A restaurant chain can collect data on what food people like by examining the plates collected from the tables – the more leftovers given the size of the dish, the less popular the food. However, looking at the plates and entering the data takes time. It would be much faster to automate the process. For example, there could be a small conveyor belt for dirty dishes brought back from the eating area. The dishes would be weighed to record the amount of leftovers before scraping and washing. To detect which food was left over, one option is that a camera above the belt photographs the leftovers and then a computer tries to identify the food. This is a complicated machine vision and machine learning problem. A simpler option is to serve different dishes on plates with different shapes, or patterns such as lines and circles that are easily distinguished by computer. Then the plate identifies the dish for the camera, similarly to colour-coded plates identifying the price at sushi-train restaurants.
Even less costly in terms of computation (and without any camera requirement) would be to put RFID tags or other remote-id technology in plates. Each dish would have to be served on a plate with a dish-specific RFID, so the returned plates can be exactly matched to the food served on them. Each plate becomes more costly, but not by much, because RFID tags are cheap.
A single restaurant could also collect data on leftovers, but a chain of restaurants would get a larger dataset faster, thus useful information sooner on which dishes to keep and which to discontinue.
The accelerometers in phones can detect vibrations, such as when the car that the phone is in drives through a pothole. The GPS in the phone can detect the location and speed of the car. An app that connects the jolt, location and speed (and detects whether the phone is in a moving car based on its past speed and location) can automatically measure the quality of the road. The resulting data can be automatically uploaded to a database to create an almost real-time map of road quality. The same detection and reporting would work for bike paths.
Perhaps such an app has already been created, but if not, then it would complement map software nicely. Drivers and cyclists are interested in the quality of the roads as well as the route, time and distance of getting to the destination. Map software already provides congestion data and takes traffic density into account when predicting arrival time at a destination. Road quality data would help drivers select routes to minimise damage to vehicles (and the resulting maintenance cost) and to sensitive cargo. This would be useful to trucking and delivery companies, and ambulances.
A less direct use of data on road quality collected by the app is in evaluating the level of local public services provided (one aspect of the quality of local government). Municipalities with the same climate, soil and traffic density with worse roads are probably less well run. For developing countries where data on governance quality and spending is difficult to get, road quality may be a useful proxy. The public services are correlated with the wealth of a region, so road quality is also a proxy for poverty.
Some of the downsides of automatic transmission in cars are that it does not anticipate hills or overtaking, and does not respond to slippery conditions appropriately. The technology that could enable the transmission to anticipate hills or overtaking is already available and incorporated in some cars, namely GPS, maps and sensors that look ahead of the car. If the map data includes altitude, then the location and movement direction of the car on the map predicts the slope that the car will be on in the near future. This information could be sent to the automatic transmission to enable it to shift gears in anticipation of a hill. A forward-looking sensor that has a range of a few hundred metres can also see a hill if the road does not curve too much. The sensor data could also be sent to the transmission. Similarly, a sensor could detect the nearing of the car in front and shift to a lower gear to prepare to accelerate for overtaking.
Slippery conditions can be predicted using the car’s thermometer, perhaps with the addition of a humidity sensor, or detected using a wheel slip sensor. This information could also be sent to the computer controlling the automatic transmission, to prevent it from spinning the wheels too fast when there is little grip. The GPS or forward-looking sensor could also tell whether the car is moving relative to the landscape. Comparing the movement data with the wheel spinning speed reveals whether the wheels are slipping.
Empty housing is wasteful from society’s point of view. Both landlords and renters would benefit from finding a suitable counterparty to contract with faster. There are already online systems for listing housing for rent and sale, and also notice boards for people seeking housing. This is a good start, but a predictive system would be better. Given enough data, computers could forecast who is a good tenant or landlord and which apartment or house suits a given person’s preferences. Less searching would be needed by all involved.
Rental agencies already have a tenant database where they exchange references for renters. A similar online system should be created for landlords and housing (distinguishing the two). Also, the rental agency or real estate bureau should be rated separately from the people working in it, otherwise bad agents may move from one employer to another and escape their reputation. A bad notoriety may even motivate a person to change their name. For good agents, the loss of a reputation not tied to their person may make it difficult to change jobs.
Instead of chancing on complaints or praise in forums, a renter could see a summary rating of many rental agencies, agents and buildings in one place. The building database should include objective measures like the distance of a building to the city centre and the nearest supermarket, the yearly electricity and heating bills, the outdoors noise level in decibels, some average air pollution measure, school catchment areas, floor plan and area, etc. This saves labour for prospective tenants, so each of them does not have to search for the same data from various sources. Information entered by past renters is hopefully objective and protects novice tenants like students from being misled by advertisements like “five minute drive to the city centre” (only at 3 am when the roads are empty, in a Formula 1 car), “short walk to the supermarket” (short compared to the Shackleton Solo expedition), “safe neighbourhood” (compared to a war zone), “quiet” (relative to a rock concert), “spacious” (roomier than a shoebox), “close to nature” (insects and rodents inside). Distances to various landmarks could be automatically downloaded from Google Maps when the building address is known. Crime, pollution and traffic density statistics could similarly be autocompleted.
Renters should be able to select the measures they consider important in the data and get a ranking of the housing on offer according to these. Once someone has rated several apartments, the system could potentially predict the housing that would please that person.
If a text has already been translated to a couple of languages with high quality, then it may be possible to improve the quality of machine translation to another language by translating separately from each original language and averaging in some sense. I do not know whether a program currently exists that is able to take into account multiple starting languages – Google Translate and other online automatic translation services I have seen only use one. Several different translations should contain more information than one, so by comparing them, some errors may be eliminated. At least inconsistencies can be discovered by computer and then checked by a human, saving labour.
Facebook makes it easy to remember people’s birthdays, it just displays an automatic reminder. Other calendar programs like Outlook can also be made to do that. Every time someone receives a reminder of an acquaintance’s birthday, they send a birthday greeting – it is almost automatic. So why not make it fully automatic by writing a program to check Facebook every day and send a happy birthday message to anyone whose birthday is on that day?
The person receiving a birthday greeting usually replies with a thank-you note, which is also a repetitive action on a computer and can therefore be automated. Continuing this way, Facebook conversations can be made fully automatic without any human input whatsoever, apart from the initial writing of programs. But Facebook accounts could come with these programs built in, so anyone creating an account will automatically start participating in these computerized conversations. This takes the idea of virtual friendships to its logical limit.
The same virtual conversations can be created using other email and calendar programs – if the calendar displays someone’s birthday, an automatic email is sent with a greeting, and the recipient’s email program sends an automated reply.