Everybody and everything, it seems, is against weight these days. The Hummer is being replaced by the Prius. People are even writing weight-restriction clauses into their pre-nuptial agreements. You can't pull up a news site these days without reading something about solving the childhood obesity "epidemic."
Heck, even my poor 15-year-old dog is on a diet.
Smaller in technology isn't exactly new, but with two emerging technologies, computers could be headed for a major crash diet.
First, there's IBM's recent announcement on research that shows pulses of light can replace electricity as a means of transferring data between processor cores. This method would not only make today's supercomputers 100 times faster, it would make them way smaller -- say, the size of a laptop.
The technology is called silicon nanophotonics, and it's based on optical fiber. IBM research scientist Will Green is quoted in InfoWorld as saying this breakthrough means IBM could put hundreds or thousands of cores on a chip. And then he adds that this will mean a better gaming experience, according to InfoWorld. That just feeds into my paranoid belief that all technology research is about building a better video game - even if the research is funded by military organizations, such as Defense Advanced Research Projects Agency (DARPA) -- which, by the way, this was.
Of course, as diets go, this is the King of Slow Weight Loss. Researchers say it could be 10 to 12 years before the technology will be able to pack the power of today's supercomputer into something you can easily carry into Starbucks.
Unfortunately, we may not be able to wait that long, given our current rate of software bloat. Fortunately, there's a diet of sorts for software, too: The Department of Computer Science at the University of Leicester is working with ATX Software -- code name: Leg2Net team -- to reduce software bloat.
We've all complained about this every time Microsoft releases a new version of Windows, right? But the truth is, Microsoft gets a bad rap for something almost all software companies do. And to some extent, as this piece explains, it's inevitable that software becomes too "fat," and less agile, given software is patched and features are added. This article from Phsyorg does a good job of explaining how these practices quickly create agility problems for software.
Unfortunately, it doesn't do a good job of explaining how the Leg2Net team will fix it. Apparently, the research is still in the early stages, but they hope to find a way to cut the fat through careful analysis of the code. The method will separate the code into chunks and, hopefully, allow a specialist to trim the fat -- just like a surgeon.
The Leg2Net team hopes to find a way to re-engineer legacy software for use in a service-oriented architecture.