Stranded
I’ve very often gone via Stansted on my way to Denmark, and I’ve always made the connexion, although Ryanair don’t guarantee it.
Today seemed OK – we were supposed to land around 6pm, and the check-in didn’t close till 19:05, so it didn’t seem likely it would be delayed more than 30 minutes.
But it was!
We arrived here at 19:15, so now we’re trying to find another way to Denmark.
Unfortunately, all flights are full, so the next affordable plane to Denmark is on Monday!
We might have to go via Bremen!!!
Update (4am): Yes, Bremen turned out to be the best bet. Ryanair were willing to book us onto that for a “missed flight fee” of £250 for all of us. The flight is at 6:40am, and we’ll arrive there around 9. Hopefully it won’t be too expensive to get the train to Denmark.
Because the flight is so early, we didn’t think it made sense to get a hotel room, so we’ve all been sleeping on the benches or the stone floor of Stansted Airport. I guess it’s all an experience for the kids! 
Heathcliff
I forgot to blog this at the time, but what was Brown thinking about when he compared himself to Heathcliff from Wuthering Heights?
As this brilliant blog posting says, many people haven’t read the book but instead know it from “the 1939 film adaptation, with Laurence Olivier playing Heathcliff in best matinee idol style”.
So has Brown ever read the book, or was he comparing himself to Laurence Olivier?
Or does he really see himself as the following? “A half-savage Gypsy boy who skulks around the Yorkshire moors in the freezing cold, sleeps in stables and drives the woman who loves him to an early grave. He is mostly a hair-trigger away from violence and can be guaranteed to lash out with his fists if anyone so much as gives him a funny look. Oh, and he also speaks in such an impenetrable dialect that it’s all but impossible to know what he’s going on about.”
Succesful move
As far as I can tell, the move to this new site has been successful.
Perhaps I’m being optimistic, but at least my traffic statistics package tells me that I’ve had 83 views since last nice, which is roughly three times as many as usual for that amount of time.
And how did people find it? For instance by searching for “iskaffe”, “all junk for sale”, “3 dages feber”, “normal weight”, “warehouse humor”, “udslæt”, “belly button men” and “spider man wegge”.
It’s amazing how people come to my blog to learn about things that I didn’t even know existed!
Move completed!
I’ve finally completed moving this blog from twid.bibulus.org to the new address, blog.widmann.org.uk.
I didn’t just change the address, but I also moved it to a different hosting company, Prime Hosting, and I switched from using Blogger to running my own WordPress installation.
I’m sure quite a few things aren’t working perfectly yet – in particular, I’m a bit anxious as to how the redirect from the old site will work – but I hope most readers will appreciate the new site.
For a start, I think the new site looks better, and there are nice new features on the right, such as a label cloud and a nice calendar. Also, the search works much better.
Please let me know whether you’re missing any features from the old site, and any other ideas for improvement!
Under renovation
I’m planning to move this blog to a different location, using different software and doing a lot of fancy things in the process.
Unfortunately, I think I’ll need to freeze this blog in the meantime, which will mean no new blog postings and no access to the comments.
It shouldn’t take too long, but of course there might be some unforeseen problems.
I haven’t yet decided when to start the conversion, but if the blog suddenly seems eerily quiet, that’s probably what’s up.
I Danmark
Så er det blevet tid til den årlige tur til Danmark for hele familien.
Anna, Phyllis og mig var naturligvis i Danmark i begyndelsen af april, men Léon, Charlotte og Marcel har ikke været der i nu et år.
Vi er der fra fredag 1/8 til lørdag 16/8 (begge dage inkl.).
Vi regner med være det meste af tiden hos mine forældre i Sdr. Vissing.
Sagen er den, at hvor man kan leje en bil med fem sæder til en rimelig pris, koster en med syv sæder det dobbelte, så vi vil prøve, om vi kan klare os med offentlig transport og nogle taxature.
Men der er jo også god plads i præstegården, hvis nogen skulle få lyst til at komme forbi.
Font design
Normally when you design a font these days, you use a program like FontForge, which is basically a glorified drawing program.
You can do anything you want, but it’s basically up to you to make the individual characters looks similar so that they combine to form a unified and beautiful font.
If you use a program like METAFONT instead, you have more options. You can define a “pen”, for instance a rotated oval, that you can draw shapes with, and you can define functions, for instance to produce serifs in a given location. However, you have to sacrifice drawing the basic shape with your mouse and have to use mathematical expressions instead.
So I’ve been wondering for some years whether it would be possible/feasible to write a font design program that lets you define a sophisticated “pen” that not only has a shape, but also comes with serifs to leave at the start and end points of a line. The serifs and the shape could depend on the direction of the stroke.
So with such a pen defined, you could then draw the basic shape in your drawing program and all the sophisticated bits would appear automagically.
In other words, I suggest drawing the top character on the right manually in the drawing program, but using fancy “pens” to turn that basic shape into all the variants beneath.
The resulting font could then be loaded into FontForge for fine-tuning, but it should be much faster than drawing an entire font by hand.
Has something like this ever been done yet?
Sudoku website
I’ve just bought the domain widmann.org.uk, and it’s my plan to move this blog there eventually.
However, I’m not going to do this anytime soon, but I’ve already started playing around with it.
The hosting company claimed it was OK to put up Perl CGI scripts, so I thought that would be fun to test.
So I took a small Sudoku generator that I wrote a couple of years ago and added an HTML interface to it.
The result is here.
I probably should disable the 16×16 option – it’s really slow, and the servers might have better things to do than doing this kind of number-crunching.
Apart from that, I guess I should optionally show the solution to each puzzle. Some other day…
Phonetic destiny
In Danish and its ancestral languages, medial (and final) stops have been undergoing reduction for more than a millennium: Old Norse often has voiced fricatives corresponding to stops in other Germanic languages (rauð corresponding to English red, German rot), and Danish then created more (ON hvít > hvid /viðʔ/). And now the same thing seems to be happening in colloquial Danish: bakke /ˈbɑɡə/ can be pronounced as [ˈbɑɣ(ə)] in rapid speech.
Likewise, German seems to be obsessed with hardening initial consonants. First the Germanic one, then the High German one, and now in many varieties, /ptk/ are aspirated and thus ready for another shift.
And Slavonic seems to have undergone several palatalisations, one after another.
It can thus feel like languages are destined to move in a specific directions, thus the title for the posting.
However, destiny is not a very scientific concept, so what is really going on?
One possibility is that it’s random. This cannot be dismissed out of hand, and further study is needed to determine whether phonetic destiny is statistically significant.
If it isn’t random, my best guess is that it’s caused by ways of speaking. Some languages are on average spoken faster than others, some are spoken more sloppily or with less jaw movement, and then there’s the contrast of syllable-timed vs. stress-timed languages.
My guess is that certain ways of speaking leads to certain sound changes being likely while effectively ruling out other sound changes.
For this to be the cause of phonetic destiny, it would require such ways of speaking to be much more stable than specific pronunciations, lasting centuries if not millennia.
This long time-frame makes this very hard to investigate scientifically. I guess one could try to find old perceptions of other languages. For instance, have any Norwegian or Swedish medieval sources commented on how the Danes pronounced their language?
One could also try to find unrelated languages that are known to have gone through similar sound changes recently and investigate whether their pronunciations are similar in other respects, e.g., whether they are spoken equally fast.
Does anybody know whether anybody has already investigated this topic?
Regional polls
I’ve been a fan of Electoral Calculus for some time, especially of the pages that allows you to predict Westminster elections, either for the UK or for Scotland only.
However, it’s often annoyed me that you had to do either, but never both at the same time. Because the Conservative revival hasn’t happened north of the border, where we’ve instead seen an SNP surge instead, it’s been a problem for accurately predicting the results of an election.
So I wrote to Martin Baxter some weeks ago, and now he has implemented my suggestion!
It’s only a beta version for now, but it seems to work really well. You can put in national values and it will then fill in regional values that you can then override where necessary. Do try it!
I tried it out using two recent opinion polls. Counting seats, instead of Cons 450, Lab 148 and LD 24, I get Cons 450, Lab 138, LD 21 and Nat 21.