Friday, 1 August 2003	Sydney	Images for 1 August 2003
		Top of page
		next day
		last day

Up relatively late this morning; it's nice that our meetings don't start until 10 am. Met Michael Still in the lobby of the hotel, and soon David Purdue came along and took us by train to the Sun office, where we had our Chapter Council meeting. An interesting enough event, and we came away with a number of ideas about how to revitalize the chapters.

Lunch was Yum Cha at a restaurant not far away, where I managed to crown myself in glory (but fortunately not soya sauce, which just missed me) by spinning the lazy Susan a little too fast.

After the meeting to the Duxton Hotel to look at the rooms for the AUUG conference, after which off to look for a nice location for the speakers' drinks on 2 September. Found a couple of nice places at the very base of the north side of the Sydney Harbour Bridge, but unfortunately they didn't want us. Looked at a couple more, both of which might be possibilities, and then to dinner at Had To Happen, American Mexican food. It doesn't measure up to the Texan places.

Saturday, 2 August 2003	Sydney –> Echunga
		Top of page
		previous day
		next day
		last day

Spent the day in the AUUG Board Meeting at ActiveSky's Sydney office. About the only thing of interest to report is that we got the agenda finished on time; minutes to follow, of course.

Off to the airport with Michael Still, where we looked at the wireless networking in the Qantas Club. In the past, it had been done in an astoundingly incompetent manner by SkyNet Global—I have reported on the problems in this diary on a number of occasions.. Since then, it has been taken over by Telstra, who appear to be continuing in the fine tradition of SkyNet Global. Admittedly, firing up a browser brought a much more detailed web page explaining that the service is free until 27 August, but I still needed authorization, which I could get either by SMS to my mobile phone, or by credit card (does that sound warning bells?). I tried both: the web page told me an SMS message had been sent, but nothing arrived. Tried the credit card method (“15 minutes: $0 ... 120 minutes: $0”), but that caused the authentication application to crash. Played around and discovered that the authentication server is now 6 hops away, over a routed RFC 1439 network (alternating 10.x.x.x and 144.x.x.x networks), and that DNS still worked, but only via uneeda.telstra.net. Considered the potential for abuse of that particular service, but I think it might be relatively secure.

Sunday, 3 August 2003	Echunga
		Top of page
		previous day
		next day
		last day

A slow day catching up (also slowly) on my mail, but only managed to get down from 5,500 messages to 2,500. There must be a less tiring way.

Bernd Wulf over in the evening with a computer and a satellite receiver card. It looks like he's right: the card is dead. I'm a little paranoid about saying things like this, mainly because I don't want to look silly, but if this card is functional, it's doing a very good job of hiding it.

Monday, 4 August 2003	Echunga
		Top of page
		previous day
		next day
		last day

Still in catch-up mode. I suppose I'm going to have to accept the fact that the idea of letting mail pile up and then removing it in a binge may be attractive, but it doesn't make for responsiveness. By the end of the day I still had nearly 1000 messages hanging around.

Back to work on my debugging tutorial. I have 50 pages already, and it looks like becoming more. A little more and I can make a book out of it.

Trying to make satay (Malay skewered meat with a hot peanut sauce) this evening. The satay itself was OK, though the meat you get from the butchers here is far too thick, but the ketupat (rice cakes, served with the satay) continue to be impossible. This time tried putting Arborio rice in a plastic bag, perforating it and stapling down the opening. The bag split, and the rice tasted terrible. Still, that's what experiments are for.

Tuesday, 5 August 2003	Echunga
		Top of page
		previous day
		next day
		last day

Finally somebody has done something more effective against SCO and their Linux farce: Red Hat has filed a complaint about FUD. Didn't read all the text, but it looks eminently sensible, a far cry from SCO's complaint in March.

While thinking about that, noted that SCO keeps harping on the “enterprise” aspects of their operating systems. It occurred to me that, back in the days when they had been a reasonable player in the industry, they had given me some free CD-ROMs, so thought of installing them on my AMD machine to see how they measure up: it would be funny if they didn't do as well as Linux. Neither CD-ROM was bootable, so I had to use boot floppies. What a pain. The newer one (UnixWare 7.0.1) didn't finish booting before hanging the system, so that was literally a non-starter. The older one (2.1.2) booted and asked me for an activation key. I wonder where I put that. Did a little looking around, and then gave up. Yet another nuisance of proprietary software, though I suppose Microsoft people are used to it.

Spent the rest of the day working on my tutorial notes. I'm pushing 60 pages, and there's enough meat in there for a couple of hundred, making it a book, but obviously I'm going to have to cut things short if I want to have something distributable next week.

Wednesday, 6 August 2003	Echunga	Images for 6 August 2003
		Top of page
		previous day
		next day
		last day

It's interesting to note how the animals round our pond interact. Our duck Daisy is gone, but Donald is still there, and he seems to be attracting wild ducks, specifically Pacific Black Ducks, which are much smaller. We've been putting him away over night to ensure his survival, and this morning one of the moorhens was overjoyed to see him again. Strange.

More stuff about SCO today. Called up Kieran O'Shaughnessy of SCO Australia, who confirmed that they intend to introduce their silly licensing thing in Australia next week, and that he would keep me informed. Called up the Australian Consumer and Competition Commission again, and spoke to Nick, who said they were getting a lot of interest, and that they want to hear from anybody who is approached by SCO for royalties.

In the afternoon, some people came along offering to clean out the water tank. Not the sort of thing that many people have to do. They apparently found quite a mess in there.

Image title: water tank          Dimensions:          2048 x 1536, 1531 kB Make a single page with this image Hide this image Make this image a thumbnail Make thumbnails of all images on this page Make this image small again Display small version of all images on this page All images taken on Wednesday, 6 August 2003, thumbnails          All images taken on Wednesday, 6 August 2003, small Diary entry for Wednesday, 6 August 2003 Complete exposure details

 

Thursday, 7 August 2003	Echunga
		Top of page
		previous day
		next day
		last day

Finally mail is getting back to manageable proportions. I'm trying to cut down by 50 messages every day now, and today got down to 250. I think I should keep a goal of not more than one page of mail displayed (on my main screen that's 86 messages).

Apart from that, working on the course notes for the debugging tutorial. I should have something useful to present next week, but it'll still look more like an advanced draft than a book. Somehow, though, that's par for the course for debugging stuff.

Friday, 8 August 2003	Echunga
		Top of page
		previous day
		next day
		last day

For no apparent reason, wantadilla panicked this morning. After rebooting, I got a SIGBUS from Emacs, both things that seem pretty unusual. Decided that software doesn't really rot, and the last time wantadilla went down was a few months ago, for reasons not related to reliability. Started looking at other issues, such as processor temperature. Found xmbmon again—the last time I used it was a year ago, and the name somehow isn't very memorable.

The results were interesting. Here's last year and two examples from this year. Same machine, almost no changes:

Note that the temperature scale is in units of 12°. I've changed it to a 10° scale since then, but I reverted for purposes of comparison. It looks like time for a new power supply.

Things are happening thick and fast on the IBM front. Finally Big Blue has filed a countersuit against SCO. It's interesting to note that they're accusing them of violation of the GPL, and they have terminated SCO's rights to redistribute IBM's software because of the alleged breaches. That's an amusing one. I've never been too happy with putting the GPL to the test in court. I prefer to look upon it as a gentlemen's agreement, and I suspect that stringent examination in court may damage it. But IBM's in favour of the GPL, so it's unlikely to have taken these steps unless it was pretty sure that it would come through. Groklaw, which appears to be somebody's pseudonym, has done quite an interesting analysis.

Apart from that, more work on the tutorial notes. Pain.

Saturday, 9 August 2003	Echunga
		Top of page
		previous day
		next day
		last day

More breakage in Vinum on FreeBSD: changes elsewhere in the kernel have caused a number of panics. It would be really nice to know about these changes ahead of time so that I could change things before they break things for people. In at least one case it appears to have wiped out a Vinum configuration. Spent some time bringing zaphod up to date, not helped by the fact that it still can't build a world without hanging several times. It's about time to get another machine, I fear: I can't imagine that all SMP boards work this badly.

It's been 30 years since I last brewed my own beer, but at the time it was quite good. I was studying Chemical Engineering at the time, so I put a little more effort into it than the can and bucket brigade. Since then, I haven't found time to do it, but things have changed, and it might be fun to try again. As a result, out in the afternoon to Verdun to visit Grumpy's Brewhouse, which turned out to be a pub as well. They seem to go to a little more trouble than most. I got a package deal with a brew starting with a Coopers' “Munich Lager” brew kit, they supplied a supplement bag to make it taste more like beer. Also got some real yeast. How can Coopers' expect you to brew lager with a yeast that ferments at 25°?

Back home, spent surprisingly little time starting my first brew since 18 November 1972. The wort looks decidedly too dark for the style of beer. We'll see.

For some reason, didn't get much further with the course notes.

Sunday, 10 August 2003	Echunga
		Top of page
		previous day
		next day
		last day

zaphod died yet again building a world. I really should find out if there's something wrong with the motherboard. Gave up and rebooted monorchid and started there instead. Just building software takes too long, especially now that we're using gcc release 3.

On with the documentation. It's surprisingly tiring. I'll make the deadline, but only just.

Monday, 11 August 2003	Echunga
		Top of page
		previous day
		next day
		last day

Today everything and nothing happened. They have published a scan of the IBM countersuit on the web, so spent a lot of time writing an analysis. Also a number of interruptions: Berndt Wulf in to pick up a new satellite receiver card (the first one he got was DoA). Also decided that it was time to find a second ABit BP6 motherboard to decide whether zaphod's problems were due to hardware or software. Sent a message out to the local mailing lists, and got a number of replies almost immediately. It's amazing how many people have old hardware lying around. Within hours I had not only arranged to buy a new board, but had it delivered to my door.

In addition, a lot of stuff to do for AUUG 2003, as well as a planned new miniconference for the Linux.Conf.Au conference next January. If that weren't enough, Jorge de Moya called up and invited us to dinner. Ended up with him staying here. Pleasant evening, but somehow I didn't get any time to work on tomorrow's deadline for the tutorial notes. sigh.

Tuesday, 12 August 2003	Echunga
		Top of page
		previous day
		next day
		last day

Last day for the tutorial notes! Dropped just about everything else and tore through the document so that at least it looks consistent. We're now up to 70 odd pages of A4; in a book it's more likely to be 100 pages. Hopefully I'll be able to make some reasonable slides for the tutorial.

Wednesday, 13 August 2003	Echunga
		Top of page
		previous day
		next day
		last day

One of those tidy-up days. The tutorial notes went out in the morning (yesterday in the USA), but all sorts of other things kept me busy. Put the new BP6 motherboard into zaphod, with the encouraging result that it seemed to work fine: the problem must have been with the old motherboard after all.

In the evening to the AUUG SA Chapter meeting, where David Lloyd did a presentation, significantly better prepared than last time, about apt and other strangenesses.

Thursday, 14 August 2003	Echunga
		Top of page
		previous day
		next day
		last day

More drudgery. Not much to show for it, but at least I don't have as many things on my plate as I did a few days back. I'm gradually cutting back on my outstanding mail list, which had the interesting effect that replying to some stuff I've been avoiding generated a whole lot more mail; the AUUG lists alone must have accounted for 30 messages.

Friday, 15 August 2003	Echunga	Images for 15 August 2003
		Top of page
		previous day
		next day
		last day

I've decided: no matter how difficult it is to keep up to date with mail every day, it beats the hell out of batching. My experiments over the past couple of months made it easier to get other work done, in the short term, but they make it really difficult to catch up. In addition, the amount of mail that my replies generate suggests that I've been deferring important mail after all.

More problems with wantadilla today, so changed the power supply as planned. The core voltage didn't look any more stable, and I notice that there was nothing untoward with the voltages which I had been logging. It panicked shortly afterwards again, and this time I noticed that the fan on the northbridge wasn't running. On closer investigation, it had almost completely seized up. Took a look in my junk box and, to my surprise, found a functional fan of the right kind, screwed to a different heat sink. It seems that these fans are attached with self-tappers.

After that, things worked again, and I started writing up a new chapter for my debugging tutorial for debugging this kind of dump. Off to dinner, and came back to find the machine had rebooted again. The heat must have fried the northbridge, all the more plausible in view of the fact that disk resets were taking up to a minute on random occasions. Gave up on that and put wantadilla's disks, Ethernet card and display cards into monorchid and left it like that for a while. Looks like we need new hardware.

In the evening, more experimental cookery. It's interesting that, though the Chinese probably invented pasta, I have never seen anything like lasagne in Chinese cookery. The Italian style with lots of béchamel doesn't sound appropriate, but we have a recipe by Paul Bocuse with salmon, spinach and broth, which tastes very good, so tried adapting that to Chinese cooking styles, with surprising success. Watch this space for the recipe.

Saturday, 16 August 2003	Echunga
		Top of page
		previous day
		next day
		last day

Came in to the office this morning, and the new wantadilla had rebooted during the night, without leaving any evidence as to why it has died. Grrr. Did a bit of wondering whether I was still on the wrong track, but the old machine (now in the position of daikon and monorchid) still seemed to be sick. Spent some time trying to upgrade daikon, during which the system hung multiple times. The new machine, on the other hand, seemed to be stable enough, but with only 128 MB of memory it was remarkably slow. The old machine (512 MB) had different RAM, so I couldn't use that; instead raided zaphod for another 128 MB, after which it ran reasonably well, but put most of my test equipment out of commission.

Took the opportunity of a couple of dumps from wantadilla to write another case study for the kernel debug tutorial. It was interesting: somehow the system had been running with a stack in a completely wrong location. Normally it grows down from 0xc1000000, but on these dumps it was round 0xdd000000. Didn't finish the analysis; too much else going on.

Sunday, 17 August 2003	Echunga
		Top of page
		previous day
		next day
		last day

Spent today doing much the same as yesterday. Tomorrow's the deadline for the Call for Papers for Linux.conf.au, so put in a few suggestions. It'll be interesting to see which, if any, they choose. Apart from that, which didn't take too long, carried on trying to upgrade the sick wantadilla, now masquerading as daikon, to the latest NetBSD-CURRENT, and finally succeeded. It's interesting to note how it sometimes works for long times on end, and at other times it fails repeatedly at short intervals.

Also finished working on the dumps from wantadilla, and came to some interesting conclusions which, in retrospect, were all wrong. Looks like I'll have to have another go.

Monday, 18 August 2003	Echunga	Images for 18 August 2003
		Top of page
		previous day
		next day
		last day

More conference stuff today. Why do we always have these last-minute problems with the proceedings? Spent a lot of time looking at that.

Also got some time to look at what I had written yesterday about the debug tutorial. I wonder what I was smoking; my analysis was all wrong. That's certainly a good reason not to do this live.

Yvonne back in the evening with new hardware for wantadilla: an MSI K7N2 Delta-L with an Athlon 2500XP+ (Barton). Put it together and discovered I couldn't insert the AGP display board: it's a Matrox G200, only 3 years old, but already obsolete (wrong voltage). It's amazing how quickly hardware is changing. Found that the “spage” board in daikon was in fact a low-voltage AGP board, and with a bit of fiddling with the XFree86 configuration, managed to get it working. I would have needed to change things anyway: it looks as if this board has two PCI busses, and both displays had different identifiers. Apart from that, the system ran fine, but once again (see the last time), the processor information was not consistent with the processor: it claimed to be an 1100 MHz machine, which I suppose corresponds to an Athlon 1600XP+, if there were such a Barton processor. Played around with the BIOS, which for some reason had set a 100 MHz FSB, and set it to the correct value. Voilà! It claimed to be an Athlon 3200XP+, and ran happily as such.

This is ridiculous. I didn't set out to overclock the processor; it just turned out that way. On the other hand, what's the issue with overclocking? Overheating. So I fired up xmbmon and found that the temperature was much lower than the old wantadilla with an Athlon 1700XP+. Left it running and found more interesting details: it runs much cooler with the cover off. Here are four temperature measurements with the cover off, the cover on, the cover slightly open (which makes it worse) and the cover off again:

Tuesday, 19 August 2003	Echunga	Images for 19 August 2003
		Top of page
		previous day
		next day
		last day

Things are ramping up to the AUUG conference, and the time I thought I had didn't quite materialize. Spent the day fighting various fires; in addition, we're planning a miniconference for Linux.Conf.Au, and today was a status meeting.

Despite all that, found time to try putting together other machines. I seem to be low on display boards, and now that the ISA standard is effectively dead, I can't even use my old, mouldy ISA display boards. zaphod, one of the Abit BP6 motherboards, has a slot, and it can drive an ISA display board, but only in conjunction with an AGP board, which is somewhat counterproductive. Played around with the old, fanless Pentium (ex kimchi) which I wanted to use to replace the current satellite gateway box. It was supposed to be diskless, but it needed swap, and it seems that the IDE interfaces on the board are dead, probably the reason I used SCSI in kimchi. Running SCSI with Linux is its own fun, so gave up on that.

Wednesday, 20 August 2003	Echunga
		Top of page
		previous day
		next day
		last day

Another day with so many distractions that I couldn't get any work done. Yesterday evening, somebody told me that Heise Verlag had published photos of SCO showing alleged UNIX code in Linux. Took a look at it, and it was obvious that they were wrong. Spent a lot of time refuting that.

After that, working on the conference proceedings. For rather complicated reasons we only have them in PostScript, and they needed fixing. Spent a lot of time doing that, and in the end agreed with the printers to bring the stuff in first thing tomorrow morning and sort things out.

Later checked on the Heise forum about SCO, joined up and made a comment, and suddenly I was an expert. Sent a message to the author, who turned out to be Jürgen Kuri, and had a bit of a discussion, in which it eventuated that Bruce Perens had also written an article claiming it came from the Third Edition. Spent some time looking at that, decided that Bruce was wrong, but that so was I, and discovered to my astonishment (with a little help from a friend) that the code did, indeed, come from System V.4. There's no reason to believe it was stolen, especially since it's really almost identical to the code in Research UNIX, but it's pretty clear where it came from. Spent a lot of time researching that, late to bed, and almost immediately up again to update my web page with a refutation. What a day! And not a stroke of work on my tutorial, for which I fortunately have time until the end of next week.

Thursday, 21 August 2003	Echunga
		Top of page
		previous day
		next day
		last day

Into town this morning to take the conference proceedings to the printer. I had know that the PDF had problems, so I brought the PostScript version as well. They took the PostScript version and... converted it to PDF, recreating the same problems. After a lot of difficulty, discovered that we could print it out on a LaserJet with PostScript, but that their toy Microsoft box couldn't spool correctly. sigh

Back home and started printing out the PostScript; had no less than 10 pages which the printer rejected. Spent most of the day printing it out before sending Yvonne off to take most of the pages in. What a pain. We must look more carefully at the software we use for creating these documents. I wonder if there's a PostScript validator somewhere.

Apart from that, writing up more on the SCO story. It's getting more and more interesting, but it's also taking up a lot of time. What with starting a new batch of beer and Essey Deayton coming along with a new digital video camera (with USB, which didn't react at all on my laptop), finished another day with no time to look at the tutorial notes. I've been working on one stack frame since Sunday.

Friday, 22 August 2003	Echunga
		Top of page
		previous day
		next day
		last day

As a result of the last couple of days' activities with SCO, my carefully pruned mail had grown to nearly 2000 messages again. Spent most of the morning trying to limit that, not helped by the fact that some new Microsoft exploit is around, and it's generating mail messages like fury. The annoying thing is that many of them spoof my address as the sender, so I end up getting all sorts of error messages back from remote MTAs. In addition, it looks as if the virus scanner people have teamed up with the spammers: I had dozens of messages telling mailing lists that they had sent spam. It's time for legislation to require people to stay off the Net if they're infectious. Removing all Microsoft equipment from the Net would be a good start.

Then into town for an ADUUG lunch, which was more active than most, and ended up walking down Rundle Street with John Perry looking for active 802.11 networks–with a Microsoft box running dstumbler.

Back home and still in catch-up mode. New things are happening all the time. Today's interesting recognition is that it would seem that SCO have incorporated the Berkeley Packet Filter without acknowledgment: they used BPF code as their second example of “stolen” code in the slides they presented earlier this week. What a success: they presented two examples, one of code which is no longer in production versions of System V, but in SCO, and also no longer in Linux. The other is of BSD code which they haven't noticed is not their own. Bravo, SCO!

Saturday, 23 August 2003	Echunga
		Top of page
		previous day
		next day
		last day

The SCO affair is still keeping people busy, and spent some time on that, but finally got my mail down to manageable proportions. Also more work on my presentations: I now have a week to finish my tutorial notes, write two columns for AUUGN and do two sets of slides for my presentations in the following weeks. At least managed to get the slides started, and made surprisingly good progress. It's still looking to be a pretty busy week to come.

Sunday, 24 August 2003	Echunga
		Top of page
		previous day
		next day
		last day

Things are getting a bit quieter now (the lull before the storm?). Mail is back to manageable proportions, and the SCO affair has slowed down for the weekend. I wonder when the mainline press will seize on the likelihood that they have violated the BSD license. Spent most of the day working on my slides for the AUUG conference, and to my surprise finished them. There's a serious problem with the rendition of EPS images in PDF, though, and I'm going to have to think about how to handle that.

Monday, 25 August 2003	Echunga
		Top of page
		previous day
		next day
		last day

Still lots of things happening, but I seem to be on top of problems again. Looking back at the experiment with batching mail, I think it's fair to say that it didn't work. It's convenient not to have to read mail every day, but there's always this worry hanging over you.

The approaching conferences took up a lot of time, but I did find some time to work on the tutorial notes, and finally finished analysing the stack frame that I started on the 17th. I wonder if the tutorial participants will find this as tiring as I do. But then, hopefully we won't have the same number of interruptions.

Tuesday, 26 August 2003	Echunga
		Top of page
		previous day
		next day
		last day

How bad does spam have to get before the law does something about it? I've successfully blocked a number of domains which seem to send only spam. One is Road Runner, owned and operated by Time Warner Cable. They don't annoy me directly any more, of course, but for months they've tolerated a spammer, apparently on a fixed IP address, who spends all his time forging other people's sender addresses:

Date: Wed, 27 Aug 2003 11:22:56 +0930 (CST)
From: MAILER-DAEMON@lemis.com (Mail Delivery System)
To: postmaster@lemis.com (Postmaster)
Subject: Postfix SMTP server: errors from 653270hfc158.tampabay.rr.com[65.32.70.158]

Transcript of session follows.

 Out: 220 echunga.lemis.com ESMTP Postfix
 In:  EHLO GARY
 Out: 250-echunga.lemis.com
 Out: 250-PIPELINING
 Out: 250-SIZE 10240000
 Out: 250-ETRN
 Out: 250 8BITMIME
 In:  MAIL FROM: <jbs@quiotix.com>
 Out: 250 Ok
 In:  RCPT TO: <grog@lemis.com>
 Out: 504 <GARY>: Helo command rejected: need fully-qualified hostname
 In:  QUIT
 Out: 221 Bye

These go on at the rate of hundreds a day, and though none get through, I do get bounce messages indicating that he's been abusing my name too:

Return-Path: <grog@lemis.com>
Received: from GARY (653270hfc158.tampabay.rr.com [65.32.70.158])
    by penguin.postmodern.com (8.12.8/mcb-20021115-1) with ESMTP id h7P3JwPw030805
    for <Majordomo@GreatCircle.COM>; Sun, 24 Aug 2003 20:19:59 -0700
Message-Id: <200308250319.h7P3JwPw030805@penguin.postmodern.com>
From: <grog@lemis.com>
To: <Majordomo@GreatCircle.COM>
Subject: Thank you!
Date: Sun, 24 Aug 2003 23:20:08 --0700

In case there's any doubt, this is not me. The trouble is that most people won't know that. I wish I knew how to get them to stop. You'd think that a company like Time-Warner would have some integrity, but mail and faxes to them have been without effect. It looks as if the only thing which will work will be DDoS against the provider or the spammer.

Much the same work as yesterday: preparing for the conference, finishing my tutorial notes, for some definition of “finishing”: this sort of thing never gets finished.

Wednesday, 27 August 2003	Echunga
		Top of page
		previous day
		next day
		last day

More fun this morning with our joint conference with Linux Australia. Why do things need to be so complicated?

In the afternoon, into town to Parliament House to meet The Hon Ian Gilfillan, who has recently submitted a bill mandating a preference for open source software to the parliament. Last week we (AUUG) issued a press release opposing the bill on the grounds that preferences are bad in either direction. The press, of course, had a field day, and it seems that we should have discussed the implications more carefully. It turns out, of course, that we're really aiming for the same thing in a different way, and so we'll work more closely towards the next version of the bill, due to be submitted in a couple of weeks. Quite a successful meeting.

Thursday, 28 August 2003	Echunga
		Top of page
		previous day
		next day
		last day

With the impending arrival of the two back-to-back conferences, I had intended to slow down today and take it easy for a day or two before life goes crazy. It didn't happen.

Mail's crazy again. Got back to my goal of under 86 messages (the number that fits on a single page of my xterm), but it kept me busy. In addition, Duncan Barclay has fixed his Broadcom 4400 Ethernet driver, so installed that on adelaide, my Inspiron 5100 laptop, and confirmed that it worked. Also tried to finish my kernel debugging tutorial, and more or less managed that.

That only left the president's column for AUUGN, the president's report for the AUUG Annual General Meeting, and a book review of “Jeff Duntemann's Drive-by Wi-Fi Guide” (summary: I don't like it). Got started on all three, but somehow ran out of time.

Friday, 29 August 2003	Echunga
		Top of page
		previous day
		next day
		last day

Hah! The very idea of wanting to slow down before the conference! Everything happened today, and I spent half the day on the phone trying to solve various crises. I've given up hope of getting my AUUGN articles finished before the conference, though I did manage to more or less complete my president's report for the annual general meeting, so I might even have a little time to spare on Tuesday.

I'll be glad when I'm back from the USA.

Saturday, 30 August 2003	Echunga
		Top of page
		previous day
		next day
		last day

More hard slog today. Things can't go on like this–they'll probably get worse, at least in the week to come. After that I really need to slow down.

Did manage to find some time for a job I had been planning for months: taking some of the eucalyptus trees that had sprung up unannounced round the driveway and planting them in the paddocks, which could do with more shade. Also planted some shrubs in the garden. I should have more time for things like that.

Sunday, 31 August 2003	Echunga –> Sydney
		Top of page
		previous day
		next day
		last day

For once, I was relatively prepared for my trip to Sydney today, and spent the morning bottling beer and starting a new experimental brew instead. Sent Yana off to pick up some satellite equipment from Bernd Wulf in Nairne, which took two attempts to complete, but apart from that all went smoothly.

In the afternoon to Sydney to the AUUG conference and met with Frank Crawford and Joel Sing, who were trying without spectacular success to set up the satellite downlink. At the point where we gave up, we didn't even have the Ethernet link from the roof to the ground floor working reliably.

Off with a number of people looking for food, and to an Indian restaurant with a big sign outside offering goat, kangaroo and crocodile curry, but had a more conventional “banquet” which was, by Australian standards, not too bad.

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.