41–49 of 49 posts
Awesome. It'll sure be nifty when 2029 rolls around and the patent expires, so applications can actually use the algorithm. (No, I don't have any information that an sFFT patent has been filed, but this would be standard practice at MIT). Tornado/fountain codes are a similar case. Pardon my bitterness, but it's an interesting question to wonder whether, by funding researchers to invent algorithms of this type and loc…
Earlier quoted context omitted.
Algorithms and mathematical methods are not supposed to be patentable in the US.
As a famous American statesman once put it: "Justice Marshall has made his decision - now let him enforce it." It's commonly supposed that the Supreme Court could void these illegal patents - could dissolve this entire illegal industry - with a swish of their shiny polyester robes. Actually they already tried that, in Flook . Jedi mind tricks, which in the end is all the Court really has, worked about as well on Andr…
A large fraction of the native population that was rounded up and marched in that incident died. Accurate records were not kept, but according to estimates about a quarter of those evicted died as a direct result. Including some of my relatives.
they make this seem like a huge deal, anyone know why? is this going to like put HD content on an iPhone? I don't expect that the signal processing hardware is the bottleneck here.
Earlier quoted context omitted.
> I don't expect that the signal processing hardware is the bottleneck here. Computational resources is always the bottleneck in bioinformatics, quantum chemistry, or any sort of high data volume analysis or simulation, and the FFT is a fundamental and commonly used transform in all fields. At least for people who still use computers to, well, compute things, a faster FFT is a huge deal.
How many domains still use FFT specifically though as opposed to some other transform? Most of the signal processing papers of the last decade that I've read or read about have used either wavelets, FFT on a small window such that this isn't really applicable, or some arbitrary non-orthogonal basis.
Awesome. It'll sure be nifty when 2029 rolls around and the patent expires, so applications can actually use the algorithm. (No, I don't have any information that an sFFT patent has been filed, but this would be standard practice at MIT). Tornado/fountain codes are a similar case. Pardon my bitterness, but it's an interesting question to wonder whether, by funding researchers to invent algorithms of this type and loc…
Algorithms and mathematical methods are not supposed to be patentable in the US.
Earlier quoted context omitted.
As a famous American statesman once put it: "Justice Marshall has made his decision - now let him enforce it." It's commonly supposed that the Supreme Court could void these illegal patents - could dissolve this entire illegal industry - with a swish of their shiny polyester robes. Actually they already tried that, in Flook . Jedi mind tricks, which in the end is all the Court really has, worked about as well on Andr…
Please choose less offensive examples. A large fraction of the native population that was rounded up and marched in that incident died. Accurate records were not kept, but according to estimates about a quarter of those evicted died as a direct result. Including some of my relatives.
Some of them died justly, others not. I'm not an Albanian and have no interest in investing emotional energy in multi-generational blood feuds. Eg, some of my relatives were killed by the Germans, who make a really fine motorcycle.
Also, life was pretty cheap before the 20th century. It's easy to appall kids these days with the death rate among chained slaves crossing the Atlantic. In fact the death rate among Europeans voluntarily crossing the Atlantic was quite similar. It also wasn't easy or safe to walk across half of North America, whether voluntarily or not. Basically, to anyone who lived in this era - regardless of race, color or creed - we're all a bunch of whining pussies.
Really the main difference between the expulsion of the Cherokees and my proposed ethnic cleansing of the Palo Alto patent lawyers is that the Cherokees didn't deserve it, whereas the patent lawyers do. Lawless force can work good as well as evil. And when we've already left law behind, what's the alternative?
Earlier quoted context omitted.
...where k is the number of FFT coefficients computed, and n is the time series length. Has anyone else been turned off by the sensationalism of MIT TR articles?
With all due respect, this is anything but sensationalism. It might help to understand that the number of FFT coefficients computed is not necessarily the number of data points processed. Theoretically, if we have a perfect cosine, for example, we'd get TWO non-zero coefficients (k=2) even if n was a huge number. Compare O(k logn log(n/k)) to O(nlogn) in those cases! Granted, this doesn't apply to every single situat…
The big-O results of the paper are interesting and with more work could prove a starting point for improvements in very long FFTs. Why long? Because, for big-O asymptotics to kick in, you'll need large n.
But, the domain of applicability seems to be a niche. The "7 non-negligible coefficients" result you refer to is for 8x8 DCT's (as in JPEG). That's n = 8. There are optimized implementations for the small-n cases that will utterly dominate a big-O optimal algorithm like this. Compare Strassen's algorithm for matrix *.
Have you looked at the code complexity of the algorithm in the paper? There are several interlocking components (hash functions, rebinning) that have their own big-O optimality claims and, in some cases, randomization. Getting it all to work together efficiently, given cache issues, even for large n (say, 10^6) would be a serious engineering challenge. People have been working on optimizing data flow for FFT for decades now, and the state of the art is very advanced.
I agree, that the result is interesting in principle. Runtime is actually sub-linear in n, the sequence length.
I just had a somewhat closer look at the MIT Technology review list of emerging technologies, and in particular the people behind them. I made a rather sad discovery. Please take a look at the following list: - Jonathan Tilly (stem cell research) - John A. Rogers, Ralph Nuzzo, George M. Whitesides, Etienne Menard (Semprius founders) - Ren Ng (light field photography, Lytro founder) - Nikhil Jaisinghani, Brian Shaad (…
Earlier quoted context omitted.
Algorithms and mathematical methods are not supposed to be patentable in the US.
As a famous American statesman once put it: "Justice Marshall has made his decision - now let him enforce it." It's commonly supposed that the Supreme Court could void these illegal patents - could dissolve this entire illegal industry - with a swish of their shiny polyester robes. Actually they already tried that, in Flook . Jedi mind tricks, which in the end is all the Court really has, worked about as well on Andr…
http://www.loper-os.org/?p=309
Transparently offload the "patent sin" to a system's users. (With plausible deniability.)