xkcd

Tub

Also, you're supposed to put the BWT before your RLE. Why RLE first? bzip has two RLE steps. RLE -> BTW -> MTF -> RLE Both RLEs have different encodings and maximum run lengths. The first one is supposed to guard against worst-case inputs; due to its inefficient (but fast) encoding it will actually...

Tub

I've done some benchmarks. const LOOPS = 10000000; const ITERATIONS = 10; (function() { const PROP = '_private_data'; let o = {}; bench('prop-single', function() { o[PROP] = { a: 1, b: 1 }; }, function() { let p = o[PROP]; let sum = p.a + p.b; p.a = p.b; p.b = sum; } ); console.log(o[PROP].a); })();...

Tub

I'm aware, but being collision-free and non-enumerable is enough to prevent accidental bugs when multiple separate modules mess with the same object. You can access or change private members in Java with the reflection API, but we still consider them private. You can mangle symbol properties in Java...

Tub

Has anyone found a valid use case for a WeakMap in JavaScript? It's supposed to prevent memory leaks, by allowing you to attach data to objects without leaking the data. The example for a key is often a DOM element, because those tend to disappear every now and then. let wm = new WeakMap(); let e = ...

Tub

psst… use a run-length encoder first. Backreferences are just as good as run-length encoding, the limit here is that common implementations (zip/gzip/deflate etc) have a fixed bit-size for the "length" parameter. For example, a backreference in DEFLATE is at least 24 bits long but has jus...

Tub

Each step is a transformation of a 2-vector (a, b) into a new 2-vector (a+b, a). This is a simple linear transformation, so you can find a 2x2-matrix M such that M * (a, b) = (a+b, a) = T(a, b) Repeated application of a linear transformation is simply repeated multiplication T^n(a, b) = M * M * ... ...

Tub

But there is no trivial assignment at the start. If the assignment w2={j1} is tried, then no solution is possible. w1={j2, j5} w2={j1, j4} w3={j3, j6} I'm a bit short on time right now, so that's all I'm going to add. Implementing the algorithm might find us a better counter-example. This just scre...

Tub

Your reduction is invalid - you cannot "add a vertex of the worker's color"; the coloring problem starts with an entirely uncolored graph. We don't know the complexity of your variant of the coloring problem - adding constraints can both increase or decrease the complexity. Doesn't matter ...

Tub

(Besides, I can write "Graham's number" on the back of a postage stamp. How big could it be?) Roughly twice as big as 1e340, because I can write 1e340 on the back of that stamp twice. :roll: There is probably some plank minimum number of joules below which the concept stops making sense (...

Tub

That last image eloquently illustrates one of the problems inherent in a 2 dimensional world. The only option for the poor surface folk is to attempt to outrun the pole or find an underground bunker instead of simply stepping to the side. They can't really outrun the pole when there's an unpassable...

Tub

You could spell out your information will 2mm ballbearings on a huge flat floor, like a warehouse. Men in Black come calling? Just tilt the floor by a degree or two. a suspicious pattern of dust-free spots emerges on the floor. You really need to turn that warehouse upside-down and shake it a coupl...

Tub

If your build system is as simple as "compile every single file in this directory into an executable", then a simple Makefile will do. See here, for example . This has minimal dependencies, but isn't easily portable to toolchains that don't use gcc/clang's command line syntax. For example,...

Tub

Technically , nothing a person does can actually destroy information due to the Laws of Thermodynamics. Being true to the comic, everything following "Technically" is complete nonsense. As far as I know, you still cant beat one-time-pads for secure encryption. All asymmetrical cyphers rel...

Tub

I think that any cryptographic approach to secrecy is going to be vastly superior to an option where you have to be physically present to set something on fire. It's probably superior against most attacks, but it has a downside: encryption does not destroy information. If a spy runs off with an enc...

Tub

Most video or audio formats are already compressed. Applying a lossless compression algorithm (like zip or rar) won't do anything, as you've seen.

If you need them to be smaller, you need to re-encode them with a different lossy method, but that's going to degrade quality.

Tub

If such a black hole were to instantly evaporate into radiation, that would be equivalent to 100 earth-sized balls of TNT going off. "Fortunately", such a black hole would actually take 2.7e24 years to evaporate. True if gradual evaporation via hawking radiation is the only relevant effec...

Tub

There are several lower bounds for black hole size. The uncertainty principle says that you cannot confine mass in an arbitrarily small volume, so any black hole's size (schwarzschild radius) cannot be smaller than the planck length (which is tiny), but such a black hole would have a mass of planck ...

Tub

If Gardevoir is capable of conjuring objects inside his crank case, it really wouldnt have to be a black hole to wreck his day. How about just a modest sized iron canonball? How about 1kg of glitter in the driver's compartment? Bigger takes longer, and an earth-radius black hole would require almos...

Tub

There's plenty of research about Linear Congruential Generators, if you feel like diving in. They have a few weaknesses, but AFAIK none that would help you here. The state of the RNG has 48 bits, thus only 2^48 of the possible 2^64 different longs can ever be returned by nextLong. According to the d...

Tub

WanderingLinguist wrote:[...]

Thank you for your insights, that was helpful.

(My phone is for phone calls, alarm clock, notes and a web browser. I'm totally out of the loop with the app market.)

Tub

Speaking for my part of the industry (keep in mind, this is not general advice): Mobile app development is big these days, and there's a shortage of programmers who can go beyond the simple "static" mobile app that just duplicates a web site. If I may pop in for a quick question, what's y...

Tub

What Yakk said isn't entirely wrong, but only a small part of all software development is meant to produce software sold to customers. Most software development is meant to solve some internal business need - automating processes, monitoring work results, data mining, totally boring business databas...

Tub

Warning, strong opinions about jQuery ahead. I did use jQuery a lot back in the day, when it had its uses. But believe it or not, for a couple of years I've been programming without ever thinking "This problem would be easier to solve with jQuery." I haven't had much issues with br...

Tub

The first thing you should learn is to write code and comments in english. Always. If you want feedback from others, then english is the one language that everyone in the tech world understands. You also need to get rid of the Allman brace style. As much as I love that style, there are some unfortun...

Tub

I have a suggestion which is not directly "easy to program" because you have to learn some new things first, but: in the long term, you can make programming (in JavaScript) easier for yourself by using some libraries. You could start with jQuery and Lodash . jQuery is largely obsolete (un...

Tub

A dark matter black hole may or may not be a tiny bit darker than a regular black hole :roll: Seriously, if you put heavy stuff within a region smaller than its schwarzschild radius, then you'll get a black hole. Dark matter has mass, so it'll work. From the outside, it won't look any different - al...

Tub

There's extensive documentation about the calculation and interpretation of those values. I bet you haven't produced a single UML diagram, barely did any testing, and there's zero documentation, so your 700 loc aren't finished yet. Sloccount is also highly inaccurate in small projects. Writing the f...

Tub

Assuming determinism and no cp-violations, time reversal gets you back to exactly the previous state, not to a different random low-entropy state. In a hypothetical classical world, time reversal means turning those particles around by negating all velocities. In a quantum world, time reversal is a ...

Tub

If you understand the state of the universe as the whole wave function, then its temporal evolution is deterministic (see: Schrödinger equation ) and it is time reversible. Randomness only appears when you try to force said wave function into a classical notion of particles with fixed positions and ...

Tub

it implies that there are just as many possible states of the universe that, left to their own devices, would tend toward lower entropy, as there are possible states of the universe that tend toward higher entropy. If you pick a random configuration with a given low entropy, then you'll find that b...

Tub

You cannot have a single giant particle in this universe. You cannot push multiple matter particles together as close as you like, either. The closest you can compress matter (without converting it into something else) is what you'll find in a neutron star. So for some fun with actual science, I rec...

Tub

Is what is true/truth with mathematicians not necessarily true/truth in the empirical world of non-mathematician mankind? In a mathematical formal system, you start with a set of axioms, and you try to figure out everything that follows from those axioms. An axiom is just a premise, something that ...

Tub

Even booting from the corrupt drive will cause further writes to the drive. If that drive contains the only copies of important files, I'd avoid that. See if you can get your time machine working before you touch the corrupt mac again. If the time machine is lost, the first step would be to make a c...

Tub

The convention should've been "don't write a language where..." Unfortunately, end-user-programmers don't get to design their own language[...] End users can choose to enable compiler flags that turn accidental assignments into warnings or errors. For scripting languages, the same can be ...

Tub

Hm. Is a high heat capacity a good thing? I thought a lower heat capacity means less energy is required to generate a particular temperature difference? The carnot efficiency is independent of the working fluid. Lower heat capacity just means that you need a larger volume to transform the same amou...

Tub

A heat engine requires neither combustion nor vaporization. It requires a hot end, a cold end and a working substance in between. Taking any real-world heat engine and replacing coal fire with a magic spell should be straightforward. Look at engines or power plants developed by humanity when we had ...

Tub

Haven't seen the movie (not that popular, apparently), but here's the phase diagram of water: https://upload.wikimedia.org/wikipedia/commons/0/08/Phase_diagram_of_water.svg There are plenty of youtube videos of water in a vacuum chamber, and in many of them the water will boil violently for a while,...

Tub

I'm glad the pedants decided to step forward. Now go start working on that list!

Tub

Still, I contend that in analysis of algorithmic complexity, the n is usually the length of the input , so n is already the log of the input value, i.e. the prime itself. On that basis, by your argument, the algorithm for factorising a known prime would be O(n). I usually expect to read O(b) if we'...

Tub

orthogon wrote:You can factorise primes in O(1) if you know they're primes. Non-primes are trickier to factorise.

Well, a prime number n has O(log(n)) bits. If numpy makes a copy somewhere, then the algorithm is O(log(n)) as stated in the comic.

Search found 382 matches