Hi everybody,
So a reportback on OSCON 2006. This is the publicly consumable part, as with my MySQL UC report.
The trip down was a pain - this is worthy of mention because none of us are American. The US INS (immigration/customs) guy that I had to speak to had absolutely no understand of open source software, or that we could give something away. He asked if I had some proof that I was associated with phpMyAdmin as a business (he kept referring to it as such, no matter how many times I pointed out we were closest to a community organization) - so I showed him one of my phpMyAdmin business cards, and the 11x17 posters I had
Do not be tempted to lie to these guys, they use infra-red cameras trained on faces to try and detect it. This still did not help his understanding of phpMyAdmin and open source, and he opened his drawer and got out his dreaded orange rejection pad (by which they deny you entry into the USA) - I had to rack my brains and make very fast analogies at this point - and eventually I succeeded, but it is NOT an experience I would like to repeat again.
A number of the other folk at the conference admitted to having similar troubles while crossing various borders.
The rest of the journey down to Portland was uneventful - I had time for a brief bite of lunch in Seattle once the bus stopped, and before the Amtrak train left (I'll take the European rail system, even that of Italy over Amtrak anyday however).
On reaching Portland, I was grabbing a cab to where I was staying, which turned out to be interesting in itself - the cab driver was Eritrean, and could not understand my South African accent whatsoever, to the point where I had to write down the address.
Originally, I was going to be staying with a fellow Gentoo developer, but about a week ahead of time, he had roommate issues, and wasn't able to offer me a place. A few other Open Source leads didn't help either.
So, in the same vein as MySQL UC 2006, I used CouchSurfing.com, and ended up on a fold-out futon bed, 1.5 miles from the convention centre. My host said she could loan me a spare bicycle, so I brought my helmet and other gear down. She asked that I keep her identity to myself, for various reasons.
CouchSurfing as a concept works out reasonably well. You get a couch or better to stay on, and generally just try to lend a helping hand while you are there (I helped in the kitchen, read a bedtime story to my host's daughter and her friend, tuned up the bicycles, and a few more things).
The Wednesday morning cycle to the convention centre was ok - I had previously made a brief study of Google Maps, and was told by my host that the convention centre had two very large spires that were good landmarks to orient by within the NE district of Portland. The ride was a bit interesting, the route I picked based on Google Maps turned out to go via some of the slummiest parts of town, which felt marginally unsafe at 8.15am.
Got to the convention centre, signed in, rigged up the 11x17 posters in the booth (as visible in the photo that Marc linked to, which was taken by the ZDNet photographer, early on Thursday morning, right after the doors opened for non-exhibitors), and said Hi to the other dot-org booths around my area - a bunch of the same people that were at MySQL UC - Apache, WITI, FSF. Fired up the CVS HEAD demo on my fiance's laptop, and waited for the first crowd to come in.
As a general summary, while exhibit hall passes were free for OSCON - it was much subjectively quieter than MySQL UC - most of those that did come in exhibit hall passes were just there to scoop freebies - and did not interact with any booth without freebies. At the breaks between sessions, there was a good deal more activity in the exhibit hall - due to the refreshments and snacks. For folks approaching the phpMyAdmin booth I had a number of folk complaining of not getting it working, but after some targeted debugging questions, the source of their troubles were Apache/PHP, not phpMyAdmin in any form.
I had a number of UI designers make interesting comments about the PMA design - such as the cookie login message (from a Databases professor at Stanford) that I brought up for Marc's message freeze request. Some of these I see we are already working on (like the present thread for tracking server/database etc in a cookie). Some more of them I intend to dig out my UI design textbooks and examine more closely to see if there are logical ways that we can improve them.
There were no comments about the lack of bleeding edge functionality - even those that I specifically asked about a stored procedure editor admitted they did not have a need for one.
There was one notable functionality request, and it's got some low-hanging fruit we can consider - an explicit tab-separated value (TSV) export format - in the same vein as our CSV export. The second part of the request was considerably more complication - and that was IMPORT functionality for CSV/CSV-Excel/TSV/XML data.
I went for lunch Wednesday with teh Gentoo developers, courtesy of E-Trade - who are apparently very large Gentoo users - discussions had about application packaging were very in-depth.
The afternoon was similar to the morning, just hanging in the booth, and speaking to interested conference attendees.
There were a lot of full conference attendees that just breezed by the booth shouting out things like 'Great software!', 'We use it everyday!', and similar slogans of support.
Starting at 15h30, the free beer came out. Nothing as elaborate as the party at LinuxTag last year, but still very good quality micro-brewery german-style beers (which Portland is famous for), incl. a nice triple Bock, and an Altbier similar to that which Alexander had going last year at LinuxTag for us.
The exhibit hall for Wednesday wrapped up at 5pm, and everybody (incl. exhibitors) was kicked out so the caterers could set up the IBM-sponsored reception. The 6pm reception was reasonble - good appetizers, and a lot more beer and wine.
From there, some of the Gentoo developers and myself went to two of the open-bar parties that were taking place in nearby hotels - the first was slow, and badly attended, with a Jewish kleizmer (sp?) [with a story attached linked to Metallica], and more beer, but no food. The second party was a lot more active, held by O'Reilly - their hotel suite was shoulder-to-shoulder people. I engaged in some interesting conversations with Google employees, before calling it a night around 11pm, so I could cycle back while still marginally sober (and able to stay upright on the bicycle).
Thursday morning saw me figure out a different cycleroute to the convention centre - one that turned out to be safer from people and traffic (there are cycling lanes painted on the road, but DON'T cycle in them with the traffic unless you have a deathwish, they do NOT like cyclists). Rigged up the laptop demo again, the ZDNet guy came by and grabbed the photo of the booth with me busy setting up the laptop demo.
Thursday seemed much quieter. I had an interesting discussion with a group of high-school youth that were developing a MySQL-based library book tracking system for their school, and helped out with their database design. I arranged for 50 blank CACert CAP forms and also did CACert assurances at the booth during Thursday, as I found that I was not going to be able to attend the keysigning event Thursday night.
Ate the conference boxed lunch at lunchtime, it wasn't noteworthy in any way. Come the 5pm teardown time, I grabbed my stuff, and departed. I was informed of a few more parties, but I turned them down due to my pending early departure on Friday morning. Stopped at a bookstore on the way back.
I had a light dinner out with my host, cooked up my famed near-diabetic-inducing chocolate sauce which was then served over ice cream and strawberries. Grabbed some sleep before the morning and leaving to catch the public bus to the train station at 7am. The return journey was much better, baring the initial 1 hour delay due to a train derailment south of Portland.
The private report portion will be a followup email to the usual location.
Robin H. Johnson schrieb:
For folks approaching the phpMyAdmin booth I had a number of folk complaining of not getting it working, but after some targeted debugging questions, the source of their troubles were Apache/PHP, not phpMyAdmin in any form.
is there anything for what we could add some checks in the setup script (or an additional check script) and alert the user of his misconfigured php/Apache?
On Fri, Aug 04, 2006 at 10:04:44AM +0200, Sebastian Mendel wrote:
Robin H. Johnson schrieb:
For folks approaching the phpMyAdmin booth I had a number of folk complaining of not getting it working, but after some targeted debugging questions, the source of their troubles were Apache/PHP, not phpMyAdmin in any form.
is there anything for what we could add some checks in the setup script (or an additional check script) and alert the user of his misconfigured php/Apache?
Their php/Apache was broken enough that NO php scripts were running at all.
Robin H. Johnson schrieb:
I had a number of UI designers make interesting comments about the PMA design - such as the cookie login message (from a Databases professor at Stanford) that I brought up for Marc's message freeze request. Some of these I see we are already working on (like the present thread for tracking server/database etc in a cookie). Some more of them I intend to dig out my UI design textbooks and examine more closely to see if there are logical ways that we can improve them.
please share ... ;-)