Symbolic Forest

An intriguing claim appeared in The Guardian yesterday, buried in its corrections column. Insurance-comparing website GoCompare has stated that there is little connection between its Google positioning and its income. More specifically: dropping from the first page of Google results for a common search term did result in a big drop in traffic, but had no financial effect on the company.

GoCompare is an internet business. As far as I understand it, they rely entirely on their website for income. What they’re claiming is: when Google lowered their ranking, they lost a large amount of traffic - but that none of that traffic, apparently, was making them any money.*

There are entire companies based solely around the premise that they will push you up to the top of Google’s pages. I’ve known marketers spend huge amounts of time and effort on it, taking common search phrases and analyzing them with a fine-tooth comb, trying to find out why they’re below their competition. I’d expect GoCompare to have been doing exactly the same thing themselves, in fact. And they’ve found, apparently, that it was all for nothing - because when Google pulled the rug from under their efforts,** they say, it had no effect at all on their income. Maybe they’re a special case because (unlike smaller online companies) they do spend a lot on irritating minimal-budget TV adverts.*** Even so, it’s intriguing. If their claim is true, there’s probably hardly any point at all in any large organisations spending much effort on search-engine optimisation. Small companies who can’t afford TV adverts, or who produce specialist products - well, that probably doesn’t apply. It probably doesn’t apply to AdWords campaigns either. But general search results? “Overall sales figures were not affected”.

* They seem to be talking about general search results. If they were talking about paid-for adverts, then the drop in traffic would also mean a drop in outgoings; but as they’re not, the change in traffic shouldn’t have any effect on their running costs.

** The Guardian had previously speculated that Google did this deliberately - it was that article which prompted the correction.

*** which, at least, aren’t as annoying as those produced by their competitor Confused.com. I’m never going to go near confused.com, no less than barge-pole distance, because their adverts are that bad. “I’m so stupid I’m trying to get money from a piece of cardboard with a cartoon cash machine printed on it! I’m confused! Dot com!”

Keyword noise: gocompare, insurance, google, search, internet, seo

A couple of weeks ago, I discovered that my site had, apparently, been compromised, but my hosting company had handled it. Wordpress had been broken in the process; but I’m not entirely surprised. Nevertheless, I thought everything was now happy.

This morning, though, a letter from Google lets me know: it hadn’t been fixed. My site has, for the past fortnight, been serving up crap to any passing search engine. This can’t be good. I don’t blame my hosting company: what they did do was above and beyond the call of duty, and they can’t be expected to understand and trace every twisty little maze of code paths in Wordpress that might result on something being sent back to the client’s screen.*** What it does make me want to rant about, though, is PHP.

PHP is - if you’re not a geek and haven’t heard about it - by far the most common “web-programming” language around today. Its modus operandi is: you intersperse chunks of programming code in and around the static content in your web pages. When your webserver reads a page, it will run the chunks of code as a program. In Wordpress’s case, the chunks of code run off to a database and fetch my posts, your comments, and so on, from it, and send them back to a client. Thus, one web page can output many posts, managing them is much easier, and so on. All well and good.

PHP, though, is … well. It’s not exactly the best language for the job, which is being polite about it. I’ve been doing lots of programming in it myself lately, for our Office Intranet, and it’s just not as rigorous as other languages. The syntax doesn’t somehow seem as thorough. Apart from the little differences you always get between languages,* it has little corners that feel slightly wrong when I use them, as if I’m transgressing the boundaries between types of programming object in a bad and dirty way.

That’s just a minor thing, really, just me quibbling. What my big problem is, what makes PHP an utterly unsuitable programming language for its job, is one particular feature much adored by people who want to take control of your website and use it to advertise pr0n and drugs. It’s a feature which is unutterably stupid, so stupid I can’t believe anyone thought it should have been created. PHP will, if you like, go and read a file from anywhere on the internet, and run it for you. Which means that a shifty-looking programmer who gets illicit access to the files on your website only has to add a couple of lines of code, to get complete control of everything. Bang. Like that.

Now, you could say: well, FP, you shouldn’t have been using FTP. And you’re right.** My hosts offer SFTP instead, and I should have been using that. There’s no good reason to use FTP either if you have an alternative available. But that doesn’t mean that the next hole along the line shouldn’t be blocked either. It’s called: defence in depth. At work, we have a high fence round the whole site, and an alarm system just inside it; but that doesn’t mean that we leave the office buildings unlocked. Security shouldn’t be brittle; ideally it shouldn’t be thin either. Once you’ve breached the first layer, the tools to complete the job shouldn’t be left lying around.

* The difference between ‘elseif’ - which is a PHP keyword - and ‘elsif’, Perl’s spelling of the same thing - will forever damage my brain.

** I have a good story about how weak FTP can be - but it can wait for another time.

*** and, indeed, it’s my own fault; I should right away have compared the live files with my known-good backups.

Keyword noise: php, programming, security

It’s a hard thing to do.

The other week, as there’s a long weekend coming up, I booked a camping holiday, in Wales. Only a day or two later, the news outlets started running stories about how awful the Easter weekend weather was going to be; wind, rain, sleet and snow. Oh dear.

There’s still snow on the forecast for Northumbria; but the forecast for the Welsh weather, though, has got noticeably better over the past few days. It’s gone from sleet, to showers, to sunny periods. And I’ve noticed this happening before. There seems to be a tendency now for the forecasts to be more extreme further off, before calming down as the date approaches.

Which is statistically what you’d expect, of course. Extreme weather is, by definition, unlikely, and shorter-range forecasts are always more accurate, so any extreme weather in a long-range forecast is likely to mellow as the forecast gets closer. That doesn’t stop the news jumping on any forecasts of horrible blizzards, though. I’m still worried that the snow forecast for the north Pennines is going to creep southwards over the weekend.

“My dog’s got no nose.”

“Haven’t we been through this?”

“Shush. My dog’s got no nose.”

“How does he smell?”

Well, funny you should ask that. We’ve just joined this scheme called Smelling-Nose Dogs. You know how, in America, guide dogs are called seeing-eye dogs? My dog with no nose now has his own guide dog, who goes around, sniffs things, guides him away from odorous obstacles and generally lets him in on all the latest dog-gossip.* And it’s given him a whole new lease of life! He’s happy, and bouncy, and has a shiny coat!** He’s always bounding around and eager for his smelling-nose dog and him to go for a walk together. Completely unlike how he used to be, always moping in his basket unable to smell anything.”

“That’s nice.”

“Yes, it’s really done him the world of good.”

“Still not very funny, though.”

“Er, no.”

* You know - which dogs have urinated on which lamp-posts and that sort of thing. Which is, I’m told, very important information for dogs.

** Not that that has much to do with anything. Maybe I should be a copywriter for Evil Nestlé’s dog-food arm.

“My dog’s got no nose”

“How does he smell?”

“He doesn’t. He sits around all day getting into a deeper and deeper cycle of depression, because he can’t smell anything, in one huge cloud of nose-related ennui. He never even comes out of his basket.”

“That’s quite sad, really.”

“Yeah, I don’t know why the title says it’s a joke.”

Driving to the office today, I was stuck behind a van which, like small tradesmen everywhere, proudly advertised a free 0800 phone number on the back. And - with little more to occupy my mind - I started wondering: is there any point, any more, in buying yourself a free phone number?

Who pays much for phone calls nowadays? Landline calls cost pence. Does anyone think: “Hmmm. Plumber A and Plumber B are both nearby - but it won’t cost me anything to call Plumber B! Hurrah!” Secondly, and even more important: most people use mobiles, now. Most people, calling from a mobile, have to pay more to call a “free” 0800 number than a normal geographical number.

The only people who are going to care about calling you for free are the people who are going to carefully weigh up the benefits of every single penny they spend. The very people, in other words, who are least likely to bring in money once they’ve phoned you. The very people who are going to query every single item on the bill. So who still buys an 0800 number?

Maps are wonderful, lovely artefacts. I love to spread one out and read it like a book, analysing every square. Nowadays, though, I only do it for pleasure. Because, for practical reasons, if I want to plan a route or look somewhere up, it’s usually much much easier to go online for it.

There are downsides to this. Google Maps are nowhere near as good as a paper map. Their cartography just isn’t up to the same standard. They include roads, railways, rivers, and that’s about all. No buildings, no landmarks, no landscape. In Britain, Google Maps have the slightly odd habit of only including railway lines with passenger services,* and

Google, though, are nowhere near as low-quality as Yahoo Maps. Since I bought a full Flickr account, I’ve used Yahoo Maps a lot, to record where I take my photos; and there are just so many places where Yahoo’s map doesn’t match reality. Take, for example, somewhere I visited recently with my camera: Battersby, on the edge of the North York Moors national park. This is Google:

For comparison, this is the Yahoo version at the same scale:

Never mind the lack of street names, and the general lack of contrast which makes it difficult to see where the roads run, especially within the National Park. Where exactly does that railway run? Where is the park boundary? The park boundary does indeed follow the line of the railway; but what shape is it? There’s a big difference there, just because Yahoo’s maps don’t include enough detail, apart from for roads, to be at all accurate when zoomed in. Rivers are just as bad, and apparently have zero width too.

If you really wanted to know the answer to the park boundary question, incidentally: it’s Google that’s right, as you can see by looking at Streetmap, who license the Ordnance Survey’s maps. Now that’s what a map is supposed to look like.

* It probably derives from the Ordnance Survey’s long-standing division between “railways” and “freight lines, sidings or tramways”, which dates right back to the start of the One-Inch series. It was bad enough when Landrangers, at I think the Second Edition, dropped the distinction between single-track and multiple-track railways. I have some First and Second Edition Landrangers somewhere, so I’ll have to check when the single-track railway symbol disappeared. First Edition were the last One-Inch maps photographically enlarged, which leads to some odd discrepancies on them;** the Second Edition were redrawn.

** for example, on the First Edition Sheffield and Huddersfield Landranger map (sheet 110), a chunk towards the south of the map is lettered in a different, older, font, which suggests that part of the map was derived from a rather earlier original than the rest of the series. I’ll scan a section some time to show you.

Keyword noise: maps, cartography

In the summer, we had big floods up here, worse floods than anyone in this village could remember. It was, apparently, a once in fifty years event.

Now: we’ve got floods again, six months later. Maybe not a once in fifty years event, true, but let’s say (for the sake of argument) that this is a once-in-twenty-five year flood.

Maths time: in any 6 months, your chance of having a 1-in-50 year flood is 1/100. 1/50 for the more likely 25-year flood. The chance of having both, though, is those numbers multiplied together. 1 in 5000. Which doesn’t, at face value, look like a particularly big number; but that’s because we’re not great at judging magnitude. Something that has that chance of happening within 6 months should, on average, have happened once in the last 2,500 years. That’s once, since the start of the Iron Age.*

The problem with probability, though, is that you can’t say: this will definitely only happen once. It could happen three times within a week,** and still be within the bounds of probability. It could still happen, within the rules of our simple model; it is just highly unlikely to happen. If it does, you’ve just seen something amazing, but it doesn’t necessarily mean that your starting figures are off. On the other hand: if something happens that, according to your figures, is highly unlikely, it does make more sense sometimes to decide that the numbers you’re basing your statistics on are out of date. Suddenly, big floods aren’t rare any more.

* slightly more than once, to be honest, because the Iron Age started about 2,700 years ago.

** Hull was flooded twice, 14 days apart, in summer 2007. Some of the floodwaters in unimportant places, such as verges and parks, still hadn’t drained from the first flood when the second (and worse) flood came. That, though, means that normal “multiply the two numbers together” probabilities don’t work. The two floods weren’t independent of each other, because of all that water lying about, so the probability of the second was rather lower than it would have been.

Today’s blog is like one of those spot-the-difference puzzles where you have to spot hard-to-find differences between two apparently identical pictures. To make it a little bit different, though: here’s a carefully-prepared Spot The Non-Difference puzzle, where (for a change) you have to spot the hard-to-find connection between two apparently little-related pictures.

Firstly, we have a photo I first spotted in today’s Guardian. It’s a publicity still from the award-winning film Atonement, and shows James McAvoy hard at work apparently invading war-torn France:

Secondly, this photo, taken by Dimitra, some years ago now:

Yes, I’m pretty sure they were taken at almost the same location, although, to be quite honest, if I didn’t already know that Atonement was filmed in the Symbolic Forest area, I’m not sure I would have spotted the link between them.

Keyword noise: film, Atonement, location, Cleethorpes, Lincolnshire, beach

It’s nearly Yuletide, and all shall rejoice. For Yuletide means: the King William’s College General Knowledge Paper. Hurrah!

If you’ve never seen it before: the General Knowledge Paper is both an exam paper, and one of the hardest general knowledge quizzes around. Its questions are succinct, cryptic, and intriguing, and range over huge areas of knowledge.* On a quick run through it today, I reckon I scored about 32 points out of 360;** doing particularly well on London and Russians. Answers probably include I. P. Pavlov, Martin Chuzzlewit, Greyfriars School and Waterloo - unless I’m deliberately trying to confuse you.

* so much so that my friend K claims it isn’t a general knowledge quiz at all, because the answers are that obscure.

** There are 180 questions; you score 2 points per answer.

Keyword noise: quiz, trivia, general knowledge, King William’s College