During the slow-news week between Christmas and New Year's, it becomes astoundingly clear that humans are good at predicting the future of things over which we have no influence -- movement of the planets, behavior of particles we can't see even with our most powerful instruments, the inevitable decay of order into the universe to a state of decoherent chaos and nothingness.
We're incredibly bad at predicting anything that has anything to do directly with us, however, especially when the predictions depend on how technology will develop and what we'll do about it.
In 1950, "Miracles You'll See in the Next 50 Years," included the ability to recycle underwear into candy and the assumption that most women would be housewives.
It also assumed it would be a good way to save the little lady some effort by making dishes out of "cheap plastic that would melt in hot water" so she could just let those carcinogens and long-chain polymers run down the drain and almost unimpeded out into the environment and water supply.
The predictions IBM releases every year are more substantial; they're all technologies IBM is developing in the lab and expects to be able to build into one specialty product or another before long.
Star Wars video chat
Its big news -- the sexiest thing in IBM's list of five things that will be real and practical technologies in five years -- is the three-dimensional hologram that will become the interface of choice for software and video chat.
Sometimes the "futuristic" predictions seem a little retro, like 3D holographic phones that will let us communicate the same way they did in Star Wars -- badly, with a lot of static and horrible catastrophes following immediately after the message.
Light, convenient power
Way more practical -- or at least desirable to those of us (all of us) who use portable computing devices -- is a design for batteries built around "energy-dense metal" (IBM didn't say which one) that would produce steady, long-lasting power for 10 times as long as current batteries when exposed to ambient air.
That detail-free teaser includes a promise that IBM is reducing the amount of energy each transistor uses to reduce the energy demand of small devices, to the point they could power themselves using the kinetic motion of the person carrying them, just as some wristwatches do now.
Kinetic energy is a possibility; power-saving transistors are great.
"Energy-dense metal" that generates power combined with room air? Sounds like sodium, lithium or one of the other unstable metals that do things like catch fire or explode when exposed to oxygen, water or other things found in ambient air.
I won't say it wouldn't ever be safe, but I'm not betting the TSA is going to let you on a plane carrying one any time soon.
Personalized commute, conserving heat
If the first three were a little mundane, the last two are mopwater dull -- though more likely to have an actual impact on consumers.
The first is a "personalized commute" that is actually the result of a big-system traffic analysis network that looks at the flow of commuters on all the roads, railroads and public transportation available, linking with adaptive traffic systems that can divert commuters from one route to another in real time to keep the roads moving.
The personalization comes in when you plug in the start and finish location and time of your commute; the IBM system would give you a reliable prediction of the best way to get to or from work, and give you real-time updates as you travel.
The last is a waste-heat-reclamation project aimed at data-center servers rather than the engines and brakes of Priuses.
Rather than spend 40 percent to 50 percent of the energy bill for your data center, servers would come with innovative cooling systems that run heat sinks to within microns of the processor to pull away heat and re-route it into the heating system for the building or city.
That would be cool from a tech standpoint, but you could manage the heat that way now with better ducting and airflow, which is a notorious weak point in most data centers.
Useful, but not sexy, and not the kind of technology that could be implemented in just one place, or by just one company. Often they're hard to even track.
You can always track how fast a new technology is developing by reading the research papers on it. Systems like adaptive traffic or municipal heat-capture from data centers depend on a lot of committee non-decisions and changes to facilities or utilities.
They're a lot more 'We'll see.' than 'Look at that!' And when it takes years for them to develop, that turns quickly into 'Who cares.'