Sam Kington

It’s not even December yet, but already the Christmas decorations are out on people’s houses. In this year of Covid I’m less inclined to grumble than normal - let people get fleeting moments of joy when and where they can - but I do feel I need to point the avoidable error in some of the decorations this year.

Namely, the “Santa stop here” signs. Specifically, the ones that include a road traffic STOP sign.

The immediate problem is that the novelty sign designers have betrayed an elementary misunderstanding of the Highway Code (and indeed its European counterparts). Circles give orders, triangles warn, rectangles inform; it’s pretty clear. The sign should have been e.g. a Santa and reindeers parked on a roof as a white outline on a blue circle; or, possibly, a tourist information sign, depending on how uncertain you were that your children had been good this year.

But OK. We can chalk that up to (a) merchandise manufacturers not caring, (b) children not being sufficiently versed in the highway code to be well-informed consumers, or indeed that (c) the highway code itself says “there are a few exceptions to the shape and colour rules, to give certain signs greater prominence”, and one of them is the standard STOP sign. Santa is important enough to get the same healthcare as the US President, so having his own personal sign isn’t that big a deal.

No, the problem is that you’ve put a sign on a destination where it belongs on an intersection.

The standard STOP sign, after all, is designed to look different because its purpose it to say to a motorist: if you see this sign and don’t do anything, you might well plow into traffic and die. Given the speeds Santa travels at so he can deliver presents to the entire world in one day, this is a pretty terrifying message to send. The implication is that while ordinary houses are just meat and potatoes to him, this one is home to e.g. the ride of the Valkyries, a cavalcade of witches from Walpurgisnacht, or some other huge high-volume highway full of supernatural creatures that would send Father Christmas’s reindeers and sled flying if he blithely flew into it without taking precautions.

Or: “You better watch out”.

Some time in the mid-nineties, my life changed. I was at University, and had decided that this time, I knew what I wanted to do. I’d made new friends, I was in an exciting new city, I was happily three years into a four-year Arts degree with time to kill, and I was pretty sure that I would become a journalist.

And then I discovered the Internet, and I realised that a career in computing could be interesting: not just a matter of writing a slightly better spreadsheet app. I dived into all sorts of weird new-fangled technologies that, I was certain, would be the making of me.

With the exception of the web browser, though, all of them involved a terminal.

This is my terminal. There are many like it, but this one is mine.

You’d log onto a machine that had a WYSIWYG word processor with as many as 256 colours, and could play a mean game of Solitaire, and then you’d disappear down a rabbit-hole into yesterdecade’s monochrome and monospaced 80x25 screens. You started off with Windows, Icons, a Mouse and a Pointer; you’d end up with a editor that was designed for an era where sending a Ctrl keystroke down the wire was a waste of bandwidth, so as a result lesser mortals couldn’t even quit it.

The learning curve was so taken for granted that when AOL brought the Internet to the masses a few years later, wits quickly described it as the September that never ended. The novelty wasn’t that the Internet was difficult to understand, but that the influx of newbies was no longer circumscribed in time.

“Sit down and be quiet for a week or two”, the oldbies would say; “get an idea of what this community is like rather than barging in and asking stupid questions”. And that was generally good advice: while the alt.fan.pratchett newsgroup might have officially been about Terry Pratchett’s books, what the regulars, in their hearts of hearts, really wanted to talk about was what various anglophone countries called different bread products. (That the modern Internet that replaced Usenet still cares about that sort of thing is the best example I know of the universality of the human condition.)

Often from an academic background, the oldbies would look down at people replying with no real content, which they’d summarise in cod-HTML as “Me too!”. “Post something interesting or don’t post!”, they’d say. They’d despair at flame wars and spam, which weren’t part of the plan at all.

“Also”, they’d grumble, “learn to quote”.

Quoting is hard; let’s go shopping!

“How do you quote?”, newbies would ask, and the instructions would basically be:

• Your news reader will automatically quote the post you’re replying to;
• Delete anything that doesn’t matter, because (a) we’ve seen it before, and don’t want to scroll past it again, and (b) bandwidth is expensive

Notably missing from these instructions would be what do do if the news reader’s attempt at quoting failed.

For one thing, there were many news readers, most of them were configurable, and as a result you’d see people quoting in a number of ways. Look at this thread on comp.os.linux. People mostly quote with “>” (although whether to put a space after it or not varies), but also use “:”; elsewhere you’ll see “|” or even “AB>” (where AB are the original poster’s initials). Everything’s mostly OK, until you get to this post where word wrap has fatally intervened: the quoting characters have been misunderstood to be part of the message, and are now mixed in with the original message.

Microsoft Entourage for Mac OS 9 was the first email client I ever used that had the ability to increase or decrease the quoting level of the selected text, and that shipped in something like 2000. If you used anything else, manual editing, writing your own macros from scratch (the accepted solution by text editor elitists), or acceptance that the quoting was broken, were your only choices.

Please welcome the top-posters. They don’t use email like you do.

Or, of course, you could top-post.

Bottom-posters like to say stuff like this about the controversial style of replying that Microsoft Outlook popularised (possibly because it was such a terrible text editor that it couldn’t support bottom-posting properly):

> Because it breaks the flow of conversation
>
> > Why is top-posting bad?

But, in truth, top-posters didn’t care about that, because they weren’t reading emails individually; they were reading them historically. They knew which emails in the thread they’d read before, so when they read a new message, they’d read until they saw something that they’d seen before, and then stop.

Suddenly, all sorts of new things were possible. You could say, AOL-style, “I agree”, hit send, be happy, and move on. (Years later, Facebook would invent the Like button, which served the same purpose.)

And crucially, you could forward just one email, or reply to a thread copying someone else in, and they’d see the entire history. For businesses that was a life-changer; previously, you’d have had to have manually forwarded every single email, because replies would have edited out the context the individual emailer didn’t care about at the time, but which a newcomer needed.

(This isn’t perfect: it assumes that people reply sequentially. If two people reply to the same message at the same time, the email chain forks and becomes two threads. But the speed that top-posting gives you makes it less likely that this will happen; and if you see a reply arrive while you’re writing yours, you can just cut and paste what you wrote into a new message without having to redo all your formatting.)

Why is email still bad? Because we don’t understand the changes we made.

Of course, not everything in the land of top-posting is sweetness and light.

The use of HTML in email can lead to problems: if something somewhere (typically code or a table) was marked as “do not word-wrap”, and someone in the chain is quoting old messages and indenting them, the message can eventually end up being wider than your mail client’s window and impossible to read without constant scrolling back and forth.

And sometimes you do need to reply inline to something someone said. The argumentative, nit-picking style of academia doesn’t fit the more conversational style of a business; but occasionally you do need to delve into the details of someone’s proposition. Posters who aren’t used to dealing with multiple levels of quoting end up using colours or bold to differentiate their words from others’, and it can be painful to watch.

But these problems both stem from the mistake Outlook and its successors made: including the message you’re replying to in the message you’re writing.

If, when you reply to an email, your message is a standalone section, and you include the original email as an attachment, then you don’t have to worry about any formatting interference between the two. Someone’s monster sig stays in their own message and doesn’t affect yours. Previous messages are stacked below yours rather than repeatedly indented until your monitor runs out of space.

And if there’s no way of editing the message you’re replying to (and why should there be? That would be a devious and subtle way of gaslighting people into thinking that the thread history was something other than what it actually was), then, sure, you need to learn how to quote. (Or, rather, to cut and paste the parts that are appropriate.) But (a) this is easier in modern email clients which know about HTML and e.g. paragraphs and blockquoting, and (b) emphasises how rare it is to have to quote parts of a previous email in detail.

Learn from the US Constitution: ideas don’t always stand the test of time

Americans often have a quasi-religious attachment to the US Constitution; but whenever the US has been involved in writing a constitution for another country, it’s favoured a parliamentary system rather than the weird hybrid presidential system the US uses. And so defenders of traditional email should similarly be prepared to answer: if you were designing email again, would you do it that way? And have, in fact, any similar systems used your type of design?

The answer is clearly no. Web-based commenting systems (forums, Facebook, Reddit, Disqus; hell, even that dinosaur LiveJournal) either disable quoting the original message by default, or more often force you to quote manually. And they all agree that text is written in paragraphs and word-wraps, with those rare exceptions when you need to control the exact formatting being made a feature or flat-out not supported.

I have sympathy for those people who bemoan the overuse of HTML in email, and yearn for the good old days when emails were just text. But we didn’t write 80-character fixed-width text documents because we wanted to, but because that’s all our tools could support at the time. Now that we have better tools, it’s time to move on.

COVID-19 is upon us, and that means encouraging people to self-isolate, even if they don’t yet apparently have, as a recently-ex-colleague put it, ‘OMG TEH LURGI!’.

For those of us whose livelihood involves going around to people’s houses fixing things, there are interesting debates about using self-employed people as guinea pigs for universal basic income. Regarding those of us who work in retail there are important debates about who you should favour when deciding that people should be allowed to miss mortgage payments. And I’m fascinated with the idea that COVID-19 is bringing about the next phase of globalisation, where we both collapse into pan-regional concerns and become more global in outlook in different areas.

But I’ve been working from home for almost 20 years, so I’m going to talk about working from home.

It “helps” that plenty of people have been writing about this who don’t really understand it. Earlier this month the New Yorker ran a piece about working from home that cunningly combined the standard clichés of working from home with the trope structure of Dog 911. And while I gladly pay for the Guardian, their article about the pitfalls of video-conferencing from home had as one of its featured comments someone directly contradicting one of its main points, and extolling the virtues of “winnie-the-poohing” (an age-old practice of “only formally dress the top half of your body” that foreign correspondents and TV presenters safely sitting behind a desk have long been proponents of).

And it’s true that you just might not be cut out for working from home. You need the self-discipline to do some work without having a boss in the neighbouring cubicle glowering over you (so bad luck to Mr. “my girlfriend hurt her knee and I had to be with her, so that’s why I did no work the other day”). But in truth, the opposite temptation is true, especially for geeks who are actually really interested in working on cool stuff (this is why Silicon Valley provides gourmet meals for young, single, male programmers: because it’s much cheaper than overtime). You can easily end up working in your spare time because, well, you had no other plans anyway!

And as soon as anybody in the team is working remotely, everyone else has to. There has to be a mic and a camera in any meeting, for the people who can’t be there physically; general chit-chat has to happen over email and some form of IM (IRC, Slack, Teams etc.); you have to make sure that nobody is disadvantaged by not being part of the serendipitous hallway conversations that happen in any large organisation.

This sounds like a hurdle, and for a small team it totally can be! But in truth, if you’re part of a large company, especially a multiple-office company, you’re already working remotely. You just haven’t realised it yet.

At $WORK, we have a London office, a bunch of Manchester offices, a large Utah office and a smaller one, a Lviv office and we also have some people in India somewhere. That’s just the offices I’m aware of; we could well have more. In truth, as with all things regarding programming, the important transition is between one and more than one: as soon as there are two offices, you need to think about which office is involved, and adding a third, fourth, fifth and so on doesn’t change this to any significant degree.

If you can’t just wander around the office, you need tools to find out if someone’s in an office today; you may need to email someone rather than just asking them something; in extremis, you may need to think about time zones. Rather than roping people into a meeting, you might fire off an email; and a meeting might have to involve people teleconferencing from many small rooms rather than everyone congregating in a large room.

And when you do that, an interesting thing happens: people get flexitime, almost emergently.

Because as soon as you get rid of the crazies who hate email and want to bug you at every time of day (scroll down to “Use other means of communication” - sorry, Guardian!), and you recognise institutionally that not everybody works flat out from 9 to 5 your time, you then suddenly allow all sorts of more positive working patterns.

For one thing, you can encourage non-uniform work patterns that still achieve the same desired effect. A colleague of mine takes time off to take his children to and from school, and makes the time up later on in the day when they’ve gone to bed. As long as Jason from Accounts doesn’t insist on him answering his phone every possible minute from 9 to 5, who the hell cares when the work gets done, as long as it gets done?

Similarly, sometimes I go to one of our offices down-South, and eventually find myself mentally blocked; often the very act of upping sticks and heading back to the hotel is enough to get things sorted again. Some office enforcer saying “stay! you haven’t done your allotted hours!” is missing the point here.

Even people who work in an office might decide to come in really early, and/or really late, to avoid rush hour. (If you do both you can presumably eventually accrue extra holidays.)

And if I have no meetings scheduled for this afternoon, I might decide at about 2pm “yeah, it’s the first nice day for about 4 months so we’re taking Ella for a walk in the park”.

So while we’re all learning, belatedly, how to work from home, the potential ramifications are far more significant, and far more interesting.

Wings over Scotland says, amongst other things:

a second EU referendum and a second indyref do not exist in isolation from each other. Each one damages the interests of the other. An independent Scotland would reduce the rUK’s chances of reversing Brexit, and reversing Brexit would damage the arguments for an independent Scotland

The first part is obvious: given that Scotland voted en bloc for Remain, rivalled in its intensity only by Gibraltar, London and Northern Ireland, then if Scotland had already been independent by the time the Brexit referendum was held, Leave would definitely have won. And if by any chance Scotland managed to declare independence while the rUK was out of the EU, the task of rUK Remainers to turn around an increasingly ideologically-polarised polity and rejoin the EU would be that much harder.

As for reversing Brexit making independence less likely, there are, I think, two reasons, neither of which completely satisfy me. The first and most trivial is that a People’s Vote which reverses Brexit but ignores IndyRef 2 (the subject of Wings’ most recent ire) might bring to power Unionists who would screw over Scotland yet again. (How exactly this would differ from the current Unionists who are quite happily screwing over Scotland is left as an exercise for the reader.)

The second is that Wings believes that Indyref 2 can only be won by appealing to leavers - specifically, yes voters who are now no voters, and voted leave. The basic principle is sound: whether it’s because the EU27 are principled (the four freedoms are sacrosanct, and the EU will not survive if countries are allowed to cherry-pick), or ruthless (throw the UK against a wall just to show the world that the EU means business), let some other country try leaving the EU first. If that works, an independent Scotland can join the rUK in declaring article 50. If it doesn’t, well, now we know.

Still, though, it seems … odd to argue that Scotland and Northern Ireland remaining in the EU is the only way to fix the Northern Ireland border problem, but all of the UK remaining in the EU would somehow scupper independence. If the Good Friday Agreement has taught us anything, it’s that answering questions like “is Northern Ireland part of the UK or part of the Republic?” with “all three are part of the EU, so does it matter?” is a very good idea. Irrespective of your opinion about GERS, it’s undeniable that the arguments against Brexit - e.g. the sheer amount of red tape and physical space for customs infrastructure required, and the corresponding economic slowdown - are also arguments against Scottish independence from a rUK outside the single market and customs union. One of the reasons the Indyref campaign did as well as it did was that Alex Salmond could confidently reassure people that we’d keep the Queen, the pound and the BBC. Do you really want to say “OK, Brexit is a shitshow, but we would do independence properly”?

And more generally: popular opinion is already turning against Brexit, and that’s before anything has really happened yet. Charlie Stross is holding off writing a third near-future Scotland book because nobody has any idea what’s going to happen in the next few months, let alone years. Talk of a new centrist party always runs up against the formidable barriers of the UK’s FPTP electoral system, but it wouldn’t be surprising if one or both of the two major parties split, and/or the UK polity reorganised itself on constitutional lines rather than the current and traditional right/left economic divide.

So by all means look at opinion polls and try to work out how to build a coalition for independence. But when Indyref2 comes, chances are that a lot of people will have changed their mind.

My standard datetime library didn’t have a standard way of saying “th” for days - you know, “Monday 1st April” vs “Tuesday 2nd April” vs “Wednesday 3rd April” vs “Any day xxth April probably”. So I went looking for solutions.

My favourite example was probably Lingua::EN::Numbers::Ordinate because of the way it iteratively works towards the proper solution:

sub ordsuf ($) {
  return 'th' if not(defined($_[0])) or not( 0 + $_[0] );
   # 'th' for undef, 0, or anything non-number.
  my $n = abs($_[0]);  # Throw away the sign.
  return 'th' unless $n == int($n); # Best possible, I guess.
  $n %= 100;
  return 'th' if $n == 11 or $n == 12 or $n == 13;
  $n %= 10;
  return 'st' if $n == 1; 
  return 'nd' if $n == 2;
  return 'rd' if $n == 3;
  return 'th';
}

And the approach I least favoured, even though it’s far more “efficient”, was this one from Date::Format:

@Dsuf = (qw(th st nd rd th th th th th th)) x 3;
@Dsuf[11,12,13] = qw(th th th);
@Dsuf[30,31] = qw(th st);

It’s not just the way that it decides “let’s just do 0..9, 10..19, 20..29 automatically, then manually add 30 and 31” (presumably because that’s part of its data validation - if you ask it for an ordinal of 32 it will tell you that there’s no such day). Or the repetition of many “th” between 4 and 9.

It’s the fact that the programmer has decided “OK, what’s the problem?”, found a solution, and then decided “OK, how do we make the solution the most efficient possible?” and golfed their way towards the implementation. This is read-only code: it assumes that the problem has been solved, that this is the best way of solving it, and provides no information about why any of this ever happened (which is usually the way of finding bugs: to realise that the previous implementor’s approach was wrong).

The most annoying commonplace falsehood about programming is “it’s all about ones and zeroes”. It’s not: it’s been a lifetime since anyone actually programmed a computer by inputting ones and zeroes into anything. It might be ones and zeroes under the hood, but that’s as interesting as saying that the life as a materials chemist is all about quarks.

The second most annoying commonplace falsehood about programming is slightly more interesting: it says that programming is all about science, and maths in particular. That may be true in the more refined parts of our industry, but for the most part we’re writing systems that interact with humans rather than particle accelerators or lunar landers. And the way we write code should reflect that.

So: if your code is all about O(n) efficiency or what have you, by all means make it efficient. But if you’re just trying to deal with real-world problems, write the code in a way that resembles the real-world problem that you’re trying to solve. The future maintainer will thank you.

I used to wake up effectively mummified: two dogs on top of the duvet, on either side of my legs, penning me in. With Taji and now Habibi gone, that’s not going to happen any more.

Ella realised something was Wrong yesterday morning, and she spent some time in places she normally wouldn’t - under my desk, or at the back of the office where there’s normally no room. She slept beside the bed when I took a nap that evening, even though (I’m guessing) she would have preferred to be outside in the garden. Did she think that Habibi had been guarding us, and now that Habibi was gone that that was now her job?

It’s possible. Ella is unusually emotionally-sensitive for a dog: whenever I shout at the computer, I can expect her to be there within seconds, concerned that I’m upset. So I hug her, because I want her to be happy, and it turns out that hugging a big happy walking carpet is good for you as well, so we’re both better-off.

There’s a thing some people do as part of US Thanksgiving, I understand: where you go round the table and each person says “I’m thankful for thing, because reason”. Well, losing Habibi was heart-breaking, but the saving grace was that we still had Ella. So we bundled her up into the car and took her on a good long walk around Mugdock Country Park, and for a while we all had a great time.

We couldn’t have done that with Habibi, because she was ill and didn’t want to go far. And chances are that Ella wouldn’t have got to sniff and play with a number of the dogs we met even if Habibi had been in perfect health, because she’d have barked her head off and we’d have moved on quickly.

And when Berkeley was alive and I’d yell at the computer, he thought I was angry at him and would run away.

Now, simple is not better. Part of loving someone is accepting the baggage that comes with them; and you do this because it’s their baggage. And it’s going to be very strange tonight when Cleodhna takes Ella to Jessie’s and the house has no animals in it for, quite possibly, the first time ever since we moved in.

Still: that Ella is, objectively, a superb and wonderful dog, makes things a little better.

Some time in June 2007, a small brown dog was abandoned in Glasgow’s Botanic Gardens. She spent almost a week cowering under a bridge before Patsy, a sometime friend of ours who can’t help rescuing dogs, managed to coax her out. Patsy nursed her back to health, but then needed to leave Glasgow for a few days, and asked us if we could dog-sit her for a few days?

That was in the afternoon. By the evening we were thinking of names.

(We thought she was part-Basenji, so decided to call her something Egyptian, only to discover that after Cleopatra, Nefertiti and Neferet, all other female pharoah names are unpronounceable. So “Beloved” in arabic it was. Then we realised that a Natasha Atlas song we were listening to at the time Mondegreened as “Habibi leash”. Whoops.)

Habibi had a beautiful coat, like a tortoiseshell cat’s but subtler, each hair a slightly different colour from its neighbour; and with a fun little white bit on her bib that would change shape when you frobbed it. But most importantly, she was delightful. Our cat Helen had a switch-on purr; well, Habibi had an insta-wag. You’d just have to walk into the room and say her name and that thing with its little twist on the end would spring into action. Often, when you were cuddling her at the foot of the bed, she’d get so excited by tummy rubs that she’d have to jump off the bed and run around. One day she asked to go outside, and when I got up to go downstairs with her, she jumped up and made an excited sound like a skeksis opening a screen door.

She was also smart. Most dogs get so excited about going outside that they start running around while you’re trying to put their harnesses on, which of course means you can’t put the damn harness on them because the damn dog won’t stay still, so they don’t get to go outside after all. Habibi did all of that, of course, but once you’d put her harness around her neck she’d lift her paw up to make it easier for you to put the rest on. (When we were in France, she would also lift her paw up to make it easier for you to carry her up the stairs; stairs that she could go down perfectly well but refused to go up. We tried putting beeswax on her paws; that didn’t help.)

We don’t know how old she was; we thought she was about 1 or 2 when we found her, but then she got sclerosis in her eyes, which normally happens to older dogs. No matter; she got grey in her muzzle, and she progressed from barking just at the postman and black dogs to barking at pretty much everything, but she stayed the same wonderful little brown mutt.

She started having off days, not wanting to go out on walks, towards the end of last year - nothing you could reliably put down to anything in particular. She had insurance, so we did blood tests and scans, but nothing came of them until our vet noticed a bulge in her abdomen. At that point things proceeded quite quickly: she had her 1kg (!) spleen removed, and the diagnosis came back. Hemangiosarcoma, the same cancer that killed Laszlo.

You can’t do anything about that, and she didn’t like being prodded by strangers in vet surgeries, so we settled in to pamper her outrageously while we still could. She had a bad moment last Sunday, so we arranged for our vets to come over today and put her out of her misery. It’s strange to set a date for your dog to die, especially when she appeared to be doing a bit better today; but we didn’t want her to be in any more pain. Think of it not as a death sentence, but a death promise.

Goodbye Habibi. Pretty girl! Clever girl! I miss you already. This doesn’t get any easier.

As she got older, my mother decided to exercise her brain by learning new things. She learned new Romance languages until they formed factions inside her brain and ganged up on each other when she was busy. She went to ballroom-dancing classes. She made pots.

Cleodhna does crochet. She has long been known to make little figurines out of yarn for people’s birthdays, but the hobby has accelerated recently. She has storage thingies that go under the bed like a normal geek, but hers are full of yarn rather than cables or Lego. She has favourite yarn shops.

She has this overarching project to make something like 48 large granny squares and 96 small granny squares, and then to sew them all together and make a quilt. She’s about half-way there. She started off following this enthusiastic woman on YouTube who had a “do one granny square every day!” channel, but she’s now branched out. She buys patterns on the Internet from fellow enthusiasts. And she knows how to make more than one hat out of yarn.

Here’s how I know: I walked through the door and asked her something, and she lost count.

As far as I can tell, crochet is about doing the most complicated things you can possibly do with a bunch of thread, while doing the strict minimum amount of cutting or knotting. Any old idiot can take an embroidery grid and a bunch of coloured threads and make a picture, and never mind that if you turn the thing over you see a complete rat’s nest of cut off threads. The platonic ideal of crochet, I vaguely imagine, is to build the Sistine Chapel out of a humungous complicated amalgam of intertwining loops, and then say to the marvelled onlooker “if you cut this knot here, right at the end where The Man said we really needed one, and then pull on this thread, it will all come crashing down around you - very slowly and softly”.

(Marcel Pagnol, in one of his novels about his childhood growing up in Provence in the 1910s and 1920s, had a wonderful bit about the pecking order of people who built walls. The Proper People built dry-stone walls, carefully stacking one stone on top of another until you got a wall that would never fall down. They looked down on stone-cutters, who did the same except that they cheated and cut the awkward bits off individual stones rather than finding the perfectly-interlocking shapes in nature. The stone-cutters in turn looked down on masons, who used mortar to glue their stuff together rather than relying on brilliance and physics. Crochet seems to me like it aspires to the dry-stone-wall school of unnecessarily brilliant excellence.)

As I was saying, I don’t really know about crochet. Cleodhna does it and she’s happy; I get to make jokes about it. And I occasionally ask her something when she’s counting… er, crochet things, and she gets mildly annoyed.

“You should have some way of indicating that you’re doing a finickety crochet thing and you shouldn’t be disturbed”, I said to her this evening, “but that’s going to be difficult as you’re going to have your hands busy.”

“I can make a hat!” she beamed. And it turns out that she has a pattern for a dwarf hat, which includes a yarn beard because of course it does. That wouldn’t be appropriate, because the yarn beard would annoy her unnecessarily while she was being necessarily annoyed by crochet stuff, so she clearly needs another kind of hat. That’s OK, she’s got a pattern for a simpler “I’m busy” hat as well.

It is, clearly, turtles all the way down.

I think I know some people who speak Libertarian, so here’s a question. To my mind, this tragic story is about puppy mills, but it’s more generally about regulation.

Cavalier King Charles Spaniels are notoriously prone to syringomyelia, which means “your brain is too large for your skull” (!), but they’re not the only breed to have been messed about by breeders and dog fanciers. The BBC decided to no longer show Crufts because, amongst other things, the standards for German Shepherds amounted to animal cruelty. (The 2016 winner could barely walk.) The Cavalier King Charles Spaniel people in the UK have a kitemark about trying to do something about the terrible health of the breed, and hooray for them. But the breeder of this tragic puppy said (effectively) “we abide by all laws”, and said nothing about making sure that their breeding stock was healthy - because they didn’t have to.

If you decide to buy a mobile phone from a particular company, and they then confess that it catches fire and explodes, so for your safety you shouldn’t use it, it’s perfectly reasonable to decide to not buy another phone from them again, because you did, after all, plan on buying another phone in a few years’ time. If you decide that, because you dislike Nestlé’s efforts to promote baby milk, or Amazon’s attempts to screw over other publishing companies, you’re going to boycott Nestlé or Amazon, well, other brands are available.

But you don’t have this choice when faced with a buying decision you’ll make at most a handful of times in your life, and that you’ll share with nobody else. Unless you’ve decided that it’s your life’s ambition to live in the same place for 40 years, and own a succession of Cavalier King Charles Spaniels, how are you going to be able to tell which are the good and bad breeders?

And unless you’re stunningly unlucky, are you ever going to deal with a funeral parlour more than two or three times, ever? No - so in the US and the UK, costs are rising. You can hope that industry trends will mean a reversal of the trend eventually, but that’s as much wishful thinking as paying attention to the IMF forecasts: they might well be right, but would you bet on it? No.

So: what do we do about industries like dog breeding (or funeral parlours), where dodgy vendors sell to many people but will hardly ever have repeat customers or word of mouth? The social-democratic response is obvious: regulation enforced by the courts. What do libertarians say?

Something my sister-in-law posted on Facebook today reminded me that I hadn’t told everyone this story!

As a kid and then a teenager, I had this wonderfully eccentric piano teacher, Jacqueline Gallon. (No, you’re right, there indeed aren’t many names that are more French.) She’d move all of the furniture in her house around every few weeks, because she said it would help her relax. Wonderful woman. Anyway, amongst other things, she had cats: a siamese queen, and a persian tom. They had kittens, and the one that she kept was a gorgeous pure-white cat with blue eyes - and therefore deaf. (This is A Thing!. And yes, Deaf White Cat would be a pretty good band name.)

Athar, Attarre, or however you spelled his name, liked to sleep in the baby grand piano that you got to play on if you were good. (Less good students were relegated to uprights.) You know how a grand piano is curved, like a pregnant B if you’re viewing it from above? If you prop open the lid at the front, you’ll realise that on the left hand side the strings go a constant length and then stop, so there’s a semi-circular bit of space at the back between the end of the strings and the wooden wall of the piano casing. Well, the lid was always open, and Athar liked to sleep there. Presumably the vibrations from the hammers hitting the strings felt nice, like a convenient cat-massage.

Anyway, this one day I’m playing this particular Chopin piece for Mme Gallon, and Athar is asleep in his usual spot. And as I play on, I realise that Athar has woken up. It’s not hard to miss; I’m playing the piece from memory so the music stand is down. I’m staring right at him, in increasing dread, as he gets up, stretches, and starts walking along the strings towards me. And I realise that there’s nothing I can do.

There are two things that must happen, and they are in direct opposition. First of all, I have to play this Chopin piece, well, and not stop. Secondly, Athar has decided that being asleep inside the piano is no longer his Thing, so he’s going to walk along to a point where he can comfortably jump out of the piano. Which, given the geometry of the piano, inevitably means the front. You know, the bit with the keys that I’m currently doing things to.

Did I mention that this particular Chopin thing used plenty of pedal? So whenever Athar jumps out from inside the piano onto the keyboard, the keys he jumps onto aren’t just going to sound, they’re going to keep on sounding?

Well, I did the only thing I could. I kept on playing for as long as I could. Athar did indeed jump out of the piano onto the keyboard, produced the greatest cacophony you’ve heard since that time your two-year-old nephew got into a fight with a bagpipe factory, and I don’t remember what happened next because Mme Gallon and I creased up with laughter. (Presumably Athar buggered off as if nothing had happened, which from his perspective was entirely true.)

Mme Gallon did say well done for carrying on playing even though we all knew what was going to happen next, though, which was nice.