+++*

Symbolic Forest

A homage to loading screens.

Blog : Post Category : Geekery : Page 15

Advice

In which we pass something on

Do not forward this email.

If you do not forward this email, nothing bad will happen to you. No terrible ancient email curses will be unleashed.

If you do forward this email, your wish will not come true. You will not receive unexpected love, or come into some money.

If you do forward this email, the missing child will not be found. Noone will break any world records. Bill Gates will not send any money to charity. Luck will not come your way.

This is not a virus, and there has never been any virus of this name. A websearch could have told you that. So don’t tell everyone in your address book about it.

Pass this information on to everyone you know.

Insert catchphrase here for easy headline

In which we get wary of the talent

As for Doctor Who: as you’ve probably heard, catchphrase-based comedian Catherine Tate is going to be back in the show for a whole series. It’s been in all the papers, after all, and lots and lots of people, who shudder in terror at the mere mention of the name Bonnie Langford, think it will all go horribly wrong. It might be interesting to see if Tate can act, rather than just mug through with a comic voice and lots of makeup until she gets to the catchphrase.

Russell T Davies has been widely quoted, in connection with this story, as saying:

We are delighted that one of Britain’s greatest talents has agreed to join us.

Strangely, though, his thoughts on Catherine Tate herself have not been mentioned.

Lost terminology

In which a word is snappy but fails to catch on

Jargon changes over the years; bits of it get picked up, some bits become mainstream, and some wither away.

On a trip to Wet Yorkshire the other day, I started thinking: there’s one piece of jargon which I think it’s a shame didn’t get picked up. It’s Charles Babbage‘s term mill, which he used to name something that was, for him, a new concept: a machine which would carry out arithmetic calculations according to a sequence of instructions. Today, we’d call it a computer CPU; but there isn’t really any better term for it other than that awkward three-syllable abbreviation. I’d much rather be talking about the newest Core Duo mill, or Athlon mill; it rolls off the tongue. A twin-mill machine sounds much snappier than a dual-processor one. If you look at one under a microscope, it even looks vaguely like the giant mechanical grids of 19th-century looms,* just like the mills Babbage was originally alluding to. Is there any chance of the word making a come-back? Probably not; but it would be nice if it did.

* I was tempted to take up “loom” and segue into a Doctor Who discussion, but that reference would be too geeky even for me.

Sleeping satellite

In which we notice that Google Maps is a step back in time

Google Maps has recently, it seems, spread its high-resolution satellite coverage over much more of the UK than before.* It now covers, for the first time, this part of the world.

I spent quite a while looking at various places around the area, seeing what I could spot; and it quickly became obvious that even though Google have only uploaded the pictures recently, they’re not new pictures. I asked around the office for advice, and as far as we can tell, the pictures are four or five years old. The new cinema next to the Boating Lake is, on Google, an empty field. My car isn’t anywhere to be seen, because I didn’t have a car back then. Wee Dave spotted his own car, outside his old house.** Various other buildings haven’t been built, or are still there on the pictures having been knocked down a few years ago. It’s intriguing; and I can’t help wondering just how Google picks its areas to upload, and if it’s been sitting on these tiles since Google Maps UK first started.

* Thanks to Martijn for mentioning it, otherwise I wouldn’t have noticed

** actually, he spotted “the one that the missus wrote off a few years ago”

Recent Search Requests

In which we know what you’re looking for

deglutation
wemyss bay station
why forests need to be saved – I don’t know, they just do
ravens where to see them in south east england – I’d suggest the Tower, personally
steps of doing long division computer geekery definition ball gagged police
why was war between bosnia and serbia – trust me, it’s a long story
gothic and depressive computer desktop backgrounds
goose to blame if i lose my balance
the bad things about solar collectors
splosh fetish

I think that’s enough of that

Frustration

In which things always go wrong … unless we want them to go wrong

A Work Story.

We need a new printer. The MD says: “Order a new printer!” Our manager waits until he’s out of earshot, then says: “get Spare Printer X working and use that instead.”

So, I find Spare Printer X out, and do manage to get it working. I test it. It seems to be fine. But then, a strange thing starts happening.

I give it a page to print. Let’s call it Page A. It prints it. All is well.

I test a different page. Page B. The printer happily prints another copy of Page A.

A third page to the printer? Out comes Page A again.

Let’s try a four-page document. I get: four copies of Page A.

Switch to a different application. It works! It prints what I tell it to—Page X this time.

I print Page Y from that application. I get Page X again.

Go back to the first program. Still printing Page A.

Let’s reboot the printer. Let’s print. Oh look, Page A.

OK, it’s not the printer. Let’s reboot the printer, and the computer, wait ten minutes, turn them back on. Check there are no files spooled and waiting. Print something. Out comes: Page A. Now this, surely, is physically impossible.*

The boss pops down to check how I’m getting along. “It’s borked,” I say. “It only ever prints copies of the first thing you told it to print. It’s useless. Look.” I repeat my last, failed, print request. It prints perfectly. Arse.

“Looks fine to me,” says the boss. “Put it in, and see if they have any problems.”

Of course, I know it’s never going to work now.

* or at least, extremely improbable, if you follow Sherlock Holmes’ philosophy.

Bigger on the inside

In which we play “spot the plot hole”

Whilst we’re kind of on the subject, from yesterday: Doctor Who.

Who’s this Saxon chap, then? What’s he up to, and what’s he going to turn out to be? Some dirty politicking with an eye to mad-eyed global supremacy, is my guess to the first. The second: well, it could be anything really.

As to the episode itself: I’m sure Russell T Davies has a machine hidden in his basement somewhere which stamps out little, villainous old ladies. It’s not that he uses them a lot, just that when he does, they are instantly recognisable, always virtually the same as each other. The whole thing was: well, nothing special. What was going on with the Doctor temporarily dying from lack of blood, then being revived by CPR? How did that work? If people in the middle of the hospital were on the verge of death when it was returned to Earth, did any actually die? It would take a while, after all, for oxygen levels in the centre of the building to return to normal. What about people investigating the big crater the hospital left behind it – were they all squished when it returned? Answers on a postcard to Symbolic Towers, 4 Iambic Avenue… OK, maybe I’m being slightly too serious. It’s entertainment, after all. Doctor Who has always had plot holes, and it always will; I shouldn’t expect it to be harder-than-hard SF because it clearly never has been that. It did entertain me, and that’s all I should ask for.

Facing points (part two)

In which we go over some railway history

More notes on the Lambrigg and/or Grayrigg train crash from a couple of weeks ago. Continued from here.

As I said in the first part, it was known for many years that junctions are a dangerous thing. Any place where a train has a choice of routes to take is a danger point, and the railways, for a hundred years or so, got around this by avoiding them as much as possible. A freight train, going into a siding, would have to run past it, stop, and back up slowly into the siding.

This is a very safe and careful thing to do, but it is very, very slow. Trains take a long time to slow down, and a long time to stop. Backing up has to be done very slowly, too, and the whole operation blocks the main line for rather a long time. If the train could run directly into the siding, things would be a lot faster.

Similarly, if one line of a pair has to be closed for engineering works, trains have to run in both directions over the remaining line. The old way of doing this was very slow indeed – the train would have to stop, reverse backwards onto the other track, then reverse again so it was going forwards. All very fiddly and slow,* and it would have been easier if there was a faster way to do things.**

So, in the 1960s and 1970s, an awful lot of the rail network got simplified and redesigned. In particular, “emergency crossovers,” like the ones involved in the Lambrigg crash, were installed every few miles on the main lines. Essentially, all they were there to do was let trains switch across to the other track if one line had to be closed for maintenance. This, though, meant greatly increasing the numbers of relatively dangerous, maintenance-heavy facing points on high-speed main lines. Cost was no longer so much of an issue – greater automation and mechanisation of the railways meant that all points were fitted with exactly the same locking equipment, so the legally-required and previously expensive locks on facing points were now provided for free. Maintenance still mattered, though.

Note that I said “relatively dangerous”. Facing points are maintenance-heavy, purely because they are intrinsically more dangerous than trailing points. This isn’t an issue, though, so long as the maintenance gets done. And, over the years, all points started to be given the same level of maintenance – there is in many ways no longer a distinction between facing and trailing points, maintenance-wise, because as I said above they nearly all have the same fittings.

So long as the maintenance gets done. That is the key. Railways just aren’t maintained in the same way that they used to be. There’s no longer a man walking every stretch of track, every day of the year, looking out for faults, like there used to be. If facing points aren’t maintained properly, they become dangerous, and they’re likely to cause accidents, such as Lambrigg and Potters Bar. The problem is, they’re vital to being able to run the railway smoothly and flexibly. If you want to run a flexible railway, it’s going to cost you more. You have to be willing to pay the price, however you want to pay it.

* there are lots of other rules involving people waving flags and people whose job is just to be unique, but I won’t bore you with them.

** This has nothing to do with the closing of alternative routes, incidentally, which people sometimes go on about as being a Bad Thing in connection with the rail network. Alternative routes are often a lot less useful than people think.

Facing points

In which we delve into railway history following a recent accident

Some notes on the Lambrigg rail accident (also known as the Grayrigg crash in the media).

Facing points are a bad thing. Facing points have always been known to be a bad thing. This has been known ever since the railways were first created.

A facing point is, essentially, a junction, where one line of rails splits in to two. The opposite is a trailing point: a junction where two lines come together to become one. You will quickly realise that a facing point becomes a trailing point if you stop your train and reverse it, and vice-versa.

Facing points are dangerous; trailing points are safe. You can drive through a faulty trailing point, or a trailing point set the wrong way, and you’re unlikely to have your train come off the track. You’ll probably wreck the point, unless it’s designed for you to do that to it,* but your train will be unharmed. Do that with a faulty facing point and your train is going to end up all over the place.

Now, this was never a problem, because for years main-line railways only ever had tracks in pairs, one track for each direction. Going the Wrong Way was strictly against the rules. The main reason for this was to stop trains meeting head-on, but it had a secondary benefit: it meant that engineers could get rid of as many facing points as was possible. This was partly an expense issue. Anyone who’s ever had a train set will know that if you switch a facing point whilst a train is on it, Bad Things will happen as different parts of the train try to go in different directions. This isn’t what happened at Lambrigg/Grayrigg, but it has the same result; and when the government realised, they quickly insisted that all facing points be fitted with a complex arrangement of locks and train-detectors to make sure you can’t do that. Back then, that involved mechanical locks which needed a lot of careful and regular maintenance and adjustment. Now, most of it is done electrically, but there is still a mechanical lock somewhere in the point’s machine that holds the various moving parts of a point fast when a train is nearby. Of course, that’s only any use when the rest of the point is mechanically sound too.

So, anyway, as I said, if all lines are one-way only you don’t need facing points. Not until you get to big junctions, at any rate, where you have to live with them. Freight lines didn’t need the expensive facing point locks, so freight trains always backed into sidings. And the railways happily ran like that for a hundred years or so, and facing points rarely caused accidents. In modern times, though, it didn’t really work.

*The second part of this article continues here*

* Lots of points on rural lines, nowadays, are what’s called “sprung points”. They’re not controlled, they just sit there. Use them as a facing point, and they’ll always send you the same way.** Use them as a trailing point, and you can approach them from either route without problems.

** Left, usually, on British main line railways at any rate.

Legal news

In which Microsoft are on the good side for once

Legal news of the week: Microsoft has lost a patent infringement case brought by Alcatel, the company that owns the rights to the MP3. That is, they don’t own the file format itself, but they own the patent on understanding what they mean.

Now, normally, “Microsoft losing a court case” would be Good News for computer users everywhere, because Microsoft generally aren’t a very nice company and seem to spend most of their time thinking up new ways to extract money from people.* This case isn’t, though, because software patents are a bad thing, a bad thing indeed. If you’re a geek you can skip this next bit, because you’ll already know why they’re a bad thing.

Software is, basically, a list of instructions for doing arithmetic. Forget all the flashy graphics you see on the screen. Forget your email and your IM programs. Computers are machines for pushing numbers around,** and computer software is a list of instructions for doing that. Remember doing long division at school? That was essentially a list of steps for working out division sums that are too hard to do in your head – software for your brain, in other words.

Now, imagine if the inventor of long division*** had patented it. Every time you did a long division sum, you’d have to pay him a royalty. If you invented a machine to do long divisions for you, you’d have to pay a bloody big royalty. That’s how patents work.

Software patents are even worse, because often they involve access to data which is otherwise locked up. All those MP3 files on your computer? There’s no practical use for them without decoding software. Decoding software is patented. Microsoft thought they’d paid the patent holders for the right to write a decoder and sell it with Windows – but then the patent holder changed, and the new owner thought otherwise. The courts agreed with them.

Imagine if the first person who ever thought of the idea of reading a book in the bath had patented it. They managed to get a patent on the following: “run bath, select book, get in bath, pick up book, hold book in a cunning way to avoid getting it wet, read.” That’s no different, essentially, from a software patent that involves reading data from a file. If someone had done that, then you could only read a book in the bath if you’d licensed the right to do so. That’s why software patents are bad and wrong.

In more amusing legal news, the right-wing UK Independence Party has been told to return over £350,000 in illegal donations, made by a businessman who wasn’t registered to vote at the time. The party think the ruling is ridiculous. It shines a light, though, on the underside of their philosophy. There are rules there to ensure that only British people with a stake in British politics can fund political parties. UKIP think the ruling is silly because the man is obviously British even though he couldn’t prove he was a British voter. Which just goes to show that they’re not interested in proof or evidence or process; their definition of Britishness seems to be that you’re Someone Like Us.

* which, to be fair, is what capitalist companies are supposed to do.

** that’s why they’re called “computers”, and not “communicators” or “info-readers”, despite that being their main use.

*** apparently the sixteenth-century Yorkshire mathematician Henry Briggs, according to this lecture from his old college