Diary of a geek

April 2005
Mon Tue Wed Thu Fri Sat Sun

Andrew Pollock


Other people's blogs


RSS feed

Contact me

JavaScript required

Monday, 25 April 2005

Garbage bags are quite buoyant

The problem with having a leftover tank of helium in your garage on a public holiday is that you have to inflate things. Since I didn't have any balloons, the next best thing was garbage bags.

I inflated one, and to my surprise it was a lot more buoyant than I expected. I was planning on taking it out the back and expecting it to hover to about the second story balcony or something, instead it bounced off the eaves, and ascended into the heavens and was taken by a cross-wind off in the direction of Mount Majura, rapidly disappearing out of sight.

Sarah thinks I'm acting like a 12 year-old, but how many 12 year-olds have a tank of helium in their garage?

I do feel guilty about letting loose a flying garbage bag though... If it wasn't littering, I'd do it again so I could get some photos...

[00:20] [life] [permalink]

Saturday, 23 April 2005

Well, that happened... (or reflections on the conference from a delegate's point of view)

Okay, now that I've braindumped about organisation stuff, I'll braindump about the conference in general (what I experienced of it).

The (warm body) networking was the best part for me (again). It was terrific that so many overseas Debian and Canonical/Ubuntu people were here this year (or was it just that since LCA 2004, and probably more importantly, Planet Debian, I recognise more names?). It was great to meet for the first time Scott James Remnant, Colin Watson (who has a totally awesome accent), the much maligned James Troup (who I didn't get an opportunity to buy a beer), Mako, Matthew Garrett, Matt Zimmerman, and probably a whole bunch of other people that I've forgotten to mention.

One of the definite highlights for me was the opportunity to have a one-on-one chat with Mark Shuttleworth. He is one exceptional person. He's got himself one metric spankload of money, but he's doing some really good stuff with it, rather than just pissing it up against the wall being an uber-rich dude.

He laid out his vision (and it really is visionary stuff) for where he wants to take Ubuntu and what he wants to do, and I was really impressed with the breadth, depth and clarity of what he had to say for himself. He knows exactly what he wants to do and how he wants to do it, and he's got the money to make it happen. Totally inspirational stuff. As I have said before, I think I need to jump on the Ubuntu bandwagon.

I also made his talk about going to space, and that was truly amazing. Again, here was a guy with a metric spankload of cash, and rather than just paying his way into it (granted, he did part with a wad of cash to get in) he went through all the rigorous training, and really became a cosmonaut, complete with a mission to accomplish while he was up there. I really don't think "space tourist" is a terribly accurate definition for him. From the sounds of it, it took some real determination on his behalf to get to where he got. He told his story really well, and you can tell he really enjoys talking about it. I hope that one got video recorded successfully, as I really want Sarah to see it.

I didn't catch a lot of Eben's talk, which was one I really wanted to catch, because I was running around trying to deal with drinks for lunch, which had been overlooked. Everyone was raving about him and his talk though, so that's another one I hope I can catch on video.

Unfortunately I didn't see a lot of what I really wanted to catch, which was the Debian Miniconf. I came in the tail-end of Mark Shuttleworth talking about Ubuntu and Debian. I suspect it was a similiar spiel to what I'd had when I spoke to him earlier, so hopefully I didn't miss too much. There was just too much initial registration stuff and general firefighting to do to allow me to have the first two days totally not doing organisational stuff. Oh well. I should have seen that coming.

I caught bits and pieces of Ted Ts'o's Recovering from Hard Drive Disasters tutorial, and what I caught was pretty cool. I missed the Bitkeeper part of Tridge's keynote, which was right towards the end, because I was doing morning tea preparation stuff, but based on some of the media coverage, it sounded interesting. Hopefully that one was recorded okay as well. I think I caught bits and pieces of Jeremy Allison's CIFS to the UNIX Desktop talk, but kept getting dragged out to attempt to deal with the issue whereby some flog was running a rogue wireless access point, and doing all sorts of nasty man-in-the-middle attacks on people. That really pissed me off (the fact that someone came to the conference and did that). Unfortunately due to the nature of wireless LANs, we really couldn't do a lot about it, but there was a small lynch mob of geeks (myself included) running around for the remainder of the day running iwlist scan on their laptops non-stop, attempting to get a whiff of the bastard again.

I was really looking forward to JB's talk about Asterisk. As it turned out, I had done my RHCE course with him last year in Brisbane (small world). The talk was disappointing. JB was an inexperienced speaker (but it is good to give those types an opportunity to improve) and his talk wasn't technical enough, and a lot of people actually thought he was trying to sell Asterisk, and it was perceived as being too salesy.

I successfully caught all of Martin Pool's talk about Bazaar-NG, and it was really excellent. (I still don't have the whole GNU/Arch, tla/baz/Bazaar/Bazaar-NG thing 100% clear in my head though, not being a really big user of revision control systems).

I also caught all of the OzTivo talk, unfortunately not realising it was on at the same time as Marc MERLIN's talk about spam evasion with Exim. Fortunately I did have a bit of a chat with him at the Professional Delegate's Networking Session, and he's convinced me that I need to give Exim a thorough investigation.

Andrew Morton's keynote on Friday was good. I was really interested to hear what he had to say, but was having a bit of trouble catching everything from right up the back. Fingers crossed the audio was recorded successfully. He didn't use any slides so that's all I really need.

I caught all of Elizabeth Garbee's talk on Tuxracer. I was really impressed by her speaking ability. She was really confident, spoken extremely well, and was humorous. The content probably wasn't technical enough for LCA, but it was great to see a young woman presenting, and it was a really enjoyable presentation nevertheless. I'm glad the CFP guys selected it.

That was about the extent of the talks that I made it to.

I'm really looking forward to Dunedin, where I can socialise more and generally be a normal delegate again. In the meantime, I can get back to having a lifestudying.

Oh yeah, shutterbug Michael Davies took a whole heap of photos, which I'm currently hosting for him, and are proving quite popular.

[06:31] [lca] [permalink]

Well, that happened... (or reflections on the conference from an organiser's point of view)

Phew! LCA 2005 is done, and I have to say that I'm personally fairly happy with how things went. There were a few things that we could have done better, but overall, I think it was a pretty rocking conference, which was Steven's main objective.

I figure now is a good time for a braindump, so stand back, here goes...

From an organiser's perspective (in the order they occur to me):

Waaaay too much pizza. The CLUG pizza guestimating algorithm clearly does not scale. We had something like 150 pizzas surplus to our requirements. The final batch of 100 that arrived went straight to Ainslie Village, where they were gratefully received, and about another 50 left over from the preceding 300 where dispersed around the campus of the ANU to random resident students and anyone else who happened to be in the right place at the right time. Oh, and we really didn't do a terribly good job of catering for the people with special dietary requirements. We went to the trouble of asking delegates if they had any when they registered, but didn't plan appropriate alternatives for the Saturday conference-provided pizza lunch, hence me making a rushed trip to the nearest kebab shop for half a dozen felafel kebabs for the vegans and the food^Wdairy intolerants.

Lightning talks fell off the radar. I think Steve thought that I was looking after them, and I certainly didn't think I was. They didn't even make the program, so they really got overlooked. We managed to shoehorn them into the program on the last day with me nominally coordinating them, but it was a bit too disorganised for my liking. I think they are a very important part of the conference, so they need to get factored in. Perhaps half an hour of them a day (first up, prior to the keynote?) would be a good way to do it in future.

Speaking of keynotes, giving away a laptop was certainly a great way to ensure attendance. Rob did a fantastic job of defragmenting the audience every day. I did find the latecomers, who insisted on clustering around the back rather than finding a seat, mildly annoying. The back rows of the theatres were also popular because the wireless coverage was better there. I had mixed opinions on whether people should be availing themselves to the wireless LAN during presentations, but everyone seemed to be doing it, so I guess go with the flow...

The birds of a feather sessions could have been advertised better. This was my responsibility. I had one delegate have a bit of a bitch to me at the Professional Delegates Networking Session about the sessions being too late and poorly advertised. I hope he email[ed|s] the feedback through to us so we get it straight from the horses mouth. I don't really know how we could have done that a lot better, scheduling-wise. I was keen on having 2 hour (maybe 1.5 hour would have been better?) BOFs, and with a pretty jam-packed program, this meant things had to stretch into the evening. The problem with this was that once people shot through for dinner, they didn't tend to come back again, so that realistically really leaves you with 9am until about 6 or 7pm at the latest, before people are going to want to run away and have dinner. I had 12 BOF slots, of which I think 7 I'd filled before the conference started by people emailing us. I wanted to preferably keep half the slots available for people to suggest topics during the conference, but I allocated the vacant slots to the later 2 hours, which I suspect is what the delegate I spoke to at the PDNS was pissed about. In hindsight, perhaps having them later in the week would have been better, however that would have required some serious rejiggery, because most other nights had something on, between the Penguin Dinner, and the PDNS. There was just a lot of stuff to try and cram in, and something had to give. Maybe running more BOFs in conflict with stuff would have worked.

The quiz show was a late addition to the program, and seemed very popular. It was a shame that it was up against the keysigning, with so many well-connected foreigners chosing to attend it over the keysigning.

The venue for the Penguin Dinner was a bit ordinary (mainly with respect to open space and audibility from the back of the room). We were a bit limited with where we could seat 500-odd people, within walking distance of the conference venue. I still think it was a fairly good night, even if I didn't manage to blow $2005 on a signed t-shirt :-) The food was pretty good in my opinion.

I think the conference venue itself rocked extremely hard (damn, that phrase is infectious). Having all the theatres in close proxmity worked well. Having it all in the one building was a definite bonus. The foyer ended up being big enough, even with the couches (and the couches were a brilliant idea).

The (data) networking was really good. I don't think anyone found the static IP addressing requirement humungously onerous. The proxy ARP problem that was bouncing MacOS X and Windows clients off the wireless LAN was a bit of a pain, but the fact that we could piggyback on the ANU's excellent wireless LAN was a real bonus. Bob did a fantastic job of getting a lot out of the ANU's networking guys. I think the terminal room was sufficiently good as well. Throwing a few PCs in there seemed to be well received, as they seemed to be in use most times I poked my head in the room.

I found the organisers' room was too far away from the action. It was good to go and chill out there, but the registration concession booth seemed to become the de facto organiser's room instead. That didn't seem to be a major problem though. I'm not sure how well patronised the speakers' and media rooms were. They appeared vacant the majority of the times I walked past them to go to the organisers' room (which wasn't that often).

Having ready access to a laser printer and laminator was bloody brilliant. I spent so much of the first couple of days just knocking up signage as the requirements popped up.

The slideshow in the theatres worked really well as an information dissemination technique (if I do say so myself). The technology we used to implement it was a little bit flakey (the theatre PCs were netbooted with a minimal Linux installation, and all ran svncviewer back to a central server, which had the desktop shared with rfb. If I'd had a bit more time, and done a bit more testing, I probably wouldn't have gone with something that shared the normal X desktop (or maybe a different VNC server that did), as it did some weird shit with what was exported via VNC if you switched to another virtual terminal. But it worked well enough. I had a lot of trouble finding a GNOME-based slideshow displaying app. I ended up using gqview, which was okay, but not great.

Some delegates seemed to be a bit grotty. Mikal lamented about finding apples cores under all the couches, to which I think Chris or Jeremy responded "those damn Apple users!". That was an amusing comment today. I think bins were in sufficient supply that there shouldn't have been as much mess as there was.

I think we overcatered morning and afternoon teas. I dare say LCA2005 will be forever known as the LCA where the delegates were stuffed with food to the point of popping. The coffee was good, and one of our main concerns (that we wouldn't be able to caffeinate enough people in the time alloted) was unfounded.

The cowbell worked well as an indication that the talks were restarting after the breaks.

Umm, I think my brain is starting to run out of things now...

But please, if you have some feedback, (positive or negative, but preferably constructive if it's negative) please email it to us.

I'm looking forward to attending LCA 2006 as a mere delegate again.

[05:39] [lca] [permalink]

Tuesday, 19 April 2005

And the word for the day is...


[22:01] [lca] [permalink]

Sunday, 17 April 2005

Twas the night before lca2005, and all through Manning Clark, nobody was stirring, not even a conference organiser

I started the day with a couple of airport pickup runs, picking up some Debian developers and dropping them at their accomodation. I also introduced Mako to Vegemite and Tim Tams, which I'm sure he'll be blogging about...

I'm really pleased with how things have gone today. We did our first batch of earlybird registrations (guestimates are about 20% of delegates registered this afternoon). The wireless LAN appears to be working, excluding a gratuitous ARP problem with MacOS X (it's a Linux conference, use Linux, dammit!) which I will Google for a solution shortly.

The slideshow is up and running in the theatres, and I have managed to get svncviewer playing ball with init, so I can remotely PXE boot the theatre machines and have them automatically VNC into the main server with the slideshow running on it (I can drive the whole gig from the couches in the foyer, very cool).

I think the couches in the foyer should be a big hit. They were certainly well patronised this afternoon by the delegates that turned up to register.

Bring it on, I can't wait. (But I'm glad to be at home getting ready for bed at a sane hour, rather than doing a million last-minute things).

[05:00] [lca] [permalink]

Friday, 15 April 2005

Holy bags of schwag, Batman!

Have we got some cool schwag... We've just spent this morning doing the bag brigade thing and packed 500 bags with some very cool schwag if I do say so myself.

I'm impressed that we've managed to have the whole lot done before midday...

Now I just need to organise the printing of some signage, make sure all the lecture theatre slideshow stuff works, and I'll be feeling pretty happy with things.

[18:50] [lca] [permalink]

Thursday, 14 April 2005

Twas the week before linux.conf.au, and all was busy

I've had my own personal hackfest this week, and it's been fun.

First, I tried to get Debian going on an E450 for use as a desktop in the terminal room (as you do). This worked, but I ran into some unpleasant video problems. So I tried taking dilinger's Sargeified Ubuntu xorg packages and building them, which worked, but presented a whole bunch of keymap problems, so I wrote that off as a failure.

Next, I dicked around with the slideshow presenting solution for the theatres when there isn't a presentation on. We've hijacked the PCs that are part of the theatres, and are netbooting them with Linux, and then they're going to VNC back into a central server using svncviewer (so they're pretty minimal), which will run a set of slides (which I must create tomorrow).

The piece de resistance was the access point I helped Bob with today for the backpackers hostel to provide delegates staying there with some wireless Internet access (we're hoping more than one person is actually going to stay there so they can avail themselves of this).

It's a small cased mini-ITX box, with a PCI wireless card and an iBurst modem hooked up to it, running a bit of NoCatSplash (just because I could) (I can't believe this isn't in Debian?). It's doing transparent proxying. It's not that exciting, but I think it's cool because we've essentially made our own bit fat access point with extra functionality to suit our requirements.

Tomorrow I need to make the aforementioned slides up, figure out how to make up a BOF sheet in LATEX that doesn't suck, and do some general gophering. Probably thoroughtly test the backpacker's access point too.

[04:41] [lca] [permalink]

Perfect weather for a conference

I took this week off work to help do finishing touches to linux.conf.au and the weather has been fantastic. I really hope it keeps it up for next week. If it does, the decision to hold the conference in April will have really paid off.

[04:19] [lca] [permalink]

Tuesday, 12 April 2005

In the interests of timely information dissemination

This email should have gone out sooner, but it still going through the works, so to get the information out before delegates start unplugging and getting on planes and getting out of contact, here's a sneak preview:


With the conference less than a week away, we thought we should give you some
orientation information for when you get here, to help you find your way to
the Manning Clark Centre. Once you've registered, you will receive your bag
of schwag, which will include the conference handbook, which will answer any
further questions you may have.

This email contains:

* Directions to the Manning Clark Centre from all over the place
* Emergency contact telephone number
* Important note regarding delegate badges
* A reminder about the keysigning
* Current weather conditions

> Where is the Manning Clark Centre?

The MCC is building 26a, and can be found at 

How do I get to the Manning Clark Center from Burgmann College?

Burgman College is building 52 on

Walk out of the college onto Daley Road and turn left. Take the right fork
onto Sullivan's Creek Road. Keep walking. You should pass the Hancock
Library on your left and there will be a zebra crossing across Sullivan's
Creek road to a bridge over Sullivan's Creek. Walk over this bridge and
follow the path along the rear of the Chifley Library (with Fellows Oval on
your right). Take a left turn past the entrance to the Chifley Library,
keeping the A.D. Hope building to your right and you should see a ramp
leading up to the Manning Clark Center.

How do I get to the Manning Clark Center from the City (a.k.a. Civic)?

Assuming you will be entering the campus via University Avenue, walk down
the pathway keeping the Copland Building (building 24 on
http://campusmap.anu.edu.au/displaymap.asp?grid=ef32) on your right, until
you reach the University Union Building (building 20) and hang a right up
the ramp to the Manning Clark Centre.

How do I get to the Manning Clark Center from University House?

University House is building 1 on

Walk along Liversidge Street until you come to Ellery Crescent. Follow it to
the left, until it reaches a cul de sac outside Melville Hall (on your
left). This will lead you onto the same path beside the Copland Building
referred to in the directions from Civic above.

> Emergency telephone contact

If you get completely, utterly and hopelessly lost, ring 6125 8186 
(that's +61 2 6125 8186) and it will divert to an organiser, who will try
their best to direct you to the Manning Clark Centre.

> Delegate badges, more precious than gold

As you may have already learned, linux.conf.au has sold out quite a few
weeks before the conference. Because of venue restrictions, numbers are
strictly limited to 500. To ensure this, we will be checking delegate badges
throughout the conference. Make sure you wear yours at all times and do not
lose it. We will be charging a replacement fee of $50 to replace delegate
badges (after sighting satisfactory identification). Anyone found not
wearing a badge at the venue will be asked to leave.

Your badge is also your ticket to the conference dinner and the professional
delegates networking session (if you your registration includes these).

> Keysigning

If you are interested in taking part in the GPG keysigning party at LCA
make sure you submit your public key prior to Friday 15th. More
information and instructions on how to submit your key can be found at

> Weather

If you're wondering what to pack, currently it is a bit unseasonably warm,
with maximums in the mid to high 20 degrees (Celcius). There is rain
forecast for Friday, which may drop the temperature back a bit next week, so
you might want to bring a mixture of summer-type clothes (i.e. shorts and
t-shirt type stuff) and slightly warmer clothes (i.e. jeans).

See you soon!

[04:44] [lca] [permalink]

Monday, 11 April 2005

Switching to static output

Fingers crossed I haven't screwed up and Planet Debian is going to see all of my blog all over again.

I've switched Blosxom to do static output, rather than forking off a Perl process for every connection. I can't wait until Blosxom 2 comes out and supports mod_perl.

This afternoon daedalus started to really chug. I'm either getting a lot of (spam) email and MIMEDefang is thrashing the box, or my blog just got real popular (or both). Enabling static output certainly is pretty trivial. Here's hoping the load on daedalus drops off again.


Well fakecat and static output generation don't play well, which means the custom RSS feeds for Planet Debian and Planet Linux don't get updated properly, so I'm switching back to dynamic output until I can resolve that.

[15:39] [tech] [permalink]

Sunday, 10 April 2005

I am Andrew The Blogger

I went to see one of my lecturers this morning, as I have this week as well as next week off for linux.conf.au preparations.

Thanks to Steve forwarding one of my previous blog posts to the entire Department, the first thing he asked me was "Are you Andrew the blogger?".

[20:48] [uni] [permalink]

Confucius say: Man who host website on dynamic IP address gonna have bad time

I've been following Dirk Eddelbuettel's dramas with his broadband IP address changing (but hey, they don't call it a dynamic IP address for nothing), and I thought I would write about how I survive on a dynamic IP address. Until I moved house and had an outage of a few weeks of my broadband, I had changelogs.debian.net hosted on the back of my broadband connection without any major problems using this method.

I've been using ADSL since around 2000, and always been on a dynamic IP address. I had previously enjoyed a static IP address on my dialup connection back in the good old days (gee our web design sucked back then).

I have had daedalus.andrew.net.au since around the same time I went broadband. It has always been sitting in a colo facility somewhere, so is subsequently on a static IP address, and it acts as the primary nameserver for my domains.

For secondary DNS, I use a mixture of a free account I have with UltraDNS (from back in the days when it was Secondary.COM), and the member service provided by SAGE-AU.

So my ADSL IP address changes from time to time. My ADSL connection usually drops out and reconnects every 7 to 10 days, usually coming back with a different IP address. I like to maintain home.andrew.net.au pointing at it, so that I can SSH into it from elsewhere.

So in the way of DNS zone files, I have andrew.net.au, which has everything in it, and I have a subdomain (an actual bonafide subdomain in its own zone file) called dyn.andrew.net.au. (I'll explain why in a moment, it's technically not necessary, but convenient to do so). home.andrew.net.au is a CNAME to caesar.dyn.andrew.net.au, and this is the DNS record that I update whenever my ADSL connection "redials".

My ADSL connection is PPPoE, so I use the Roaring Penguin PPPoE software, which just works with pppd. pppd allows you to run arbitrary scripts when the link comes up, so I've written a Perl script to send a signed (TSIG) DNS update request to my nameserver, and dropped it in /etc/ppp/ip-up.d/

You have to go for a signed request because the update is coming from dynamic address space. I guess I could have added a huge ACL to my BIND configuration allowing all of the address space for my ISP to be able to send my nameserver dynamic updates, but that's a bit broad for my liking. So instead, I have:

zone "dyn.andrew.net.au" {
        type master;
        file "/etc/bind/master/dyn.andrew.net.au.zone";
        allow-update {
                key zoneupdatekey;
        allow-transfer {

I then create a key with

dnssec-keygen -a HMAC-MD5 -b 512 -n ZONE dyn.andrew.net.au
This produces a couple of files like Kdyn.andrew.net.au.+157+43730.key and Kdyn.andrew.net.au.+157+43730.private, which strangely enough seem to contain the same key material, despite one supposedly being private. I add the key material to a key directive in my named.conf:
key zoneupdatekey {
        algorithm hmac-md5;
        secret "iBHthjiMEM3gqPaQy1oME9sTp87awUU65s+z9Rd9s3wxfE1BpTzfM0j/qSGKCxfDECKvVxOLyxQP459JAx5IfA==";
(and no, this is a key I generated for the purposes of this blog entry, not my actual key)

This key is also in my script, as is the domain that is being updating, and that is it. The script can be tested by hand by invoking it:

PPP_IFACE=ppp0 PPP_LOCAL= /etc/ppp/ip-up.d/dyndns
and this should set the IP address to for the A record in question. The script logs to /var/log/messages on the box it is run on, and /var/log/daemon on the box running BIND will contain some information about how the update request was handled.

caesar.dyn.andrew.net.au has a 5 minute TTL, so the most anyone with a self-respecting caching nameserver should retain the old IP address for when it changes is about 5 minutes. That's good enough for me.

Why did I bother with the whole subdomain thing? The BIND 9 Administrator Reference Manual says that you shouldn't edit a dynamic zone file by hand, so you really don't want to mix a zone file that has manually maintained entries with stuff that is updated dynamically. Also, BIND tends to make a bit of a mess of the zone file, so I give it its own zone file to make a mess of, and never touch it myself.

So in summary, home.andrew.net.au is a CNAME to caesar.dyn.andrew.net.au, which is updated dynamically via my script. My script is just a Perl implementation of something you could do with nsupdate and a few lines of shell, by the way.

[20:38] [tech] [permalink]

Friday, 08 April 2005

There are at least four security flaws in this piece of software

eval ls > $HOME/listing

This is on the cover of a brochure for a "Writing Secure Software" tutorial offered by eSec back in 2001. I kept the brochure because it made me think, and until now, I hadn't been able to find four flaws. I was just doing some cleaning up and I found it again.

So far, I have:

  1. relying on $PATH to provide ls (someone can overload it to cause something else to executed).
  2. trusting the output of the aforementioned ls command and executing it
  3. relying on $HOME to be set to something sane
  4. making an assumption about the current working directory of the script (as this is going to influence what ls returns and is thusly fed to eval)

Well, that is four things, but I'm not sure if that was the four things eSec had in mind. Now I think I will throw it out...

[21:17] [tech/security] [permalink]

Thursday, 07 April 2005


That was the sound of our current domestic bliss bubble bursting.

This afternoon, I got a phone call from a sales droid at the real estate agent that manages the place we're currently renting. He said the owner had asked for an appraisal on the property. This can only mean one thing, he is thinking about selling.

Selling means two things: tons of people traipsing through the place, and the probability of having to move again increasing.

I really don't want to move again. We're pretty settled here.

My interpretation of the ACT Residential Tenancy Act is that a lease can be terminated early if the title of a property changes hands, so we might not be safe until our lease runs out, if the property does change hands and the new owner wants to live in it.

So I can only hope that the owner just wants an appraisal for the heck of it (but who does that?) or the appraisal isn't that good, so he decides not to try and sell it. Maybe rising interest rates have something to do with it all...

Time will tell.

[04:10] [life] [permalink]

Tuesday, 05 April 2005


[15:49] [humour] [permalink]

Sunday, 03 April 2005

Pondering my future

I have to say that what Matthew Wilcox had to say regarding the Social Contract changes strikes a chord with me, and echoes the sentiments I've heard of other developers around the traps.

I hadn't been thinking about it too much until someone I respect greatly brought it up recently, and it made me think about it in about as much depth as Matthew probably has, and I didn't like it.

I definitely feel that calling the GR an "editorial change" was grossly misleading.

I guess I will have to consider whether I want to continue in the project once the new Social Contract starts to take effect on Etch.

Maybe Ubuntu is a more moderate alternative.

Like Matthew, I have an emotional attachmen to Debian. I firmly believe it is a technically superior Linux distribution. If Ubuntu builds on that strong foundation without some of the licensing fanaticism, maybe that's where I'm better off contributing my spare time...

[06:39] [debian] [permalink]

I'd like to be a Mr Mom

Sixty Minutes ran a story tonight about men being the "housewife".

I've wanted for a long time (once I had settled down and gotten married, and had kids) to be the one doing the domestic duties, provided it was economically feasible.

For a few years I've been jaded with the IT industry. I can't see myself being in it for the rest of my life. I think the realisation came when SecureNet was acquired by Betrusted. We scored a manager from the United States, and he was having one-on-ones with everyone. He remarked to me about how I'd been there 18 months and was a "veteran" (there had been significant staff turnover in those 18 months). It was like "Wow, you've been here 18 months! There's forever!"

This made me realise that if I was considered a "veteran" after 18 months to 2 years in a job, that I'd be looking at something like another 12 different jobs in my working life. I'd actually like to settle down in a job and be a 20 year veteran, not a 2 year one, and you just can't do that in IT. You either go bust, get bought, or the job gets so sucky that you just plain quit. So far, I've been in all of those situations in my 10 years (egad, it's been 10 years!) in the workforce.

I figure, if Sarah can earn enough money, and my investments can supplement that, why should I have to keep working? I love kids. I change a pretty mean nappy, I'm a clean freak (when I put in the effort). I could do Open Source stuff from home in my spare time (some mothers may laugh at me now, saying that it is a full-time job just being a parent).

I think most importantly, I'd like to be there for my kids. I've worked some pretty horrendous hours in jobs in the past, and I don't want to be working jobs like that when I have kids at home I could be spending time with.

I certainly don't agree with 60 Minutes calling the phenomenon a "social upheaval". Gawd, just because people aren't sticking to their gender stereotypes any more? Give me a break. I think it's a lifestyle and an economic decision. I know Sarah really enjoyed last year when I was a full-time student and not working at all. She had lunches made for her, all the housework got done, and I was a much nicer person to have around.

[05:06] [life] [permalink]

One assignment down

Whilst I don't necessarily agree with the motive for this particular assignment being one done in pairs, I have had a pretty fun time doing the actual work. It was nice to be able to openly collaborate on an assignment for a change.

I worked quite will with my partner, Tiane. We both have half a clue about Java, and we both seemed to be on a similiar wavelength. We were able to elaborate on our ideas really badly to each other, yet understand what we were trying to say and do, which was rather convenient.

In terms of efficiency, it probably wasn't all that good. I think we spent 3 sessions in the labs at uni for about 7 hours at a time, but the time seemed fairly productive. Doing the work in three sessions seemed to help us take a fresh approach to problems. We would solve something in 15 minutes that we'd previously been bashing our heads against in the previous session.

So I'm kind of hoping the next assignment will be a partner one as well, because it's more fun.

[04:48] [uni] [permalink]

Is Architecture: Any too much? (or If only one could control wanna-build more easily)

As I continue to wade through the backlog of mail in my debian-devel folder due to The Thread, I read a small thread starting with this message from Peter De Schrijver, which got me thinking.

Why do we bother building everything on every architecture?

I haven't seen or played with any of the more uncommon architectures that are potentially going to be affected by the Vancouver Proposal, but I know ARM is more of an embedded system architecture, and I don't actually know what m68k is good for, other than maybe revelling in a bygone era.

For some of these architectures that have trouble keeping up with building everything, do we actually need to build everything on them? Does s390 need KDE? Does it even need X? Can you sit in front of a mainframe and use it as a desktop computer? Would you even want to? Similar questions must apply for some of the other architectures.

So I wonder if we're a bit too quick to use Architecture: any in our binary packages? Well, as I just discovered on #debian-devel, the Architecture: field is irrelevant anyway (to package building), it is wanna-build that attempts to build everything, and it doesn't look at the Architecture: field of a source package.

So it's a bit of a shame package maintainers can't more directly control this from their packages, and it's a shame that it can't be done on a opt-out basis. For example, if KDE didn't need to be built on m68k, the kde source package could have something like Architecture: any, !m68k. Even if the architecture list had to be explicitly specified, wishlist bugs could be used to add an architecture to the list that a particular package was built on.

But, as it seems that the build process doesn't work this way, this approach won't work in any shape or form without a bit of reengineering of all the buildd stuff.

One upside of these musings is that I discovered a veritable wealth of knowledge about how everything hangs together, which was something I was lamenting about the lack of about 18 months ago, so I can improve my education some more.

Back to reading debian-devel...

[04:40] [debian] [permalink]