said hello before going to catch the last minutes of mjd's last (third!) tutorial on Tim's and my recommendation. Later he said he liked it, surprise surprise. (Mark, you should do more tutorials, you could have done four!)

Dinner

I'm not sure where the time went, but suddenly we figured out that it was time for dinner if we were going to make it back to the State of the Onion. So we jumped in the cars, Tom Phoenix, Hugo van der Sanden, Chip Salzenberg, Tim Bunce, Robert Spier, Dan Sugalski and myself. Nothing like that to remind you that you don't know much about Perl. Oh, and Chip is really really entertaining. We went by a few places that looked nice, but incompatible with our "need to eat and get back in less than an hour" schedule, so we ended up at a faster Mexican place. It was nice and the company was great.

Larry's State of the Onion

We got back in time for the Onion talk, where Larry Wall used a table of contents from Scientific American to relate various things to Perl. He talked about how the Perl 6 and Parrot efforts has been good for the perl 5 community and the perl 5 development. (And lots more, of course).

He talked about how Uncertainty and Doubt is a part of life, and being human. Fear is the evil part of FUD, because it plays on the other two to make us afraid of the unknown. (Or that's the short version of how I understood it).

In the end he talked about how the world is better off if people people are tolerant about and even embracing other languages. The title from Scientific American was "The Death of Languages" or some such. Not that Perl is dying. It was about how it is that languages die; what stages they go through. And he talked about There Is More Than One Way To Do It; also when it comes to choosing language. Not that we should all go and use Python, but he mentioned how parrot so far has been a place where people have been playing with odd and old languages. And how it's a good thing to have the diversity.

It's true that the perl community in the last years has been becoming increasingly clued that Perl is good and Python is good. And Ruby. And Lisp. And C. And probably VB too (although we wouldn't know when or how). And Intercal (okay, okay, I'll stop). Our community is aware that advocacy sucks. In particular when it's clueless and bashing other languages. It's great. It makes me happy to be part of this.

Quiz Show

Afterwards was the quiz show. Derek Balling and the rest of the usual team won as usual. Ian Holsman was very helpful over irc (he's sadly not here) helping me with the mod_cache stuff. Graham Barr arrived just in time to get a well deserved white camel award for search.cpan.org. I was too tired and unfocused to really have fun with the quiz show. Tim, Chris and I left a bit before to go outside and get a beer. Uuuhmn, much nicer.

Outside

Outside were Hugo, Jarkko, Elaine and a bunch of other great people. Justin and William Rowe came by to say hello. I never met Will Rowe before (or Justin before this afternoon) and it's always great to put a face to the name. I also said hello to Chromatic from #perl in person for the first time. Fun fun.

Just like last year Tim helped me polish my slides over a few beers on the patio (last time it was more like getting them from notes into powerpoint though). It was very good going through them again with someone else. Tomorrow after my talk I'll take time to put the photos up. But for now I'm back in my room, ready to crash. Actually, I already crashed on the bed... ZzzZZZZZ .... Good night.

Aaarrgh, I can not fall asleep at all. Gaaaah!

Except for visiting mjd's regexp tutorial to take photos I still haven't been in a session at all.

Around 6pm people were meeting up for dinner. Our little group quickly escalated into a larger group and morphed with another group or probably rather groups. So suddenly we were standing with ~20 other people with vague plans. Shuttle bus came. Eh, no. The little shuttle bus came. The little shuttle bus is a bit smaller than your average suburban minivan here in southern california, so obviously that didn't work so well. The driver said he would come right back though... Hhhmn... Vivek Khera chimes in that the "real" bus goes from the other side of the building, so some of us go over there. "It'll be here in just 10 minutes" we hear. She didn't sound too sure though. Anyway, Robert and I both have our cars here, so some of us went to just drive somewhere. (I think "we" were Tim, Ken, Robert, Vivek, Eric Cholet, Bill Moseley, Cindy, ... and a few more. My memory fails me horribly!)

We went to some place by the water. The food was decent and the conversation was good, so it was great. We got back just in time for some of us to catch the ending of the Perl Documentation Project BOF. We hung around afterwards there and talked about about the perl foundation, legal issues with the project and various other things. It was fun. I was getting really really tired though and went to my room to sleep not much later. That's almost 6 hours ago. Grrrrh...

I'll try sleeping again now... Good night. ZzzzzZZz...

I arrived yesterday after driving about 2 hours. That's a nice thing about the conference being in San Diego; for me anyway. :-)

My hotel room is on 11th floor (East tower) with a few over the Bay and downtown San Diego, it's actually really pretty. I must remember to take photos. Yesterday at night and earlier this morning I had internet access through some friendly accesspoint called "Wireless". It's gone now though. I can sorta, almost, nearly see the "oreilly" network, and if I walk down the hall I can see "ingynet" (Brian Ingy Ingerson's, I presume). But that doesn't help me much in my room,
so the hotel is now getting $12 to get me 24 hours access. Of course I put up my access point, so fellow attendants nearby my room can share.

I stayed up a bit too late yesterday in the bar hanging out with Ken Williams, Jan Dubois, Jesse Vincent and lots others. As Dave Cross (I think it was) wrote, when you sit in the bar at the perl conference, eventually the whole perl world will pass by.

I didn't get up out of my room until past noon when I walked down to the lunch tent and found Tim Bunce, Robert Spier, Ken and others. Robert took some photos with my camera. The food is much improved from last year; great! In particular the desserts were good. To Ken's amusement I had three slices of the fruitcake. Uuuuhmn.

Afterwards I walked around the corridors and into mjd's regexp tutorial briefly to take photos. I will get them up soon, maybe at photo.perl.org or something like that. We could collect photos from everyone who wants to contribute.

Now I'm back in my room and is supposed to make more slides for my talk. Gah.

(doh, MacStumbler just said "diiing" because the "Wireless" network from my friendly neighbor is back. I could have saved $12! Or he could have. Oh well).

Uhu, leaving for San Diego now. Tentative ETA (depending on traffic) around 6.30-7ish. I hope I'll find people quickly. Yay.

If you know stuff about mod_perl and want to review my slides; please drop me a mail.

In not too many hours I am going to drive down to San Diego. I am just on my way to bed, so it still counts as day minus 1.

Not entirely sure what sessions I will attend (except of course for Real World Performance Tuning in the mod_perl Environment Wednesday afternoon). Sunday I will hopefully make it in time for dinner with people there, and in time to meet for some meeting which agenda I momentarily have forgotten. It was at 7pm I think. Hmn.

I need to pack some stuff. (Accesspoint for wireless networking; camera; lenses; socks; battery charger; clean underwear; powerbook; cd-burner(?); empeg?; book I promised to bring Nathan; toothbrush; contact lens stuff; old laptop drive in firewire enclosure for extra storage; anything else?)

Sleepy time.

Eewwwp; in no more than 24 hours I should be in San Diego at The Conference. I hope my sleep pattern will be better next week than it has been the last weeks; otherwise I'll have a hard time making it to any of the keynotes. Or stay awake for the beer. Or finish my talk. Gaah.