The person on the frontpage of www.postgresql.org is David Fetter, and the picture is taken late afternoon on the day after pgday (which means a rather late evening the day before and a long day of sightseeing in Florence). If you're not properly tired after several days (and nights) of community activity, you didn't try enough!
So now the second day of pgday has started. Yesterday continued in the same way it started, with several more good talks. Andreas did a good talk on community relations. In the late afternoon Susanne showed up to do a talk on locales and encoding - that's another name checked off the list of PostgreSQL community people to meet.
After the conference itself, it was off to a restaurant for proper Italian pizza. In what appears to be standard Italian ways, the restaurant hadn't actually opened when we arrived, so we had to walk around the city for a while before they'd let us in and serve us food. But once we got there, it was all good.
The day finished at the PLUG bar (I can't stop being impressed that the usergroup actually has their own bar! If that's not a good driver for membership, I don't know what is..) This time they even had a band with live music!
Got back home around 2, did the mandatory email checks, and off to bed. Actually made it in time for JoshB's keynote today. For those of us well inside the development process right now much of it wasn't news, since it was focused on what's coming up in 8.3. However, one interesting part was that Sun and EnterpriseDB actually published commercial benchmarks on PostgreSQL! They should be going up on the website sometime early next week (that's Sun's website, but I'm sure there will be links from the news section on ours), but the exec summary is that PostgreSQL is very fast and very cheap (same speed, 1/3 the cost as Commercial Vendor One). Which we already knew, of course, but it certainly helps the PHBs to have nice graphs and official numbers. So kudos to Sun and EnterpriseDB for getting that done!
I arrived in Prato, Italy, yesterday and even got picked up at the airport. We then went to Monash University where the conference would be. After some time waiting to help out we (me, Josh Berkus and David Fetter) gave up on that and went for a walk around town instead. Picked up Andreas (ads) and Stefan at the hotel. Several hours later we met the Italian guys again for dinner and drinks, and it seems they somehow got things sorted without us because they were ready to go. We also picked up the guys from Dalibo (France) at the same time. Finally, after we'd all had our food and drinks, the edb guys from the UK (Dave, Simon, Greg) arrived and we just had to go out for another couple of beers...
Anyway. Today the actual pgday has started (it's actually two days, regardless of what the name implies). Talks are actually translated in realtime, which works reasonably well but not perfect (as always). The amount of merchandise they have is impressive - t-shirts, notepads, pens, more t-shirts, pins etc. And Andreas brought mugs and those cool plush elephants he's been blogging about, so they're really well set. And they have an impressive amount of volunteers, so things are going very smoothly. Even the wireless network works today...
There's a big crowd here, so from what I can tell this is so far a big success. There's of course a large amount of Italian people here, but there's also a lot of people from abroad - as far as the US and Japan, and a good number from all around Europe. Let's hope the rest is as good as the start!
Oh, and if you're not here, there's live audio streaming available!
So it's been a couple of days since I arrived for pgcon in Ottawa, so it's obviously time to blog. At least Devrim keeps telling me to...
Trip in was not too bad, though 4 hours at Heathrow airport is not exactly fun. Got here Monday evening, tried to coordinate with the EnterpriseDB UK team who arrived an hour or so after me. After much problems (ahem, phones that don't work abroad and such - hello Dave), we did manage to hook up and had a dinner and some beers in the evening. Plus, of course, PostgreSQL chatting!
Tuesday was mostly spent walking around the city and playing tourist along with Dave. Simon and Heikki hadn't done their homework properly, so they had to update talks and finish off patches. In the afternoon we headed over to the Royal Oak pub where the registration was, and spent the rest of the day chatting with the other pg people who had arrived.
Since I didn't go to any tutorials, the conference itself started for me on Wednesday, with Bruces keynote. Somehow I managed to get mentioned in that one again, I just don't know how that keeps happening... An interesting look back at some of the things that brought us to where we are now, and a good way to kick off the conference.
Next talk for me was PostgreSQL-IE, which is a system for doing context based image search in PostgreSQL, and looks really really interesting. Currently used in the medical field by them, but it seems like it could be used for several other things as well with fairly small changes.
After that, Robert Treat did a good overview on partitioning. I haven't really looked into the partitioning stuff as much as I've wanted, so this was a really good way to get started on that.
After lunch, I went to a presentation on migrating from SQL Server (primarily) to PostgreSQL. But I'll have to admit I spent a lot of that time working on my own slides, so I can't really comment on the talk.
Denis gave a talk on why the EnterpriseDB team picked PostgreSQL as the foundation for building the EnterpriseDB product and company. Most if the reasons were the ones you'd already have guessed, but he brought some nice details about the decision process.
Getting to the last actual talk, there was that weird PostgreSQL-on-Windows guy giving his talk. I think it was fairly well received, even though Dave had to point out I forgot to say that he had actually fixed the installer so that it'll work on Windows Vista out of the box on 8.3. Oops, sorry about that Dave! We'll see what the eval forms tell me for a more objective view on how well it went.
The final part on the actual conference was the BOF. The discussion we had first turned out into pretty much a repeat of things that were already said on the mailinglists, so I'm not sure it was very productive. We did end it up with a keysigning event, which worked better.
After that it was off to the EnterpriseDB sponsored drinks and dinner party. We got there hours after the other guys (who skipped the BOFs), but we got there before the food was served, so the damage wasn't that big. Food and beer and good discussions were had. When the place closed we moved on to the bar that's in the hotel where the EDB guys stayed for another beer, but we didn't end up staying there very long.
Just in for the start of the last day now, also worrying about the fact that my airline for my trip home have gone on strike, so I have no way to get home right now. But I'm sure it'll work out some way, so I'll just focus on the conference for now.
I just got booked for going to pgday in Prato, Italy. Looks like it's going to be a great gathering of the European people in the PostgreSQL community. Really looking forward to meeting those from the EU group that I haven't already had a chance to meet. I haven't decided if I'm going to submit a talk yet though, but it seems I have a few more weeks to decide.
Closer in time, we're rapidly coming up on pgcon next week. I really need to polish my slides soon. But I guess that's what long flights are for after all... Between these two events, there isn't that much of the PostgreSQL community that I won't get to meet over the next couple of months. Great times ahead!
Oh, and what has happened to planetpostgresql.org? The post-rate has really dropped drastically (and yes, I realize I'm definitely part to blame for this). There are as usual interesting exceptions, but it seems the number of posts have gone down. People, get back to blogging interesting stuff!
Dave has put up a snapshot of the new 8.3 installer on the pgFoundry page. This contains a whole lot of things for Windows users to test:
So if you're a windows user, and want to test things, please download this file and let us know (through the mailinglists!) how it works out for you.
NOTE! This is a pre-beta development snapshot of both the server and the installer. DO NOT USE IN PRODUCTION!
Thanks to Dave, we now have a couple of machines on the buildfarm that build with msvc - a definite requirement for using the msvc build for the main binary distribution of 8.3. Today had my first msvc-specific patch that the buildfarm found for us - a trivial overlook, but the buildfarm certainly helped. And even the Unix folks seem to like it - diversity is good for portability.
As of yesterday, msys/mingw is no longer used to build the CVS tip of pginstaller. Instead, it builds completely using the tools of Visual Studio 2005. This includes both the actual installer, and the custom actions written in C. There is no change to the actual installer, it's still built using WiX based on the same files, and the DLLs are built from the same source code (except for a few compatibility fixes).
The build system for the pginstaller is fairly complex. Or at least, it's fairly weird. Mainly because of the need to deal with the translations, which are applied by first generating complete installers in different languages (which in turn is done by rewriting the whole source code and replacing translatable strings) and then doing a "binary diff" between them, which is finally embedded in the main installer.
The drug of choice for this rewrite is msbuild. While primarily geared for building managed code (that's Microsoft-speak for .Net code), it can be a pretty good Makefile replacement. While it can't do everything that your typical unix makefile can do (given that it can't run arbitrary commands), it has some good features for dealing with just the situation we have with our language code. Check it out - it's all in the pginstaller cvs. It's always a matter of taste which is cleaner - I personally find the msbuild code a bit more readable.
The best news is definitely that there is no need for a msys/mingw installation to build it anymore. And if it's used to create an installer for the msvc build of the backend, there's no need for msys/mingw at all anymore.
As a final step to my updates to the download code on www.postgresql.org, I've added code so the server remember which mirror was last used to download. When you download a second file, this mirror will be automatically picked for your download, saving you a whole click. Enjoy!
I've just committed code to the website that finally allows http downloads. This has been a much requested feature for quite a long time, mainly from those users that are behind firewalls that don't work well with ftp. Most often a http download of a single file will also be faster than ftp download of the same file, since there are less roundtrips. Though for large files that difference shouldn't be noticeable.
The mirror checking script will poll the same mirrors we already have and figure out which ones support http download as well as ftp. Therefor, all mirrors listed will always support ftp, but some (currently about a third of them) will also support ftp. There will be no http-only mirrors.