Follow Slashdot blog updates by subscribing to our blog RSS feed


Forgot your password?
DEAL: For $25 - Add A Second Phone Number To Your Smartphone for life! Use promo code SLASHDOT25. Also, Slashdot's Facebook page has a chat bot now. Message it for stories and more. Check out the new SourceForge HTML5 Internet speed test! ×

Comment Re:Check the links (Score 1) 266

Actually, it's not (only?) a library bug, but a(lso) programmer bug. Using std::make_pair(x, x) makes a pair, duh! Complaining about it is silly. Hint: the initializer list version of insert() is faster than the pair version (at least on sane platforms, Microsoft can be weird about c++)

Comment Re:Many a young engineer.... (Score 4, Informative) 109

If only it were as simple as that. He's still right about one thing though, your initial statement about "conceptual debate between hole flow and electron flow" is misguided. It's just the reasoning that's ... inexact.

Both 'electron flow' and 'hole flow' are pseudo-particle descriptions of many-body transport phenomena. Heck, there are systems where the pseudo-electrons have anisotropic mass, charge/spin separation, and so on - hardly the behaviour of a free electron. Besides, that 'a free valence band' term you used is misleading - a vacancy is as ill-defined spatially as an extra electron in a strongly-interacting many-body system. 'Electron' and 'hole' flows both are the same concept - quasiparticle linearizations of otherwise (mathematically) intractable systems. So there is no 'conceptual debate', yet neither is a 'real boy^H particle'

Comment Re:Infinity (Score 1) 1067

It has no value. What you have here is a discontinuous function. The graph may appear continuous to the naked eye, but it is not. Just pretending that y = sin(0)/0 = 1 is not proper math.

That's an amusing, if not too common mistake - sinc(x) is indeed continuous at zero, no eye required. Do a Taylor expansion on sin(x) and you'll see that sinc(x) = 1 + O(x^2), which in this particular case is a series that converges everywhere on the real axis, including at x=0 (check the convergence radius using the usual methods, if you need to). But what do I know, I must have, how did you put it, 'relied on the power rule to get through Calculus 101.' Well, at least I got through it.

Comment Re:Infinity (Score 1) 1067

Thank you, if this is all you have to say in reply then I'll take it as a compliment. An argument would have been nice, for instance why exactly anyone attempting to calculate (exp(x)-1)/x at x=0 is, you know, wrong, but I wasn't holding my breath for it anyway - facts tend to be confusing wrt absolute statements. Well, if saying division by zero is 'wrong' floats your boat, more power to you. Feel free to declare irrational numbers wrong as well, since they cannot be represented using a finite-word binary CPU.

Comment Re:Infinity (Score 1) 1067

The fact that the result of division by zero is undefined is a fact of life, not just some made-up mathematical construct.

Looks like someone skipped Calculus 101.There are an infinite number of situations where division by zero is well defined, even 0/0. It all depends on how one approaches that point. To put it simply, if X/0 (with x finite or not) is a simple arithmetical calculation, you are right. If it's an algebraic one, more often than not the denominator and numerator are continuous functions and certain ways of sending one or both of them toward zero have well-defined limits. Sometimes those limits are all equal (think 1/x^2 for x -> 0, left and right limits are +Inf), sometimes not. In the first case division by zero is perfectly well defined if you compactify the real axis to include +-Inf. The same approach works (or not, if there's no well-defined limit) for expressions going to 0/0 - heck, it's how derivatives are calculated for differentiable functions, for one thing.

The common mistake that code monkeys do, which makes them state things like your affirmation, is confusing operations done by the CPU (arithmetics) with operations done in code (which come from algebraic expressions). The above should make it clear that division by zero is undefined for the CPU, but not necessarily so for code (or mathematics). Now, how you handle it in code so as not to trip the CPU is the actual problem, but it does not make it wrong (duh!)

Comment Re:Two questions need to be asked (Score 1) 546

Of course it wasn't worth it, because your privacy is far less important than your security.

Oh, ye of little minds. Go live inside a maximum security prison then, full security and zero privacy. How would you like that, then?

When your privacy is violated, you only worry about bad things that "might" happen.
When your security is violated, those bad things actually DO happen.

Either you are that simple, or you're a fearmonger. The second scenario would explain a troll modding, I suppose.

Like, right now, you actually think your privacy rights is more important than your competitive economic advantages you may have over Russian or China

I get it, those Foxconn workers' privacy has no value and the rooftop fences are for their security. (that was sarcasm, by the way). I do wonder what is your motive for arguing that a trajectory towards a concentration camp model increases the competitive economic advantage of a country.

We adults make fun of this sort of thing.

Indeed we do, although I suppose you and I mean different things by 'adults' and 'that sort of thing'.

Comment Re:It wasn't the tweet (Score 0) 185

And the fact that the poor Q1 numbers were again accompanied by outrageous executive compensation didn't help either. It must be really nice to be in mr. Noto's shoes, getting about $72.8 million dollars for half a year's work - well, presumably less now due to the stock grants part taking a plunge.

Seeing as during the last quarters executive compensation in the form of stock grants kept climbing at Twitter in spite of the company's financial performance, it hardly seems a good value investment, even with the raising tide lifting all boats nowadays.

Comment Re:Oh grow up (Score 2) 232

If some of you stopping look at every thing systemd tinted glasses you might start reacting like rational adults.

Funny you should say that, AC. I'll wait (not holding my breath though) for the rational, adult answer of G H-K to this message and a timeline for addressing the issues it raised:

To quote the last part:

Not that this complaint is not in any sense new you have been ignoring people who try to bring up meaningful issues for a long time. The fact that when people bring up uncomfortable points about the kdbus code they get routingely blown off certainly contributes to the lack of meaningful review as it is not rewarding to work with someone who does not listen to criticism. At this point the strongest possible language and the strongest possible push back are being used because everything else is routinely swept under the rug.

So, feel free to engage in that rational discussion anytime now.

Comment Re:No they don't (Score 1) 226

In the sections that don't overlap you basically have three items, more efficient visible light lasers (more than 50% socket to light, to be exact), lightweight space structures, orders of magnitude lower launch costs.

I was going to point out how your list is arbitrary (visible light lasers? try masers; and I take it launch costs were a hidden factor in that 'equation,' right?) and not even consistent with your blog post (should be larger, I'll put that to getting carried away by ... imagination) but then I read your next sentence.

Now throw darts at the wall. See how often you hit that section.

See the problem?

And yes, I do see the problem. To put it bluntly, you have no idea what you're talking about. Must be an interesting universe that you're living in, where research is driven by a dart-sampling-on-a-2D-surface process - and it's good that you picked a 2D surface, otherwise in higher dimensionality spaces you'd have had an awfully low research rate, which would make explaining the ability to post things on /. awkward.

And so, I apologize for wasting the time of any people who were interested in a meaningful discussion (a circle which apparently does not include the OP, maybe from lack of an adequate dart supply). And to the OP, I wish good luck in promoting his 'equation' and arguing that, as meaningless as it is, a factor of 2 or so of improvements - not even an order of magnitude - has not and cannot ever be overcome by technological progress. After all, focused research is a myth and it's only a matter of darts until, say, Intel's materials research will yield something entirely unrelated, like a cure for daltonism.

Comment Re:No they don't (Score 1) 226

You're making two assumptions that, I would submit, undermine your argument to some extent.

The first one is no technological progress - and with it, a corrolary that the only PV tech worth pursuing is the one working at ground level, protected by a thick atmosphere. I would expect a trade-off between conversion efficiency and lifetime to be the tech play to follow for space PV pannels - meaning different materials, different ideas that would perhaps not be economical at ground level. The constraints are different up there, and in the long run we will need some sort of space-based solar energy capture tech, as there's just too much to do off-planet once a suitable jump point is established to scale down costs. Heck, even lunar settlements will need this. Moreover, for a system with high enough fault tolerance/robustness/etc, the solar station does not even have to be placed in Earth's orbit - Earth's Lagrange points would make more sense, even if transmission will have to solve a somewhat thornier problem, so lower efficiency traded of for longer lifetime could be compensated by higher illumination. Further down the road, heliocentric orbits for multiple stations and/or multiple relays might turn out to be a must, especially if humankind ends up needing power all over the inner solar system.

The second assumption is that any solar power station has to be photovoltaic. I would expect thermoelectric to be quite interesting, at least in the short term. Longer than the current launch-cost constraints, I would not be surprised to see viable alternatives that we're not even imagining yet, mainly due to the fact that 'what works well on the ground does not work as well in space' (as you already said) is not a one-way relationship - things will probably end up working well in space that would be silly to try on the ground.

The main problem, currently, is costs. When (not if) that is solved, any space-based solar power system has a heck of a potential to scale. Besides, for now any solar station will very likely be more about advancing technology than actual power generation. One needs to start with a first step in order to reach the 10th step.

Comment Re:Not that new (Score 1) 59

but it only really works if the object has to stay within a certain limited area

Actually, it's even more trivial than that. As they explicitly say in the video, the object has to stay out of the central area. Why? because the central area is where you're focusing the light. Now if they would only take those four lenses, put them in a tube and 'cloak' an absorber around the focal point to remove stray light, they would have a marvelous invention. I suggest calling it a telescope.

Welcome to elementary optics class, now with Harry Potter themed experiments.

Comment Re:OKC's match algos suck (Score 1) 161

It's called the "tyrrany of dimensions". The more variables you have, the more data points you need exponentially to derive meaningful partitioning analysis from it, regardless of how clever your distance algorithms are.

Indeed, but only if you insist on carrying along in your analysis all the irrelevant and correlated dimensions.

And they have hundreds of questions when a dozen would be about all the entire population of Earth could support.

So do surveys, for significantly smaller sample sizes. I wouldn't be surprised if a non-trivial percentage of those questions are intentionally redundant - you know, to check *ahem* consistency, improve accuracy, etc. If, say, you have 100 questions grouped into 10 categories with 10q/cat, you have just dropped the dimensionality significantly while at the same time having more confidence in your data. A rule of thumb in surveys is don't trust the user^W^W^W^W *ahem* trust, but verify.

Comment Re:Ingress is unclear: not inverse cube force (Score 2) 26

It is the strength of the interaction that is found to be inverse cubic. The strength of magnetic force is inverse quadratic. If somebody found evidence of an inverse cubic force then this would be evidence of higher-spatial dimensions and very unexpected indeed.

How did you get modded informative? The magnetic component of the force between electrons in this case is indeed proportional to the inverse cube of the distance. Elementary magnetostatics, since it's the interaction force between two magnetic dipoles (look up dipole-dipole interaction if you want to see the formula). No higher dimensions or other mumbo jumbo required.

Comment Re:Now that Lewis's 15 minutes are up... (Score 2) 382

Until people can recognize the difference between front running (a biased ordering of particular market events) and high frequency trading (low latency response to available market data) then there really is no point in responding to this nonsense.

You seem confused about what frequency means - hint, it's not the inverse of latency. HFT is about (very) low asset holding times, not low latency of the response (although the latter is a necessary means). Case in point, the low latency part, when uses to provide liquidity (as the standard argument goes) would be indifferent to trading patterns - much like a market maker in a stock doesn't pick and choose trades and usually has a requirement to, you know, be there to make the market if needed. HFT, in the fast flipping sense that gave the name, has no such compulsions and very much cares about trading patterns, which together with trend hunting algos has a negative effect on price stability (statistically prone to abrupt swings in both directions).

So do try to understand that high frequency and low latency do not describe the same thing. Otherwise people might start to think that there really is no point in responding to your posts.

Comment Re:Multiplatform? (Score 3, Interesting) 164

It does indeed appear to be OpenBSD only at present (from ):

... and not really that multiplatform for future development, either, since it requires (as per the linked slide)

Modern C string capabilities (strl[cat,cpy]) asprintf etc.

None of the quoted functions are standard C and strl* are BSD-only - yay for GNU-BSD strn*/strl* string function wars :(

It's all nice and good practice that they want to use the best tools available to them on OpenBSD, but not caring for what's available on other platforms is not really how one does portability and *will* produce forks, regardless how much the LibreSSL authors want to 'discourage' it.

Slashdot Top Deals

In the realm of scientific observation, luck is granted only to those who are prepared. - Louis Pasteur