I’m a great admirer of MC Escher. There are numerous possible practical applications of his idea of single-shape tesselations.

The simplest and most obvious is to create a cookie cutter. Not only does this eliminate waste, but if you make one tray of dark and and one of light chocolate brownie, then you can combine tiles from each to create a pleasing (and highly calorific) pattern. Yum.

More on applications of Escher-like patterns later…

It’s a real pain to walk with a small child and it can sometimes also be dangerous to negotiate crowded pavements and road crossings. Rather than use the reins which used to be popular (but which seem somewhat punitive/agricultural) here’s an alternative.

A light, nylon harness clips around the child’s body. An extension strap runs down each of the child’s sleeves. This whole thing is cheap enough to be left inside the child’s coat and doesn’t add to the normal level of difficulty getting them dressed to go out.

When walking, the extension straps each have a plastic clip at the end which attaches to a corresponding attachment on a padded wristband worn by the parent. This reduces the required grip and arm tension for the child, when they are being led by the hand.  The clips can be operated by one hand, but not by the child.

This mechanism helps in getting them to walk in the right direction without bending down every five seconds to pick up some gravel or the ubiquitous dog excrement. It’s also strong enough for a parent to arrest a child before they topple over onto the ground or walk out in front of  a car.  In crowded places, I’d suggest that the parent also wear a thigh strap (with clip buckle) so that they can temporarily tether the youngster to them, whilst using both hands to eg fish for keys or cash.

Still in festive mood, I thought today’s invention should take the form of a new game.

A web page allows input of a search term by one player. It then displays the images which Google image search has found for that term (but only the images, without any text, URL’s or other information attached –Windows Live search displays the extra information only when the images are rolled over, but for some reason the images seem somehow less compelling).

The other players then have to view the returned images and the first to correctly come up with the search term, wins.

This is relatively easy for concrete nouns (eg ‘bicycle’) but unbelieveably difficult for abstract terms (eg ‘nature’).

One nice feature is that the game can of course be adapted to be played by any number of remotely located people.

You don’t have to be any kind of conspiracy theorist to recognise that powerful people have long used doctored images for a variety of purposes from selling papers to selling ideology (can you spot the difference?).

To make this very much more difficult, I propose using 3-D movies.

It’s well known to children reading a comic that when attempting to detect the almost imperceptible differences between two ‘alike’ images, all you have to do is ‘fuse’ these’ by crossing one’s eyes a little and visually superimposing them. Areas of disparity then all stand out simultaneously as twinkling regions.

Two small moviecameras mounted at interocular distance (~5cm) apart would be used to film every event. The two movies could later be superimposed by the viewer, possibly using a simple stereoscope, to see the scene in a single-viewpoint form of 3-D (although viewing one of the two movies in the conventional way would still be possible, of course).

These days, of course, Hollywood technologists can scan an actor’s face and insert it into a (2-D) movie so effectively that it looks as if he’s doing his own stunts. Without having scanned the objects or people in question, however, this insertion is pretty much impossible to fake convincingly in a sequence of stereo images: the twinkling effect is obvious. Thus the imagery is much more difficult to fake or manipulate.

The oddly-named “More Than” insurance company has done some ‘research’ in which they claim that one in five drivers admit to concentrating behind the wheel less than 75% of the time (mostly people answering this questionnaire said they thought about sex; hardly a surprising result but not very reassuring when you are nose-to-tail at 70mph).

Other studies have shown that driving skills are at their best when in a ‘flow-state’, so it’s not obvious that hypervigilance avoids accidents anyway.

Having said that, driving with your hands on the wheel at the ten-to-two position is apparently reliably indicative of safe, alert driving.

Today’s invention is therefore a simple alert message which is issued when the contact pattern of the hands on the steering wheel differs from the correct one for more than a second or two -a ‘dead-man’s handle’ for the modern era. This could be achieved by simple template matching of a pattern on a touchpad embedded in the wheel surface with one stored by the driver on taking possession of the vehicle. It would still allow the occasional sleight of hand required for a rapid three-point turn.

A more elaborate system could detect whether a driver was failing to undertake the recommended ‘mirror-signal-manoeuvre’ procedure. A face detecting camera could assess the orientation of the driver’s head in relation to the mirror and then confirm that the indicator was in operation before the wheel was turned. If the driver was persistently not undertaking the sequence safely, it might suggest that more training was required or that the driver was drunk (see also this article).