|
| Greg's diary |
| November 2005 |
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Tuesday, 1 November 2005 | Echunga | |
| Top of page | ||
| next day | ||
| last day |
Things are gradually getting back to normal, though one mail thread is still worrying me: over 140 messages on poor MySQL performance on recent releases of FreeBSD. Spent some time installing software on the new eucla, not overly successfully. Installing Emacs failed consistently with the following messages:
===> emacs-21.3_8 depends on shared library: Xaw3d.8 - not found ===> Verifying install for Xaw3d.8 in /usr/ports/x11-toolkits/Xaw3d ===> Vulnerability check disabled, database not found ===> Extracting for Xaw3d-1.5E_1 => Checksum OK for Xaw3d-1.5E.tar.gz. ... (Xaw3d installs) + /usr/bin/install -c -m 0444 XawImP.h /usr/X11R6/include/X11/Xaw3d install in . done install.man in . done ... ******************************************************************* ===> Running ldconfig /sbin/ldconfig -m /usr/X11R6/lib ===> Registering installation for Xaw3d-1.5E_1 ===> Returning to build of emacs-21.3_8 Error: shared library "Xaw3d.8" does not exist *** Error code 1
Investigation shows that the port was installing libXaw3d.7, not libXaw3d.8. I solved the installation problem by creating a link between /usr/X11R6/lib/libXaw3d.so.7 and /usr/X11R6/lib/libXaw3d.so.8:
=== root@eucla (/dev/ttyp2) /home/grog 1 -> l /usr/X11R6/lib/libXaw3d.*
-rw-r--r-- 1 root wheel 420930 Nov 1 17:39 /usr/X11R6/lib/libXaw3d.a
-rwxr-xr-x 2 root wheel 309646 Nov 1 17:39 /usr/X11R6/lib/libXaw3d.so
-rwxr-xr-x 2 root wheel 309646 Nov 1 17:39 /usr/X11R6/lib/libXaw3d.so.7
-rwxr-xr-x 2 root wheel 309646 Nov 1 17:39 /usr/X11R6/lib/libXaw3d.so.8
Why is software installation such a pain?
The weather is gradually getting warm. This time last week I was picking my way through the snow and slush at Helsinki airport; today the temperatures were close to 30°.
| Wednesday, 2 November 2005 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
A lot of work today, none of it particularly noteworthy. Somehow managed to clock up 12 hours anyway. Discovered, more or less by chance, that I wasn't on an important internal mailing list, and that I had been put down for a number of presentations in Athens next week.
Installing software on eucla continued unabated. In the Good Old Days, with slow machines, it was practical to compile all the software, but it no longer seems to be the case. Installing GNOME took 6 hours, and kde was still going when I went to bed.
Just before I finally did make it to bed, the phone rang: Mikael Ronström of the cluster team, talking about Athens next week. Another hour. I'll be glad when this travel and stuff is finished.
| Thursday, 3 November 2005 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Came into the office this morning to discover that my KDE build had failed after 41899.95 seconds (nearly 12 hours):
===> Registering installation for apr-nothr-gdbm-db4-1.2.2_1
===> Returning to build of subversion-1.2.3_1
Error: shared library "apr-1.0" does not exist
*** Error code 1
There's too much of that stuff, but the real question here is: why is a default installation of KDE installing subversion? Tried installing something closer to home, with other problems:
This proved to be due to KDE again:
=== root@eucla (/dev/ttyp3) ~ 122 -> pkg_info|grep mysqlmysql-client-4.1.15 Multithreaded SQL database (client)=== root@eucla (/dev/ttyp3) ~ 123 -> pkg_delete mysql-client-4.1.15pkg_delete: package 'mysql-client-4.1.15' is required by these other packages and may not be deinstalled: koffice-1.4.1_1,1=== root@eucla (/dev/ttyp3) ~ 124 -> pkg_delete -f mysql-client-4.1.15pkg_delete: package 'mysql-client-4.1.15' is required by these other packages and may not be deinstalled (but I'll delete it anyway): koffice-1.4.1_1,1
After that, MySQL installed. I wonder if it broke koffice. I can't say I care too much.
That wasn't the end of things. I finally built a new kernel for FreeBSD 7-CURRENT, but when I tried booting it, the shutdown process left three file systems dirty. On reboot, the system hung. Spent an inordinate amount of time (all the less thanks to the availability of FreeSBIE), and finally discovered:
All recent versions of FreeBSD seem to have a problem rebooting if any ext2fs file systems are mounted. This was the case with my system; umounting before running shutdown seems to work around the problem.
As the result of the unclean umount, the file /sbin/init, of all things, was truncated. That was the cause of the hang.
I had added support for dcons, the dumb console, in preparation for my debugging course later this month, and this caused the console output to stop exactly at the point that the root file system should have been mounted. In some cases, the system rebooted correctly, but the lack of output made me thing that it was hanging.
Things weren't over yet, though: in the afternoon the machine gradually ground to a halt while trying to build mutt. And further investigation showed that I had a more general problem with a problem that I had observed yesterday: for some reason, when using a character-mode console rather than an xterm, the character echo was “spongy”: it was noticeably slow. Further investigation revealed that a ping with the default 1 second delay actually took about 5 seconds between pings. In fact, a whole lot of things didn't work properly; it seems that some timer interrupt wasn't working.
That wasn't what I need 3 days before leaving for Europe for a month, so decided to install the newly-released version 6.0 instead. After several hours, came to the conclusion: the same problem appears there. What a pain, especially since it doesn't happen under 5.3 or 5.4.
| Friday, 4 November 2005 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Still more work on the laptop install today, and finally confirmed that the problems I had been seeing are related to ACPI. Disabling ACPI solved that problem, but of course it also removes the advantages; I have already been wondering whether the overheating problems I had with the last incarnation of eucla were related to ACPI being disabled. More investigation is needed.
| Saturday, 5 November 2005 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Finally I'm getting things under control. In retrospect, I would have had a pretty full week if I hadn't had the fun with the laptop; that was the straw that broke the camel's back. Today installed some multimedia software—this machine has a DVD writer—and tried copying that. Surprise! /home is an extfs file system, and at least under FreeBSD files are limited to 2 GB. Managed on a second attempt to burn a DVD over the net, but that's clearly not the right thing to do.
Apart from that, bottled some beer and finally got round to reading back issues of the local newspaper (the Mount Barker courier): I hadn't read them since the end of September.
| Sunday, 6 November 2005 | Echunga | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Spent the day in travel preparation mode again, mainly spent finding strange bugs in my laptop install. For some reason had another couple of spontaneous reboots. Not fun. I think the time for playing around with software is over, and now I want reliability.
Also spent some time looking for software to play MP3s. Tried cmp3, a curses-based program that insisted that my xterm background was black (and set it, thus leaving me with a black-on-black display, and then gqmpeg, another toy that thinks it's a kiddie's pocket MP3 player:
Finally discovered mpg123.el, an Emacs macro that interfaces with mpg123, which seems to do most of what I want, though there are some rough edges. Strange where Emacs can still hold its own.
| Monday, 7 November 2005 | Echunga → Singapore → | Images for 7 November 2005 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Last minute preparations for my journey to Europe included storing some phone numbers in my phone. And the keyboard failed! While removing it from the cover, the zip on the cover broke too. All this for a phone that's not even 2 years old, and what timing!
Off to the airport, where they weren't quite as badly prepared as last time. They still don't have networking in the Qantas Club, but there were a couple of desks just outside with network connectors:
Unfortunately, they didn't work. It's not clear whether they're intended to (the jacks look like they double as RJ11 phone jacks and RJ45 networking jacks), but it would be a nice gesture.
On without event to Singapore. Getting on the plane at Singapore, they wanted me to remove my jacket (why?). I asked if they wanted my shirt off too, but they didn't. So I asked “What's the Malay word for jacket”. Answer: “baju”. “And what's the Malay word for shirt?“ “Sorry, I don't speak Malay”. In fact, the words are the same, as another guard on the other side of the metal detector confirmed.
| Tuesday, 8 November 2005 | → Frankfurt → Athens | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Arrived in Frankfurt at 5 am, and was sent to Terminal 1, where there was absolutely no information about what to do next. My printed itinerary showed just “Terminal 1” with no indication of which part, and when we got there nobody knew what to do. Collected a couple of other lost souls before arriving in the departure area of area B. It proved that I had to go through immigration to do anything, and after a while discovered that I could get into the American Airlines Admiral's Club, which is a whole lot better than spending 6 hours outside in the terminal. Unlike the Qantas Club in Singapore, though, they have hidden their computers and network connectors where I couldn't get at them, so I couldn't tap into their network. What kind of businessman would use a public computer to read his mail?
Instead invested € 2 for 15 minutes connection to the airport wireless network, not to mention 5 minutes filling in the forms (“in triplicate” they'd say in Germany), and downloaded my mail. After processing it tried to connect again, but accidentally forgot to set the expiry date of the credit card, and some bug in the authentication software refused to accept that I had changed it, so gave up on that.
On to Athens by way of Thessaloniki, and got to the hotel without too much difficulty. It's been 26 years since I was last here, and I didn't recognize anything. The hotel room, which I'm sharing with Igor Babaev, was at 26°, and we had fun trying to cool it down. Then down to the lobby and met far too many known names at once—it'll be a while before I can associate them all. Off with Igor, Mark Matthews and Elliot Murphy to what the concierge claimed was a typical Greek restaurant, but which turned out to be a Western-style steak house. Still, had reasonable food and then early to bed.
| Wednesday, 9 November 2005 | Athens | Images for 9 November 2005 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
The heating in this room is terrible, and I slept even worse than normally. Got moving slowly and rather differently from what we had planned: contrary to what we had expected, Mårten Mickos, our CEO, showed up, and so we didn't get down to our planning meeting until late morning. Apart from MySQL people, Ann Harrison and Jim Starkey of the Firebird project were there, and we spent a lot of time comparing implementations.
Apart from that, managed to find another room with better heating—strangely enough, a cheaper room—and also participated in a long discussion about our future course. It's really refreshing to be in a company that, despite commercial ambitions, is not prepared to pursue these ambitions to the detriment of the Open Source community.
In the evening out to dinner. Had a good time, far too late to bed.
| Thursday, 10 November 2005 | Athens | Images for 10 November 2005 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
On with our discussions today. It's clear that our viewpoints cover a fair area, and spent some time (probably too much time) discussing low-level implementation alternatives. Things have changed a lot since I last looked at these issues.
I had sent out a message to the FreeBSD mailing lists planning to meet up with some people, but the main person active in FreeBSD in Athens, Giorgos Keramidas, had chosen the time for a business trip to California. Instead got a mail message from Spiros Bolis, who had discovered an interesting detail:
Funny things do happen..
Guess who's staying these very same days and at the *same* hotel: the core
devel and business administration team of MySQL!
They had planned a dinner this evening, so out with them to quite a pleasant Taverna, something we would never have found by ourselves:
|
|||||||||||||||||||||||
Interesting discussion, and of course far too late to bed.
| Friday, 11 November 2005 | Athens | Images for 11 November 2005 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
The combination of late nights and early rising doesn't work well with me, and it took a while for me to come up to speed this morning. More discussions, and we made quite good progress.
Finished at a relatively civilized time, and I decided to get an early night. That didn't quite work: ended up walking with Ann Harrison, Jim Starkey, Monty, Anthony and Sergei to the same area that we visited on Wednesday:
|
|||||||||||||||||||||||
Food was good, but somehow the whole business took over 4 hours, and didn't get to bed until 00:45.
| Saturday, 12 November 2005 | Athens | Images for 12 November 2005 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Project management training this morning, an attempt to compress 2 days of training into 4 hours:
|
|||||||||||||||||||||||
Not surprisingly, it didn't quite make it, but it was useful nonetheless.
In the afternoon we had intended to go sight-seeing, but it seems that apart from the Akropolis there are no really interesting places to see. I wouldn't have minded going to the Akropolis again, but they shut at 15:00, and we would have not had much time. Spent some time trying to find something else to do, without success.
In the evening to a little Taverna I had found, right round the corner from the hotel: I had found it by looking out of the window. It turned out to be exactly the kind of place I had been looking for all along. Ended up with quite a crowd, and finally I made it to bed early enough.
| Sunday, 13 November 2005 | Athens → Frankfurt → Stockholm | Images for 13 November 2005 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
An early flight meant that I had to wake up at 5:30 am, but I was awakened earlier than that: Igor had to leave two hours earlier. He hadn't been gone long when the phone rang again, pretty much exactly at 4 am. Not very happy, and recalling an incident in Brighton a few years ago I answered it anyway. Line immediately went dead.
I was just getting back to sleep when it rang again; answered in no uncertain terms: “Shut up”, and left it off the hook.
Next thing there was a knock on the door with a message to call Igor. He had left something important behind, and asked me to get Brian to take it back with him. Reluctantly woke Brian, who told me that he wouldn't be returning home for a week, too long for Igor. Since I was awake anyway, got up, dressed and a taxi to the airport. By the time I got there, Igor's flight had been called, but gave the object to a staff member, along with my business card, and asked him to contact me if he couldn't get it to Igor. No call, so hopefully it worked. So much for my intention to get enough sleep.
Somehow managed to while away the time; Anthony was there too, so had a bit of conversation until he left. Finally off to Frankfurt, discovering that I was sitting almost directly next to Sergei Golubchik, and to the American Airways Admiral's Club where I had been on Tuesday. The same woman let me in as last time, but this time they had decided that they had changed the regulations, and without an onward flight on a One World airline they didn't want to let me in. They did it this time anyway; maybe I should join a Star Alliance club.
Then on to Stockholm. I had booked SAS, but it turned out to be a Lufthansa code-share. I try to avoid Lufthansa, and today's experience didn't help. Getting on the plane, I couldn't find my seat. There were no seat numbers to be seen:
|
|||||||||||||||||||||||
It turns out that you have to be less than about 1.70 metres tall to see the numbers:
|
|||||||||||||||||||||||
|
|||||||||||||||||||||||
On the flight to Athens with Olympic Airlines, I had been concerned that the food might not be good enough. That concern proved to be unfounded, but I was expecting a little more from Lufthansa than what I got:
|
|||||||||||||||||||||||
Baggage claim in the Nordic countries has one advantage that I would like to see elsewhere: a yellow “do not pass” line that keeps people from milling around the belts and blocking access to people who want to get to their bags. Unfortunately, it doesn't work as well as it should:
|
|||||||||||||||||||||||
It's funny that people who mill around the belts are also those who are prepared to wait for their baggage to go halfway round the belt before picking it off.
It's been 15 years since I was last in Sweden, and I had forgotten how expensive things are here; the taxi trip from the airport soon reminded me. Early to bed.
| Monday, 14 November 2005 | Stockholm | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Called up Mikael Ronström this morning for directions about how to get to work, and discovered that I was nowhere near and had to get the underground railway (Tunnelbana). On the way discovered that I'm not very well set up for this sort of thing. My laptop bag (two laptops) is heavy and on wheels, and neither buildings nor streets are very well suited. I've developed a new sympathy for people in wheelchairs.
At work, spent a lot of time talking not only about ndb but also about the structure of the company. Somehow didn't get through my mail, a sure sign of overload.
In the evening, went shopping for books. Last month Monty had shown me an interesting cook book, Bonniers kokbok, so bought a copy of that, and also a dictionary which showed me that the “Teach yourself Swedish” book that I had bought 2 months ago was even worse than I thought. The translations of Swedish words are so context-dependent that they're misleading and completely useless. It looks as if I'm going to have to mistrust it and look up every word.
| Tuesday, 15 November 2005 | Stockholm | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Finally got through my mail and started looking at the documentation I came here to look at. We've had a lot of useful discussions, but putting the documentation together is still a lot of work. We're missing a framework in which we can do it: it's easy enough to document individual functions, and what they're useful for, but that doesn't help us understand the engine any more than describing the structure of a tree helps us find a way through a dense forest. The forest analogy has another interesting aspect: once you've found your way through a forest, you think you've understood it; but there can be many other ways through the forest, and bears lurking behind trees. Understanding the forest is an order of magnitude more complicated.
It's getting colder now. Hopefully my clothing will prove up to the task. Ate at a Korean restaurant not far from the hotel, and then had a beer at a surprisingly cosy pub, also not far away, where it was quiet enough that I could read my book about transaction processing principles in peace. In general it's surprising how quiet it is in the middle of Stockholm.
| Wednesday, 16 November 2005 | Stockholm | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Being in an office without a phone is not a good idea when I have to use my Australian phone to make any calls. On the way to work this morning, looked for a prepaid SIM card, but all the mobile phone shops don't open until 10 am. Found a Pressbyrån outlet just round the corner from the office, and there bought a Comviq Kontant card for SEK 99 with an Amigos plan for another SEK 200. It turned out even cheaper than I expected: I can call most foreign countries, including Australia, for Kr 1.00 per minute, about AUD 0.17; that's barely more than I could do in the other direction from my cheap landline connection. Amusingly, calls to landline connections inside Sweden cost SEK 2.49 per minute.
It's getting colder now, overnight temperatures below zero. Jonas came in shivering and complaining about the cold, and the sight of my (bare) feet apparently made him feel even colder.
Spent some time talking about ndb; it's interesting how much of this resembles Tandem's architecture. Certainly puts me in a better position to understand it.
Lars Thalmann also came in today to say hello, and came along in the evening when we went to another pub, the Akkurat Belgo Bar, to meet Anders Karlsson and Kris Köhntopp. Despite the name, I ended up drinking some British cask conditioned ale, including a wobbly one from Marlow. Good food, but noisier than the Spitfire last night.
| Thursday, 17 November 2005 | Stockholm | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Spent a fair amount of time today talking with Jonas about specific issues that he's looking at, and with Mikael about general database design issues. Also more work on understanding ndb, which at this stage seems easy enough to understand. At least the people have gone to some trouble to document their work.
| Friday, 18 November 2005 | Stockholm → Uppsala → Stockholm | Images for 18 November 2005 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Off to Uppsala by train today, involving a half-hour wait at the Central Station; I should have checked the times before just going there. Through the frozen countryside to an equally frozen Uppsala. Fortunately the office (Bongårdgatan 8) is in easy walking distance from the station, though getting in was not that simple: there's a telephone system with instructions on only in Swedish. Met up with Mats Kindahl and the systems admin people there, also with Ingrid Göthe, who does the accounts. Some of the others I had expected to meet were out of the office, however.
Had a good talk with Mats, but didn't get much work done. At 16:00 we had another company teleconference, only the third. I've participated in the last two, both times from Scandinavia; in that connection it's amusing to think that only six months ago I was in Norway complaining about the length of the days. Now they're correspondingly short, of course. And I've been to Norway, Finland and Sweden in that time, which wouldn't be unusual except that I was in Australia between each visit. What interesting times!
| Saturday, 19 November 2005 | Stockholm | Images for 19 November 2005 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
In to town to do some sightseeing today. I've been to Stockholm before, but I can't recall anything in particular about the places I was in those days. Rather amused to come into the Storkyrkan to the sound “Dies iræ! Dies illa!” from Mozart's requiem (yes, I know the ! isn't written, but in that work it's definitely sung that way). They're performing it tomorrow, and this was a choir rehearsal:
|
|||||||||||||||||||||||
It sounded as good as it could without the soloists, so bought a ticket and will go back to hear the real thing tomorrow.
Walked around the Gamla Stan, making a point of getting a photo of Mårten Trotzig's Gränd for Yvonne's sake:
|
|||||||||||||||||||||||
I'm not sure what's so important about it.
After that to the Medeltidsmuseet, which had quite a bit to look at. One of the exhibits (about marriage in the middle ages) surprised me by having texts in Norwegian; the more I looked, the more it pointed to Bergen. When I got out, I discovered it was on loan from Bryggens Museum, only about 200 metres from Peter Hansteen's office, and outside which I stood six months ago. It's a small world.
| Sunday, 20 November 2005 | Stockholm | Images for 20 November 2005 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Up slowly this morning. It snowed last night, and the weather forecast suggested that it would be warmer. Out round midday to visit Skansen, marvelling at how low the sun is in the sky even at midday:
|
|||||||||||||||||||||||
It wasn't any warmer, so ended up in the Vasamuseet and spent a lot of time looking at the only exhibit, the Vasa, a 370 year old ship that sank on its maiden voyage and wasn't refloated until 1961. It's really in remarkably good condition:
|
|||||||||||||||||||||||
After that, back to the Storkyrkan to hear Mozart's Requiem and also a somewhat superfluous piano concerto (KV 482). Pianos in churches sound wrong, and this was no exception. The requiem was good, but in many ways I enjoyed the rehearsal yesterday better. Too many people present, for one thing. One nice idea was to bring out the bass trombone solo in front of the orchestra for the Tuba Mirum.
| Monday, 21 November 2005 | Stockholm | Images for 21 November 2005 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Somehow didn't have much to show for today. Spent more than usual time at work, and some of it talking to Jonas about aio, an issue that I had looked at earlier this year. It's surprising how much of the work I did at Rocksoft is relevant here.
| Tuesday, 22 November 2005 | Stockholm | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Quiet day today; nobody showed up in the office, and I spent some time catching up on other work, and some investigating an idea I've had for a long time, putting multiple tables into a “table space” (i.e. a big file or a raw partition) as opposed to individual files. This is quite close to the work I did on Monkey last year, but of course the constraints are different. Spent a while talking with Monty about the approach, and it looks as if I'll have an API pretty soon.
| Wednesday, 23 November 2005 | Stockholm → Frankfurt → Basel | Images for 23 November 2005 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Up at 6:30 this morning to catch a plane, and discovered to my surprise that the breakfast room was pretty full at this time of the morning. Off to Arlanda and thence to Frankfurt. This time the plane was really SAS and not a Lufthansa. I owe Lufthansa an apology relating to my comments on their food last week: for the same fare, SAS gave me exactly—nothing. There seems to be a more expensive economy fare that does give food, but that hardly seems to be worth it.
In Frankfurt, my bag was pretty much first off the belt. A good thing too: it took 28 minutes from when I landed. The people here seem to think that that's normal; I seem to remember that there are laws in place in Massachusetts that punish delays of more than 15 minutes for baggage delivery.
Arrived in Basel without finding anybody else, so since there's a music museum in Basel, off to see that. Quite a nice collection, I suppose, if you're not a specialist. They even put some instruments out to “play”, which a group of students did with great volume on various brass instruments. It seems that they had a class assignment, answering questions about the instruments; I was left wondering how effective it was.
| Thursday, 24 November 2005 | Basel | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
FreeBSD developer summit today. Not too many core developers were there: only two members of the core team had been registered, and neither showed up. Murray Stokeley turned up to be back in the USA, and presumably Thanksgiving played a role in his inability to find a way here. Robert Watson went to more trouble: he “shot himself in the foot“ by dropping a glass and then stepping on it, requiring 14 stitches. So about the only other kernel developer left was Poul-Henning Kamp.
Had a number of discussions, most of them not too interesting for me, though we did spend some time talking about the installation process. As usual, of course, came to no useful conclusion.
Also discussed some of my ideas about selective syncs; looks as if POSIX offers a solution, which I just need to implement.
Out to dinner with a whole lot of people; for some reason, everybody's following me.
| Friday, 25 November 2005 | Basel | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Today was the day of my kernel debug tutorial, so early through sleet and snow to the university. It's nice to have free tram services here, but in weather like this you get to appreciate the comfort of having the conference in the same building.
The tutorial itself went well, though I was once again not able to get firewire debugging working. I have the horrible feeling that somebody has broken it and nobody (obviously including myself) has both time and inclination to fix it.
Out for dinner in the evening, again being followed by a large group of people, nearly 20. It's not just counterproductive to have such big groups; it makes it very difficult to get a table, and the first restaurant just threw us out altogether because they didn't have the place. The next place had the same welcome as the first: “Es wäre vorteilhaft gewesen zu reservieren”. Got a couple of tables and had a fondue whose main virtue was to emphasize how much better the others were that we had last year. Lots of talking, relatively late to bed.
| Saturday, 26 November 2005 | Basel | Images for 26 November 2005 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
No clothes to wear! Vera had arranged to have my linen laundered, and she told me she was going to return them yesterday afternoon, but the hotel reception told me it hadn't arrived. Grrr.
To the conference, hearing many things I've heard before. I don't know if it was that or the lack of sleep, but I was pretty tired, and I nearly went to sleep in the most interesting talk of all, about the future of the X window system. Off back to the hotel for a rest, first talking with Vera, who told me that she had, indeed, dropped off the clothes last night. Back at the hotel they again said that they weren't there, and it wasn't until I insisted that they looked around and found them. They weren't even very apologetic: “Well, you have your clothes, so all's well”.
Conference dinner in the evening, in a vaulted cellar not far from the university:
|
|||||||||||||||||||||||
As so often, the music drowned out the talking, so didn't stay long. I think everybody would be happier at these events if there were no music at all: then we could talk instead.
| Sunday, 27 November 2005 | Basel | Images for 27 November 2005 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Second day of the conference started with Ollivier Robert talking about Mercurial, which sounds interesting. Unfortunately he rather overestimated the time at his disposal, and most of the talk was about the (manifold) shortcomings of other version control programs.
Apart from that, didn't do too much; accidentally missed the only other paper that had really interested me, Ted Unangst's talk on a new thread library. I'll have to chase that one down later.
In the evening, yet again to the Rheinfelderhof, the restaurant that I had found last Wednesday, and which we have been visiting ever since. Pleasant enough place, and the service was good enough that (very much out of keeping) I encouraged everybody to give them a tip.
| Monday, 28 November 2005 | Basel → Carnate (Milano) | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Today was the day planned to travel to Milano by car with Massimiliano (Max) Stucchi and George Georgalis. Max woke me up early: since yesterday evening they had changed their plans, and we were off by 9 am, heading first for Zürich, mainly so that George could see it. On to Milano, stopping just before the border for George to do some shopping, though ultimately he didn't buy much.
In the evening to Montevecchio to look for a restaurant that Max knows, which unfortunately was closed. Found another very pleasant one nearby, one that put to rest the idea that Italian main courses are always small; I wasn't able to finish mine.
| Tuesday, 29 November 2005 | Carnate (Milano) | Images for 29 November 2005 |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
As if to prove Max wrong with his claims about North Italian weather, it snowed all morning today, something that's really unusual. The heating didn't keep up with it well.
In the evening off to check out a shopping mall near where Max is thinking of holding the next BSDCon Europe. Borrowed Max's mother's car, and on the way had to fill up. The filling station was unmanned, and the currency reader had feed problems, so tried to use my Australian debit card. It returned the card (good thing) and then hung displaying “validating card” for several minutes, during which we discovered that we could do nothing to cancel the attempt. We obviously couldn't leave either; if the validation had succeeded, I would have ended up footing the bill for somebody else. Fortunately it timed out after sometime, apparently resetting the machine in the process. What a mess!
| Wednesday, 30 November 2005 | Carnate (Milano) | |
| Top of page | ||
| previous day | ||
| next day | ||
| last day |
Into the office early this morning: Max had borrowed his mother's car, and she needed it back early. Spent much of the morning arranging forward travel plans: decided to go to Toscana tomorrow, and spent an inordinate amount of time negotiating broken car rental web sites, finally succeeding with Avis (after trying very hard).
Later browsed through Max's storeroom—he sells O'Reilly books—and found one that I didn't know, MySQL in a nutshell. It looked quite good, and I decided that I should probably buy it when I get home. Then out of curiosity checked the name of the author: Russell Dyer. I know him: he's a MySQL employee. That's not surprising, but it reminded me of the fact that I exchanged mail with him last week: he lives in Milano, and I had arranged to meet him here. That only left today, so called him up and arranged to have lunch with him.
It turns out that Russell was a victim of the destruction of New Orleans, and that he spent a month living with Andy Oram in Arlington before moving to Milano. He's not finding it easy to settle in here, so arranged to take him to Toscana tomorrow.
On the way back decided to pick up the car today, saving the trouble of two return train journeys. I didn't want to bring all my baggage on the train, so I would have had to go back to Carnate anyway. It's good to be in control of my own movements.
Do you have a comment about something I have written? This is a diary, not a “blog”, and there is deliberately no provision for directly adding comments. It's also not a vehicle for third-party content. But I welcome feedback and try to reply to all messages I receive. See the diary overview for more details. If you do send me a message relating to something I have written, please indicate whether you'd prefer me not to mention your name. Otherwise I'll assume that it's OK to do so.
| Top of page | Previous month | Greg's home page | Today's diary entry | Next month | Greg's photos | Copyright information |
