Keyboardless keyboard

Sometimes we need to interface with a computer without having to rely on a physical keyboard. For instance, most combat situations would preclude the use of a keyboard, as would working on power lines, or when your hands are covered in oil, etc...so, this idea, whether it already exists or not, is for those and similar situations, both current and future.

This seems problematic

Hardware
Make a glove with pressure sensors sewn into the finger tips that are "spatially aware" of one another. When activated, the sensors send wireless signals to a computer, smartphone, or HUD device via bluetooth, where each signal tells the device which finger was tapped and where that finger was, positionally, relative to all other fingers (or perhaps just relative to the thumbs, since they tend to stay in one place) when it was tapped.

Software
The software that utilizes the signals should be coded to take advantage of computer learning and pattern recognition (as well as traditional word prediction and auto-correction) so that each word in the dictionary can be turned into an approximate pattern of pressure and position signals from relevant fingers. The pattern recognition software should be coded so that it can adapt to the nuances and quirks of individuals. In this way, the longer a person uses the software, the more useful it becomes for that individual. This computer learning and bio-feedback data should be stored in separate files from the base software, and made transferable, so that if the glove is damaged, the individual can transfer the files to a new set of gloves without having to start from scratch. The glove and the software should not activate until needed, perhaps by the push of a button or by bringing the thumbs to within one inch of each other, otherwise inputs would be made randomly while wearing the gloves.

Function
This system allows the user to treat any flat surface as a keyboard.

Perks
It could also operate as a mouse by holding the left hand motionless as a reference point for the right hand. The fore and middle fingers could then be tapped as left and right mouse buttons.
This system could be built into preexisting military gloves, work gloves, or consumer gloves.

Prepaid gasoline

If you're running a gas station, this idea's for you.

Build a website that updates your advertised price of gasoline in near real time. Allow customers to prepay for a limited amount (up to say 50-80 gallons) of gasoline, redeemable within a limited amount of time (up to say 60-180 days).

Customers will watch your site to catch a good price, which allows you to see internet traffic, advertise, show deals, promote your brand etc. Customers will prepay for gasoline when the price is right, giving you an upfront payment, while allowing you to undercut the local competition. If they're buying from you, they're not buying from them. This does mean that you're not selling as much gas at peak prices, but it also means that neither are your competitors. You can support lower margins by moving more product, and supplementing the loss with advertising, brand loyalty, and sales of other products.

Walmart shopper

Build a website that allows a customer to enter a grocery list, a local address, and a preferred delivery time. Access this list with a smartphone, do the shopping for the customer, charge a 10%-20% fee of the total grocery cost, and a $10-$20 delivery fee based on distance to the address.

So, if you charged 15% and a $15 delivery fee, you could knock out the grocery run in about half an hour to an hour, depending on your shopping savvy and traffic, which means you could easily make $30/hour. That's $60,000 a year.

This number grows rapidly as you get better and/or hire out the shopping and delivery.

Smart traffic lights

Why do traffic lights have to suck? If we can afford to place a motion sensor camera on every toilet from L.A. to New York, why can't we do the same for traffic lights? How hard would it be to place a camera sensor with a special chip that could analyze traffic in real time? I'm talking about predicts-the-best-solution-as-traffic-approaches type of software. Of course, this idea isn't new. Someone code up that algorithm and end the days of waiting 5 minutes at a traffic light for no discernible reason.

Every second counts as far as I'm concerned. The more people can move efficiently and smoothly around the country, the fewer accidents we'll have, the more productivity will increase, the less gas will be wasted, the less road rage we'll feel, the less pollution we'll breathe...I mean it's all small change, but it adds up pretty quickly when implemented on a national or global scale.

And it might just keep me from going insane.