Colin's Journal: A place for thoughts about politics, software, and daily life.
Apparently EU voters are not “mature” enough for an EU wide referendum to be held. The evidence for this: that given an undisclosed list of people to vote for, a sample of French voters picked the only Frenchman on the list, the rather famous Valery Giscard d’Estaing.
As Europunditry points out, in what way does this research show that an EU wide vote on adopting the constitution would be invalid? Even if we accept the hypothesis that people will generally vote for someone from their own country, how does this invalidate an EU wide vote for adopting or rejecting a constitution?
If an EU president was elected by vote (not something that’s being considered at the moment) this research might be of some interest. The major point that is raised for me however is not that the European public is not mature enough to vote in an election of a president, rather that the voting method used is of critical importance. If a single transferable vote method was used, or multiple rounds of voting (probably impractical), you could very quickly get the electorate interested in candidates that aren’t from their own country.
I’ve released version 0.4 of TALAggregator, my RSS aggregation software. This release fixes a problem where some feeds would not be successfully parsed because of the interaction between MySQL and Python when handling Unicode strings.
The source of the bug is that the MySQL interface in Python can do automatic conversion to and from Unicode for VARCHAR columns. Unfortunately it doesn’t perform this magic for LONGTEXT columns, leading to all sorts of confusion on my part. It’s should be solved now – I’ve disabled the automatic conversion and do it all myself.
Finding near by paths suitable for in-line skating is difficult in Toronto. If you are a good in-line skater who doesn’t mind the odd hill, gravel road, or tram tracks there’s plenty of choice. With my skill level however I’ve found it best on the beaches and out on the islands, both of which take a considerable time to get there.
I decided to try my luck at Etienne Brule Park this afternoon, and it seems like a good place. In theory you can follow the trail all the way up to Lawrence, but I found that the path deteriorated in quality by Dundas West. Maybe once I’ve got back into the swing of things I’ll try going further up.
From the Old Mill subway station it’s just a short walk to the park where the path starts. There’s a separate brick path for pedestrians, and a paved path for cyclists and bladers both of which generally follow the path of the river upstream. The surface quality is mixed but generally good, with the inclines being shallow enough for me to cope with. The path does climb slowly as you work up river, which means that coming back is much more fun. It’s a shame that it’s not a little longer, but I may yet give the path into Lambton Woods another go to try and extend the run.
It seems that BitTorrent is finally getting some mainstream press, inevitably because of it’s use to download material that’s under copyright. The BBC has an article describing how that latest episodes of US TV shows are being made available for download.
The two take away points from this article seem to be that the “problem” is at this point rather small (not many people doing it), and also hard to disrupt. I’m sure that TV companies will try to shut down sites assisting in the distribution of TV shows in this fashion despite the difficulties of doing so. Other actions that they are less likely to take, but which might also have an impact:
If downloading TV shows becomes really popular it could even become just another channel of distribution. If it’s easy to find an official version (with adverts) most people are not going to bother hunting down a hard to find version which has had the adverts stripped out. Bandwidth costs can be contained through the use of the peer-to-peer network, although multiple versions would have to be provided so that the advertising is targeted to the correct audience.
I suspect the biggest challenge would be that people can skip through the adverts, which reduces the amount of income derived from them. If they are kept short enough however it might work, it takes time to skip ahead and then backtrack to the point where the advert finishes and the show resumes…
There’s a fairly detailed rant/analysis of the new on-line rail timetable in The Register and all it’s failings. Some of the things mentioned in the article have already been corrected (for example the ‘Today’ and ‘Tomorrow’ buttons are back), however the best part of the article is the link to the German alternative. Yes you can search for UK train times using the German site, and it’s better than the new UK version!
Not only can you search for train times for journeys in the UK, but it seems to handle most European countries. For example you can work out that you can take the “train” from Helsinki (yep Finland) to Preston in a mere 44 hours 52 mins!
For the curious: You take a two hour train from Helsinki to Turku, the overnight ferry to Stockholm, then back on the train to Hamburg with a change in Koebenhavn. Another train to Bruxelles (Brussels), the Eurostar to Waterloo, underground to Euston and then a direct train to Preston.
Perhaps more realistically you can do Preston to Venice in just shy of 21 hours, with only 3 changes (excluding the underground)! If you fancy having a play yourself check the timetable site out.
I’ve just had my first skate of the year, and my feet hurt. I managed to spend half an hour (I know, pathetic isn’t it?) outside in the bright sunshine skating around the neighbourhood without falling once.
If the weather is this good next weekend I think I’ll head down to the beaches, the paths there should be in far better condition than the roads around our house are, and the scenery is far more interesting.
I’ve uploaded a few pictures from Kevin and Mary’s wedding. If you would like to see them (see me in a suit!) drop me an email and I’ll send you the login details.
I think I’ve tracked down the reason for this page not being displayed correctly in IE. I was setting the width of the left hand column explicitly, which worked fine in Mozilla but caused IE to display visual glitches. Having removed this it now seems to display OK in both browsers (you may have to control-reload the page to see the fix).
Mozilla’s “DOM Inspector” was very useful in helping to find this – highly recommended for web development.
On the way in to work this morning I spotted the Metro headline, that two more people have died of SARS in Toronto, so I looked up what the Globe & Mail were saying on the subject. The article is interesting because I learnt a couple of things that I’ve not seen mentioned elsewhere:
Although SARS has been linked to a specific coronavirus, that virus is being detected in fewer and fewer cases. And, more puzzling, it has been found in some people who show no symptoms, suggesting that it is possible to contract SARS without ever knowing it.
Francis Plummer, the scientific director at the National Microbiology Laboratory in Winnipeg, told the conference at a Toronto hotel that only 40 per cent of people who have been diagnosed as probable SARS cases have tested positive for the virus. That number has declined in recent days to almost zero – something the scientists are at a loss to explain.
So it appears that cause of SARS is still a bit of a mystery. It makes sense that in some cases the virus would show up without any of the symptoms, but why would people with the symptoms of SARS not have the virus present if it does indeed cause the disease?
I happened upon “Real Programmers Don’t Use Pascal” today. It could have been re-written several times since 1982 with a whole series of languages and environments given the same treatment as that handed out to Pascal here. My favourite part:
Real Programmers are reluctant to actually edit a program that is close to working. They find it much easier to just patch the binary object code directly, using a wonderful program called SUPERZAP (or its equivalent on non-IBM machines). This works so well that many working programs on IBM systems bear no relation to the original FORTRAN code. In many cases, the original source code is no longer available. When it comes time to fix a program like this, no manager would even think of sending anything less than a Real Programmer to do the job — no Quiche Eating structured programmer would even know where to start. This is called “job security”.
The full list of my published Software
Email: colin at owlfish.com