Follow the reluctant adventures in the life of a Welsh astrophysicist sent around the world for some reason, wherein I photograph potatoes and destroy galaxies in the name of science. And don't forget about my website,

Monday, 18 November 2019

Government Reloaded

I know I haven't been active here much lately, although things on the other blogs are in a bit better shape (though I do have a series of philosophy posts in draft and almost ready to go). I've been seriously stressed out about the whole Brexit thing and the possibility of becoming an illegal immigrant, so I've been throwing myself into writing code instead of blogs, which is a more effective distraction.

Who should I vote for ?

Since the three month extension all but guarantees my secure residency status, it's time to look at how the heck I should vote in the oh-so-exciting forthcoming election. Back in 2017, I felt dissatisfied with the traditional web-based questionnaires that assess who you should vote for based on stated party policies. So I wrote a little spreadsheet that takes into account other factors, like whether I trust the party to implement policy and/or adjust their priorities based on changing evidence. For example, I like some of the current Tory party policies (who couldn't like more public investment after years of austerity ?), but I wouldn't trust them as far as I could spit on them.

Here's the result from last time :
The "policies" score is based on the results from isidewith, which is pretty objective. All the other parameters are my own subjective judgements.

Two years later, have things changed ? Well, yeah, a bit. Not all that much though. For some reason isidewith doesn't give the results for Plaid Cymru any more, so I used the results of last time. Nothing was given for the Brexit Party either*, but they're not worth bothering with anyway. Also I included a weighting factor for each property in terms of how important it is to me (this turns out not to make much difference, at least when the differences are this stark), and give a percentage of the maximum possible score as a slightly easier way to compare the final results.

*I gave UKIP slightly different scores this time not because I felt sorry for them or anything, but just because I whimsically decided that a value of 1 is good enough to denote "utter crap" and it wasn't necessary to use smaller values. The main point is the final relative ranking anyway, not the absolute numbers.

You can see that this is a much better way to distinguish between parties when their policies aren't all that different : my scores from isidewith are basically equal for the left and centrist parties, with a huge polarisation against the right. That the Tories and UKIP respectively reach 45% and 30% in terms of policies demonstrates that choosing by policy alone is a lousy idea - sure, I agree with them on some basic issues like being in favour of the judicial system and not torturing children, but that's not a sensible reflection on who I'm gonna vote for. For all the problems of polarisation, when it comes to making a decision, it's the differences that matter, not the similarities. I mean, I agree with Nigel Farage that oxygen is generally a good thing, but that doesn't make me any more likely to vote for him.

Anyway, the Lib Dems are still clear winners, but everyone scores lower than last time - nobody scores above a few percent. Most parties have one or two areas in which I agree with them to at least a reasonable degree, but none of them perform well in everything. Overall, given that things are so dominated by Brexit, I'm glad the opposition have rallied to try and stall things. I'm rather less happy that they can't agree on a damn thing besides delay : they missed their golden opportunity to chuck out Boris on a vote of No Confidence; all their talk of refusing pacts and alliances annoys me intensely. The very inability of anyone to compromise is precisely what's led the current paralysis*. Hence no party has come out of this unscathed.

* There is no contradiction between wanting more compromise and wanting Boris removed. Compromise is only possible between the reasonable and reasonably consistent, not with fickle nutcases.

(For those wondering, though I'm fiercely opposed to Brexit in any form, I have set out the sort of compromises I could accept here, and I'd probably be willing to go further than that, if pushed).

I will never for the life of me understand why Parliament didn't do the feckin' obvious thing of going back to the people. Fair enough that Theresa May decided to act like a traditional majority leader at first, but after losing a vote by the largest amount in history, it should have been obvious that a radically different strategy was necessary. At a minimum, the Brexit negotiations should have been cross-party, because this is an issue that effects everyone. More realistically, a second referendum would have been a decisive way to break the deadlock.

While I've never seen a second referendum as anti-democratic, I've always accepted that those saying this is potentially voting again until you get the right answer have an entirely valid point. This was because I lacked the general criteria as to how to judge when a second vote is legitimate and when it's not. Very helpfully, Bercow provided the answer of substance and circumstance. Repeatedly asking people an identical question in identical circumstances is not democratic, hence he chose not to allow the government to bring back a bill within 24 hours of it being shot down. That these criteria have actually been used to prevent repeated votes makes them especially appealing - no-one can say that you can apply them to get repeated votes whenever you feel like it, because you can't.

A second referendum would be clearly different in both circumstance and substance. Circumstance we need hardly bother with, that is obvious. But in substance too the difference would be clear. The original question simply asked us if we'd like to leave or not. The new question would be a choice of staying in or going out under very precise conditions. That's like asking someone, "wanna go out tonight ?" as opposed to, "wanna go for a cocaine-fuelled orgy where everyone dresses up like farmyard animals, or stay in ?". Those questions are fundamentally, undeniably different. Yes, they admittedly share one common option, but the other options are not the same at all.

But I digress. None of this helps me decide who to vote for. The Liberal Democrats get extra bonus points here because of the candidate they're fielding in the constituency I vote in :

The problem is that Cardiff North is traditionally a Labour-Tory marginal, so my vote is potentially very influential. So my conscience says "VOTE LIB DEM !", adding, in a loud shouty voice, "HIS NAME'S LIKE MY NAME !". But my brain says, "err, well, are you sure... ?" rather more quietly while biting its lip and looking worried. I'd rather have the Lib Dems over Labour, but in Cardiff North that doesn't look like a realistic choice (my ideal choice is, as it has been for some years, a Labour-LibDem coalition). Am I handing victory to the Tories by voting for a third party in this case-? I don't know - I'm going to have to wait and see how thing develop. First time for me that a campaign could actually change how I'll vote.

Had this been another case of May-vs-Corbyn, I would have chosen the Lib Dems without hesitation. May and Corbyn were and are about equal but different varieties of awful, so choosing between the two made no sense - I might as well vote according to my principles, which I did. But now it's Boris versus Corbyn, and that's rather different  Is Corbyn awful ? Oh my yes, but Boris, to me, is unquestionably worse. If Corbyn is like eating a shit sandwich, whereas May is like eating a shit hot dog, then Boris Johnson is more like being raped by a bear. It's a whole other level and kind of awful. Corbyn's weakness is that he never changes his mind; Boris' that he doesn't have one. With Corbyn, policy-wise one knows what one is getting - with Boris it's equivalent to putting one's hand in a lucky dip that contains some boiled sweets, and old sock, and a nest of scorpions.

Well, yay.

What fresh hell comes next ?

As Plato says, we shouldn't just examine issues in isolation but also consider consequences. There are all kinds of possible outcomes to whatever happens next - I don't want to guestimate probabilities, but just to set out possibilities for a roadmap of what to do in each eventuality.
  • A Tory majority. Even if this is only small, the only likely outcome is a ratification of Boris' deal. It might, perhaps, be amended, but ultimately it'll go through. That still carries with it the effective risk of No Deal at the end of 2020 if they can't agree on trade policies and refuse to extend the implementation period.
  • A Labour majority. This potentially results in a softer Brexit and a referendum on the result, assuming the EU grants the necessary extension and is willing to renegotiate. The former seems unlikely to be rejected, since it would by definition be the absolutely final extension possible. The latter, despite recent protestations, also seems credible given that Corbyn's earlier proposals were already endorsed by the EU. Getting us to stay in the Custom's Union cannot really be a complicated issue, given that May's deal allowed for this eventuality in the backstop.
  • A Liberal Democrat majority. Brexit is revoked, though this is incredibly unlikely.
These are all obvious. More complex possibilities open up with a hung Parliament. Here things get very tricky indeed.
  • The balance of power favouring the Tories, but not their Brexit policy. Say, in which the Tories plus right-leaning but Brexit-opposed independents just outnumber Labour, Lib Dems etc. This would essentially continue the current paralysis. The Tories might form a workable minority government, but BoJo's sheer cack-handeness would render it useless because the man refuses to compromise on a damn thing. This continues the current state of flux, with nothing happening until the very last days of January - probably ramming through BoJo's deal, but perhaps compelling a vote on extending the transition period beyond 2020. But after January the government would likely collapse and yet another election would be needed. 
  • The balance of power favouring Labour, but not their Brexit policy. This might be even worse. If Labour formed a minority government but could not guarantee that they'd get Parliament to agree to either their own deal or a second referendum, it's hard to assess what the EU would do. It's possible that they'd refuse an extension even with a Labour majority, but a second public vote is a very much a final, nuclear option on the whole thing, so that's unlikely. But if Labour can't actually guarantee that would happen... things get messy.
No numerically viable coalition looks workable in practise at this point. If politicians have any sense, they should start asking the EU right now what they'd do in these scenarios. Asking at the last minute, presuming that an extension will be granted, is bloody stupid. The French strategy of continuously appearing threatening and then acquiescing at the last moment every time is a good one, and perhaps that's their whole game plan (but equally they might be really getting sick of the whole sorry thing). But there's no guarantee that applying pressure to a totally deadlocked Parliament would actually help - it could well result in Parliament simply doing nothing, and crashing out without a deal through default. Even if the EU tried to save us from ourselves by strongly encouraging a second referendum, there's no guarantee that a poorly-hung Parliament would actually do it.

To my mind, in the event of a hung Parliament a second referendum offers the only way forward. But then, it's obvious to me that Boris Johnson is a dangerous buffoon, whereas apparently this - astonishingly - isn't the case to almost 40% of the public, something I cannot for the life of me understand. So what the hell would it take to force Parliament to make a bloody choice ?

The biggest problem with our electoral system is that it's incredibly limited. When I vote for a candidate, I don't specify if I'm voting for them or their party. I may or may not be choosing to put that party in power : I may want them to be in a coalition government; I may only want that specific individual in charge of my constituency; I may not even care about the candidate and just want to vote nationally. There's no way of knowing. There's no perfect solution to this and I have no revolutionary idea. I just want people to start thinking about ways to vote for things we actually want, rather than voting against thing we don't. But I expect such a vision is just too darn Utopian. Ah well.

Sunday, 10 November 2019

Cologne Cathedral Calculations

Every year there's an "all hands" meeting for anyone who's even so much as accidentally glanced at ALMA data management in Europe. This year's was in Koenigswinter, near Bonn. Having fretted like you wouldn't believe about the possibility of a No Deal Brexit, the extension meant I could go to this without worrying about having to get back into the Czech Republic afterwards.

Instead the issues were all about getting there, not coming back. I signed up at the very last minute, being convinced that I'd done this months ago. Then it turned out the hotel was full, so I spent a frantic couple of hours searching every other hotel in the vicinity only to discover they were all bloody full as well. How that happens I'll never understand - apparently the demand to see miles and miles of generic rolling forested hillside is extraordinarily high. Fortunately there was a last-minute cancellation so a room became available after all. Hooray !

Getting there wasn't much fun, though, since I had an absolutely stinking cold and it was a seven hour drive. I mostly sat in the back feeling utterly wretched and sniffling. At least there was a nice sunset to look at.

We missed the opening buffet dinner but I didn't really mind since I very soon collapsed. This was quite rejuvenating and I woke up no longer feeling like I wanted to make zombie noises every time I shuffled along, or had any particular inclination to vomit at people.

It must be said that ALMA meetings are distinctly corporate in flavour compared to other astronomy conferences. In the past, they've nonetheless been quite lively as people have rather strong opinions about how to do things. But now things have reached a steady state where everything is ticking over nicely and no-one is likely to spit in anyone's eye or start a protracted blood feud over how many proposals got accepted. They've found a formula, and it just works.

There are only two things of note about the meeting. The first is this spectacular Lego ALMA model :

The pedantic reader will immediately note that there are neither the correct number of antennas nor do the models have realistic designs. Well, sunshine, you're gonna wish you hadn't noted that, because the model is interactive and stupendously cool. Each antenna can be placed on specific pads, where some electronic equipment registers its presence and alters a virtual array in real time. What this does is simulated how real data would be processed by an interferometer of that exact configuration, and you see the result on the screen above the model. Interferometry - and I can't stress this enough - is hideously complex, and if I had my way I'd make it illegal, but this model does the damn thing in real time using bits of Lego. That's like building a microwave out of a yogurt pot and a dead bat. Yes, alright, and a PC running in the background, but you get the idea.

Anyway, this is a very nice way to illustrate how the layout of the antennas really matters; if all your antennas are far apart from each other, you'll see fine details but no large-scale structure; if you only have antennas close together you won't see any details at all. Only by having a mixture of different antenna spacings can you see something close to reality. The model lets you choose which image you want to view (a galaxy, a simple preset model, or a live feed from a camera) and how long to simulate the integration time. Obviously, I don't need to wax lyrical about how tremendously impressive this is. Suffice to say that every home should have one.

The other thing of note about the meeting is that some poor soul decided to arrange karaoke, from which I excused myself from on the grounds of not wanting to collapse on stage and/or frighten the neighbourhood cats by singing with a sore throat. This event was conducted with far, far too much sobriety, but the landslide winners (the UK group) managed - and I kid thee not - to start a mosh pit with "Anarchy in the UK" ("Anarchy for the U.K. / It's coming sometime - JANUARY 31st !").

Yes, really, that happened.

The social excursion was probably my favourite of all the ALMA meetings so far : a trip to the massive Cologne cathedral. And it really is massive. I tried to resort to panorama mode to get it all in, but it didn't work. Instead I only succeeded in chopping someone's leg's off.

There was also time to wander round a bit before exploring the cathedral, so I went across the bridge with is emblazoned with hundreds of A4 paper sings instructing readers to "Fuck Theirry Jaspart". Who's Theirry Jaspart ? I looked him up, and it turns out he's a rubbish artist trying to attract attention. And I'm not going to back down for calling him rubbish because a) he is and b) he clearly spent ages on a lame publicity stunt. I'm not going to even give him a link. Fuck you, Theirry Jaspart. Stop sticking signs on things !

Back to the cathedral. As I said, it's very, very large.

It also claims to house the mortal remains of the Three Wise Men in a shiny golden reliquary.

Entirely plausible, I'm sure. More interesting was the tour of the roof. 45 metres is really, really high, and that's still a full hundred metres or more below the peak.

They started building the cathedral in 1248 but soon got bored when they realised they hadn't a snowball's chance in hell of ever finishing the thing, probably because there were only ever about 60-80 people working on it. Only the lower levels, for the most part, are actually medieval, since the builders decided to take a quick 300 year break while they worked out what to do next. Then better construction techniques came along and they got on with it, finishing it off in the 19th and 20th centuries though sticking to the original medieval plans (which fortunately they still had). This means the interior of the roof looks extremely unusual.

Personally I thought this was great - modern materials and techniques but with medieval styling. One particular tower - which marks the exact centre of Cologne - is particularly weird, looking normal from a distance but "like the Chrysler building" (as our guide put it) close-up.

All this finally ended in a very nice sunset over Cologne, and then everything went back to normal again.

Thursday, 29 August 2019

Escape To The Old Country

Another travel post !

Most of my holidays consist of week-long jaunts back home, often featuring a non-zero fraction of time back in Cardiff's astronomy department as this is where a bunch of cool people hang out. This time was a bit different : one week in Cardiff sans physics and one week in north Wales. Cardiff is a lovely place to live (anyone who says differently is just trying to keep it to themselves) but it's not exactly the tourist highlight of the country. Still, it's nice enough. We went down the bay and saw the Lego models :

And then we went on a boat ride, which features... well, being on a boat for a while. Called the Enterprise, because Drama, I guess.

Some people say that Cardiff isn't a very Welsh city, but they're wrong. I can prove it. One morning we woke up to this :

Wet and full of sheep. What more do you want ?

But Cardiff is somewhere you normally live rather than visit. Up north, it's the reverse. It's also the exact topographical reverse of the Netherlands, making it the perfect place to teach Shirley the wonders of mountains.

(Shirley protests that she already knows about mountains, and has been to the Alps, but this strikes me as completely inconsequential.)

This turned out to be even easier than expected, as the cottage we stayed in (which happened to be extremely cosy and absolutely adorable) was in fact slap-bang in the middle of the mountains already. You could drive for many miles in one direction before seeing anything more substantial than a farmhouse, and in places not even that. It didn't merely lack a phone or WiFi signal - it barely even got a radio signal.

No internet is of course better than slow internet, which sucks donkeys. I wouldn't really want to go without for much longer, but not having any internet access for a week is just lovely. Still, it's not really sensible to sit up in the mountains reading books the whole livelong day, so our first stop was the picturesque Lake Vyrnwy with its famous fairytale castle water pump station :

Last summer everything in Europe was dead and it was so hot that birds were spontaneously combusting in the air. This year it was 15 C, everything was green, and there was so much water in the lake that the dam was overflowing.

We also walked up to a natural waterfall, which despite having been to the lake before a bajillion times back in the day, we'd somehow never noticed.

Mountains ? Check. Waterfalls ? Check. Obviously the next thing you do in Wales is visit a castle, so we did. Chirk Castle, as it happens, where I'd never been before (Wales claims the title of world's highest castle density). Although this is one of Edward I's "Iron Ring" of fortresses to suppress the unruly Welsh, it has a very different appearance to most of the others. Like Beaumaris, it was never fully completed, and it looks oddly low and squat. It's also set on a very gentle hill that isn't enough to deter a disabled postman, let alone an invading army.

But also unlike all of Edward's other fortresses, Chirk is still inhabited. This means the rooms are in quite a bit better condition than most of the ruined castles.

Chirk is very nice and well worth visiting. For a more medieval feeling, we went to Harlech. Harlech is what I think of as one of Edward's real showpiece castles, along with Conwy, Caernarvon and Beaumaris. It's also famous as the military stronghold of Owain Glyndwr's rebellion, and like the castles of the Welsh princes, it's in a very scenic location (whereas Edward tended to build in the flattest places available, presumably because he hated Wales so much that he wanted control of every scrap of flat land he could get his thieving hands on). It's an altogether more castley place than Chirk, and a lot cheaper too.

We were hoping to finish off the trip with the other Welsh necessity of visiting a mine. But oddly, we were situated in one of the few parts of the country where mines are inaccessible (Wales may be tiny, but it's also very very hilly, which means you can't just nip around however you like). So there was no time for the mine. Instead, we went to a second, bigger waterfall - that took us about 16 years to find because the main road was closed. It also has an obnoxious car park charge that can be easily avoided by parking down the road instead.

We finished off with Powys Castle, another one I've never seen before. Like Chrik, Powys is more of a stately home, albeit quite an interesting one. There's a Clive of India museum and a pretty nice free talk about the history of the castle (it does have medieval roots, even if the interior is now just a fancy house).

Powys is well worth visiting too. Parts of the interior are, if anything, even more ornate than Chirk. But neither have the same sense of silliness that's found in the still more impressive Cardiff Castle, which has paintings of deliberately ridiculous fantasy medieval-esque creatures. Personally I prefer the authenticity and atmosphere of something that's at least partly ruinous, but palace castles are also nice.

From there we went to the Netherlands, which is flat as flat can be, though you can't move for windmills. There we ate delicious home food until we nearly exploded, came back to Prague, and then Boris Johnson decided to try something that's not been done since Charles I. Fun times.