tgflynn

71 ms·

751.
by tgflynn 11y ago
I may be wrong but I was under the impression that any gas molecule with more than 2 atoms contributes to the greenhouse effect. I don't think there's currently much ammonia in the Earth's atmosphere but if we started produc
752.
by tgflynn 11y ago
I really don't understand what you are trying to say here. By "made of 2 co-primes" do you mean "a product of 2 co-primes" ? If you're trying to define a new concept here called a "mirror point number&quot
753.
by tgflynn 11y ago
Dispatch definitely looks farther along but are they even targeting food delivery ? I just saw a techcrunch article about them that doesn't mention food but packages and mail. Food delivery may require a somewhat higher speed, more ro
754.
by tgflynn 11y ago
I think something like this is going to happen fairly soon. It seems like speed will be an issue if this is limited to sidewalks. Maybe what's needed is something partway between a car and say a bicycle. I think it needs to move fast
755.
by tgflynn 11y ago
I think it would be great if there were a way to mouse-over a word and get a popup with the word's translation and information like part of speech, conjugations, declensions, etc.
756.
by tgflynn 11y ago
This seems like it would be a great idea if it can be done because the market for it would probably be very large. However it seems to me that doing this would be very difficult for at least 2 main reasons. First spreadsheets can be organiz
757.
by tgflynn 11y ago
To understand ML you'll at least need a good understanding of vector differential calculus and linear algebra. There are many free ressources available for learning math today, MOOC's, free textbooks, etc. but if you were studyin
758.
by tgflynn 11y ago
I agree that correctness is usually most important but even that's relative because with the possible exception of some life critical software virtually all software has bugs and a certain bug density is essentially tolerated in the so
759.
by tgflynn 11y ago
There's no way to answer that question in general. It all depends on how the software is going to be used, the goals of those who are developing it, the environment in which it is being developed and probably many other factors.
760.
by tgflynn 11y ago
Wouldn't that only be a problem with open loop control though ? If you have an encoder and use feedback would backlash still be an issue ?
761.
by tgflynn 11y ago
How computationally expensive ? Are we talking supercomputer time to simulate the few seconds it takes to grab an object ? Advanced robots are expensive too and usually much harder to get access to then computational ressources (ie. AWS).
762.
by tgflynn 11y ago
That's a very interesting question. My guess is that the physics of grabbing things, especially non-rigid things, is very messy and difficult to simulate. It would be great if someone here were able to give a detailed answer to this
763.
by tgflynn 11y ago
Ah OK, I hadn't thought of potential energy in permanent storage.
764.
by tgflynn 11y ago
I'm pretty sure that's not correct. If it were why would research labs have to pay for things like Matlab ?
765.
by tgflynn 11y ago
Heat pumps aren't usually called electric heaters though. Electric heaters are usually considered to be devices that generate heat by consuming electricity, not devices that transfer heat. In any case a heat pump obviously falls into
766.
by tgflynn 11y ago
Not sure why you say "very, very close". The efficiency for heating of all electrical devices is identical, unless they're somehow transmitting energy to a different location.
767.
by tgflynn 11y ago
The value seems to have depreciated even faster than for consumer hardware. I think you would expect to pay a bit more than $2 for a working laptop that sold for $2k in 2004.
768.
by tgflynn 11y ago
For learning purposes being able to run zOS would probably be of considerable interest. If the machine actually requires 6 kW to run then I guess the electricity costs would be on the order of $1 per hour, depending on your local rates. M
769.
by tgflynn 11y ago
Sure I get that but if you can't run the software that was/is most frequently used on these machines then you're only preserving half the history.
770.
by tgflynn 11y ago
In that case I'm not sure I see the point in buying something like this just to run Linux. Linux looks pretty much the same no matter what it's running on.
771.
by tgflynn 11y ago
Does buying one of these on eBay get you the right to run zOS ? If so that could be a lower cost alternative for people interested in learning about mainframes, since if you had the usage rights you could run zOS on Hercules and just store
772.
Doritex and web provider fined for security breach(bizjournals.com)
2 points by tgflynn 11y ago|0 comments
773.
by tgflynn 11y ago
I had this problem when I was an undergraduate. I think that if you have the energy and intellectual capacity to follow these various interests without it adversely affecting your grades or future prospects (I didn't) then it's no
774.
by tgflynn 11y ago
Could you explain a bit more about how you "visualize numbers" ? You say you see them along a path. How exactly does the path represent the number ? Is it the length, or the number of segments ?
775.
by tgflynn 11y ago
Is the contrast on an old CRT really better than on a modern LCD screen ? I find that a bit hard to believe but I certainly don't have any way to check. You could run a plain vanilla window manager with a single xterm if you want to
776.
by tgflynn 11y ago
Do you know of a way to measure "intrinsic dimensionality" ?
777.
by tgflynn 11y ago
N/e^D would be vanishingly small for most problems on which deep learning is used. Image recognition for example may involve more than 1000 pixels and e^1000 is a number that makes the number of the atoms in the universe look tiny.
778.
by tgflynn 11y ago
Nor do we have any proof of a lower bound > O( N ) for 3-SAT.
779.
by tgflynn 11y ago
For your second objection by "efficient algorithm", I meant an algorithm that is practically efficient on very large problems. That probably means no worse than O(N^2), but preferably more like O( N*log(N) ). For your first objec
780.
by tgflynn 11y ago
Yeah, that's the thing about this problem, virtually infinite payoff if you succeed and virtually zero chance of success. The human mind isn't very good at dealing with that kind of cost-benefit ratio, it can drive people a littl

More ›