Bike ejector

Locking your bike when you only need to leave it unattended for a couple of seconds is a pain. A better way: a device that sits on your front forks and jams the spokes if someone tries to cycle away (once they've got up a bit of speed), ejecting them from the bike.

There are a couple of ways of implementing this. A mechanical, springloaded rod that is entered between the spokes. Some kind of mechanical counter is used to ensure the villainous cyclist has got enough speed (indicated by rotations) before ejection. Or a remote controlled device (again with a springloaded rod that is entered between the spokes. This allows for more user control, and prevents injury of the owner of the device forgets to deactivate it before getting back on.

[Note: the post is intended for humor only - such a device could cause serious injury and should not be tried at home]

Finding new words

It should be relatively easy with crawl the world wide web comparing its contents to all established dictionaries (all languages, technical glossaries, and slang dictionaries), and hence finding examples of character strings that don't fit existing word definitions. And for each such string it should be possible to measure it's frequency.

Of course, there will be numerous spelling mistakes (and in fact data on common spelling mistakes is useful for both spelling correction, understanding the way people type, and the way people think), but it should be possible to filter these out based on similarity rules.

What would you want to do this? To answer the question "what is the most common undefined word?", but also to set about adding definitions for those words to relevant dictionaries to increase the extent of structured human knowledge.

Polytunnel robot

Some continued thinking on polytunnels: it is probably within the realm of existing technological capability to create a polytunnel building robot, opening up the possibility of putting vast swathes of desert under plastic potentially allowing moisture to be retained, and succession to occur.

It might also be possible for these robots to be solar-powered, and to be able to harness local materials (self-replication would be good too, but it's a ways off). Markus Kayser's SolarSinter, hints at the possibility of putting the desert under glass, using available sand.

Perhaps a polytunnel building robot could be the next project for the Global Village Construction Set. 

Analysing conversation

I am always intrigued by what people spend their time talking about (a quick google suggests we spend between 2 and 5 hrs a day talking). So how to get that data?

It would be an interesting experiment to place mics in bars, clubs, restaurants, shops, near watercoolers, etc to capture conversation, which could then be transcribed and analysed. Obviously the particular locations would need to disclose to people that the experiment was in progress, and steps would need to be taken to ensure privacy of data.

Alternatively, venues could offer subsidised drinks and food in return for this invasion of privacy, with marketing organisations paying the venues for the resulting intelligence gleaned. This could potentially be a win-win-win for consumer, venue owner and marketing organisation. There is also the potential for marketing organisations to identify people who are Connectors, Mavens and Salesmen, and follow-up with these individuals, for example offering them free samples.

An initial batch of questions:

• What topics are people talking about? (work, home life, friends, celebrities, TV shows, etc)
• What brands/companies/products are people talking about? What are they saying? What attitudes were conveyed by word choice and tone of voice?
• How do people get from one topic to another?
• What syntactic structures are common?

There are a couple of movements into the tracking consumers space, e.g. SceneTap and Shopperception, so perhaps this (potentially scary) future might become commonplace and not an experiment.

SMART objectives

The SMART criteria are a useful set of requirements for setting sensible objectives for organisations. However, when applied to employees I feel that there's something missing: controllable/influencable.

It is possible for an objective to be specific, measurable, attainable (by the organisation/team), relevant and timebound, but not be something that an individual has much control/influence over.

So I propose iSMART: influenceable; specific; measurable; attainable; relevant; timebound.

Optimising supermarkets

Until we get that robotic delivery service sorted, the widespread adoption of dropboxes and cheaper delivery costs, it's likely that we'll have to put up with the existence of supermarkets.

I suspect there is much to be done to optimise supermarkets; there are three key problems: finding goods (with subproblems of knowing whether goods are stocked, knowing whether the goods are in stock, and routing); aisle congestion; and checkout congestion.

Finding goods

The low-tech solution is better sign-posting, but there are so many types of goods in the average supermarket (let alone hypermarket) that detailed sign-posting would be counter-productive - it would be hard to find the right sign.

The high-tech solution is a digital store map, accessible by smartphone with voice recognition location search. The next step beyond this is a Sat Nav style display (or docked smartphone) on the shopping-cart that maps out a route in-store based on the items in a list (downloaded from online account).

Aisle congestion

A simple fix to aisle congestion would be narrower shopping-carts, as this would allow more shoppers to pass each other on the aisles. However, this could reduce the amount people purchase, making it an unattractive choice for supermarket owners.

Another possibility would be to monitor store traffic and aisle congestion, for example with pressure-sensitive floors or 3D imagers such as Kinect. Data on which aisles get congested could be used to rearrange stock. This should also mean that shoppers visit more aisles on average, which increases their exposure to products - something which would be attractive to supermarket owners.

Checkout congestion

There are numerous solutions to checkout congestion: self checkout; in-shopping-cart self-checkout; near-field-communication-based checkout; etc.

For classic manned checkouts, monitoring people entering the store, combined with knowledge of the average visit length, should allow supermarkets to predict and preempt checkout congestion (i.e. upmanning in advance of an anticipated period of congestion).

Overall busyness

A solution to overall busyness would be to publish store visit data, allowing customers to easily identify when the store would be quieter, and tailoring their shopping plans accordingly. This might benefit the supermarket in requiring less staff overall, as busy periods would be smoothed out.

Wall-mounted toothpaste dispenser

Why? To reduce waste by getting as much toothpaste out of the tube as possible, as easy as possible.

How? Wall-mounted holder for the toothpaste tube with rollers to squeeze the tube from the thin end.

Problems remaining? Getting out the toothpaste from near the cap.