Incorrectly for fun: umlaût
Suppose two consenting people wish to engage in having sex as an icebreaker. What is some advice for success?
It will probably only mutually work, i.e., be an icebreaker successfully, for certain kinds of people. What kinds of people? It probably depends heavily on what sex means to each person.
Probably the goal is to have a long conversational cuddling session afterwards. So don't do it too late in the day when people might fall asleep immediately afterwards.
Biochemically, we are probably wanting certain hormones to become active, perhaps pair bonding. What causes those hormones to be released?
Somewhat surprising is the lack of prevalence of restaurants, especially delivery, from which one can order (and pay) days or weeks in advance, for a discount.
Ordering in advance allows the restaurant better inventory management: how much ingredients to order and avoid wasted food. People often know their meal plans or desires in advance, so would be glad to have a discount.
In some locations, there exist multiple distinct societies layered on top of each other, living in geographically the same area but otherwise not mixing and barely interacting. Most commonly class divisions in dense urban settings.
Such locations are not suitable for geographic-based political boundaries, i.e., legislative districts. Instead better would be multiple legislators "at large" from a single district.
In contrast, geographical districting is useful when there is a very large area of a homogenous society: the residents of each district can have a legislator who is physically close to them, perhaps for face-to-face discussions.
To what extent does the experience of growing up as an African-African male include informal education to steer clear of white women, or some certain subset of white women, who can wield much political power (playing the victim card) when accusing rape or other sex crimes against them? (This would be similar to the informal education teaching African-Americans to steer clear of cops.) If true, it would be a good subject for a documentary.
If you don't want to be hanging by your neck from a tree next morning, don't even risk interacting with them. Many historical lynchings were about sex crimes, or broadly unacceptable sexual actions that the official justice system could not prosecute to the death that the lynchers deemed necessary.
(If the lynchings induced the current (hypothetical) education, then it is another case of terrorism actually working as planned.)
The issue is probably much larger than a race boundary, probably class warfare also, despite seemingly fewer actual lynchings not at a race boundary. If the issue is common, then if one is a person (stereotypically a middle or upper class white woman) bearing social class markers indicating that people will listen to, and believe, an accusation of rape or other sex crime, then courtship life will be lonely: rationally, no one wants to risk their neck. Game theoretically, such women will seek to conceal or shed such social markers, though this may be difficult because those markers may define identity.
Inspired by the extrajudicial social and online lynchings seen these days against sex crimes.
Create a TV show depicting behind the scenes of producing another TV show, and in particular the behind-the-scenes discussion of providing the audience with the entertainment they crave (to make the most money) but obliging the legal, political, and moral censors.
In the past with the Hays Code: (implied) premarital sex was permitted so long as it was depicted as bad, that the characters doing it suffered punishment in the plot.
Modern similar example: (implied) child pornography depicted only as bad -- the criminals who do it getting punished -- on Law & Order SVU, but indirectly satiating that demand.
Given an algebraic expression, test whether it is identical to zero for all possible assignments of variables. How difficult is this problem? It probably depends on the operators and functions in the expression. I wouldn't be surprised if it is Turing incomputable.
Suppose we limit to asking whether it is zero for all but a set of measure zero, where it could be undefined or take a value different from zero. For example, (x/x)-1. Is the problem easier?
For all but a set of finite measure (area)? Probably look at asymptotic behavior.
Pick a few random values and plug them into variables. Evaluate and test if the result is equal to zero (this is potentially tricky). If it is nonzero on a set of measure zero, then one has to have been tremendously unlucky to find a nonzero.
Difficult cases: a expression that is nonzero only for small neighborhood around the first hypothetical counterexample to the Riemann hypothesis. Or around all counterexamples, perhaps summing up to an infinite area.
An expression which zero outside of a sphere of radius G, where G is some huge finite number.
Zero for all irrational numbers, nonzero for rationals (Dirichlet function). Or, zero for all non-computable numbers, nonzero for computable numbers.
Intimidation is an important skill for success in various cultures and subcultures. Where do people learn it, and in particular, practice it during the process of learning?
For many things, one enters the world in the most subservient status, so not an appropriate venue for practicing intimidation: child subservient to parent, new employee subservient to bosses. Then, one can imagine lack of opportunities for first-hand practice in intimidation preventing advancement, so a feedback loop.
Learning by intimidating younger siblings. Or perhaps male siblings learning on female siblings. Not everyone has siblings.
Obviously, schools. But policies against bullying may prevent people learning this socially important skill.
Of course, the fact that it is an important and socially useful skill is a larger problem.
Previous similar, learning cheating and deception as life skill.
Technology marches forward at (seemingly) exponentially faster rates. Network effects, broadly, might induce everyone to adopt some technology. However, the effect of a technology might require generations to properly evaluate.
Consider making people who refuse to adopt some technology a protected class, withe government providing them social assistance, in order to generate socially useful data about the delta effect of the technology.
(Of we currently already get such differential data from third-world countries who simply lack access to some technology.)
The protection lasts only a generation or two from the introduction of the technology, probably invalidating protection for most instances of people stupidly refusing vaccines.
Tell the story of a superhero whose one superpower is the ability to solve all NP problems in polynomial time.
As a comic superhero with a mostly useless power, maybe the exponent of the polynomial is extremely large (but finite).
What competitive sport offers the greatest opportunities for surreptitious cheating? Hypothesize correspondence chess with a prohibition against external assistance including computers. Then, form a correspondence chess league among friends who trust each other not to cheat.
More interesting than the game is the trust network. When will such trust networks form? Can they be leveraged for something else?
Consider a tar file, or some other archive format of many files, which is becoming available as blocks in random order, perhaps through bittorrent. Unpack the files in the underlying archive as they becomes available, instead of storing the whole thing then unpacking it. This is to save disk space.
Blocks corresponding to incomplete files will require some overhead.
Bittorrent itself however already has a mechanism for distributing collections of files as one torrent.
Probably the easiest non-Euclidean 3D manifold to model, perhaps for flying around in virtual reality, is the surface of the hypercube.
What happens at the corners? Probably easiest is you can see into neighboring cubes, but you cannot see "around the corner" into neighbors of neighbors. Or maybe you can; things get weirdly replicated as if some strange mirror.
Consider a gluing matching faces of polyhedra, building structures. For example, regular dodecahedra will not tesselate space, but I suspect one can build interesting helices (spirals). Boerdijk-Coxeter helix.
Limit to polyhedra with faces with regular polygons of unit side length to keep things simple. Though star faces and star polyhedra could be included.
Hinges if relaxing to join only at an edge. Joints for joining only at a vertex. Restrictions of faces to regular, and edges to be unit length, can correspondingly be relaxed.
Easiest to do it in virtual reality. In 2D or 3D we seek user interface tools which provide nice feedback when a polygon or polyhedron clicks into place correctly next to another one.
Consider a drawing or modeling program in which the unit line segment joining others at its endpoints is the only primitive. Shortcuts for regular polygons. Also including stars, apeirogons. Shortcut for rhombus of useful angles, probably constructing an edge parallel to another one, or complete a rhombus given two edges meeting at a vertex. Shortcut for a polygonal arc between two points.
Higher dimensions (wireframe nets), hyperbolic spaces. Abstract polyhedra.
In a Rubik's cube program controlled by keystrokes, useful would be animations depicting what each key does. Alternatively, click on an animation to operate (awkwardly) by mouse.
12 quarter turns of faces, 6 whole cube rotations. More for slices, 2 layer turns.
Where should one donate to maximize the amount of political support in return, from recipients of the donation or their supporters? This has probably been worked out by the rich.
One scenario is money which has been obtained through nefarious means. Donate (some of) it in return for political support in case you get in trouble for the nefarious means by which you obtained the money.
Inspired by Sandy Jenkins, Collins Street Bakery.
Probably the best is to donate to political campaigns, though there are some devils in the details. Does this help explain why there is so much money in politics?
Probably not donate to the poor, because the poor have little political power.
Using a stream cipher in the background, one can simulate uniform random samples of infinite precision floating point numbers within a bound, e.g., between 0 and 1. Behind the scenes, only a random key for the stream cipher is being sampled. The sampled infinite precision floating point number can have as many digits or bits as needed generated on demand.
Comparing such a number against another, or against a rational number, is easy. Arithmetic is harder, including radix conversion.
Can one do non-uniform random samples, e.g., normal distribution?
Consider slides in random order for which the time serves as an index. However, if slides take varying amounts of time to display, things become tricky.
One way to do it is to schedule blocks of slides at a time. Perhaps an hour worth of slides is precomputed from that hour's random seed. The total number of slides shown in the hour is varying, depending on which slides took longer. One can later know what slide was shown when from recovering the random seed computed from the hour, then linear search by time.
Avoid repeats between blocks by the following approximation: compute not just the current hour but the blocks 2 and 1 hour before. Assume all slides were shown 2 hours before: this is an approximation. Compute the approximated block 1 hour before avoiding repeats with the approximated block 2 hours before. Compute the current block avoiding repeats with approximated the block 1 hour before. We must use approximations and not reality to avoid an infinitely long Markov chain into the past.
Need to do slight fudging to cause each block to be exactly 1 hour. For example, scale everything by time.
Traditionally, learn, or be exposed to the source work first, then the derivative work, to understand and be able to enjoy the context.
But people can and do do it the other way around, an interesting mental task, perhaps mentally filling in the gaps or remaining not unpleasantly confused.
Sometimes the much later derivative work has been better adapted to a modern audience.
Watch the movie based on the classic book first, or after?
The cover of an old song?
A player races through a manifold, seeking the shortest path. Unlike most racing games, speed does not matter. Animation could depict where other past racers were for your current distance.
Perhaps race through a tube of varying curvature and width. The racecar travels "in a straight line" if given no input, though that might be a tricky concept on curved manifolds.
Even finding the shortest path between two points on the surface of a cube is nontrivial.
Perhaps different skill classes for corresponding to the minimum speed permitted. Higher minimum speed is harder.
2D and 3D manifolds.
An artist publishes a work, then solicits suggestions for improvement. The artist then incorporates some of the suggestions into additional versions of the work.
Most ambitious would be to do it for a movie, after it comes out in theaters. The released DVD contains additional scenes filmed from the suggestions, that the audience can splice in if they want.
Some sort of crowdsourcing platform in which the good ideas get voted up. Need ability for collaboration, but also forks and merges, an unholy union of git and reddit.
Issues of copyright: Creative Commons CC-BY, or ambitiously CC-BY-SA probably works. Cannot have NC.
Nightclubs, by design, induce in their attendees the extremes of human emotion: extreme joy and extreme sorrow. Hopefully more of the former, which is why people attend, but there are sort of zero-sum or conservation law effects that can occur.
If a nightclub doesn't have a mass shooting every once in a while (where once in a while might be something like once in a thousand years), it's probably doing it wrong.
Inspired by Orlando.
How are contracts and other rules enforced in regions that have strict gun control? There remains a mutual demand for some system of dispute resolution in parts of society, usually the bottom or untouchable classes, not adequately served by the normal justice system.
Create a character set to express numbers from 0 to 99, as two small characters stacked vertically, so they take up no more space than a single typical Latin character. We want it to be legible at low resolutions, so traditional numbers reduced in size probably won't work. Here is an idea based on various accents:
single double grave acute accents
left right wedge
Consider radically modifying baseball by eliminating the pitcher. Instead, the ball is hit off a tee, like in youth leagues.
Probably also eliminate the home run.
Need some way of making hitting harder, so the batter cannot easily control where the ball will go. This difficulty exists in golf driving. Maybe lighter, more textured ball.
Ball could be levitated by Bernoulli effect: it bobs in the air.
Base stealing either no longer happens, which would be sad, or needs to be significantly modified.
A live webcam of a bug zapper in action would be entertaining. ("You might be a redneck if...") Difficulties / desired features:
Brightness and contrast: ideally the camera's vantage point would be from inside, so the light is behind the camera. And cylindrical fisheye, probably synthesized from multiple cameras.
If not filming from the inside of the device, one way to avoid the brightness of the light is to have it flicker at a very high rate, then have the camera shutter synchronized to be open when the light is off.
Desirability of slow motion and magnification. The sound it makes could be a useful cue of what to replay in slow motion.
12 hour replay, 6 month replay.
Some sort of machine learning to detect when the bugs are out and the lamp is most effective. Of course, a light sensor can be one component. Some reports that it can also be effective during they day when it rains the the bug zapper is covered from the rain: bugs seek shelter from the rain.
Some way of cheaply attracting mosquitoes. Attractant (octenol) is reportedly expensive. Generating warm CO2 should be easy.
Some way of easily cleaning it, including the bug bits that get splattered on the camera lens.
Electric bug zappers preferentially kill bugs attracted to the light. It could therefore increase bugs not attracted to light, for example mosquitoes, as they expand to fill the ecological niche made empty. Also, as it decreases bugs, it decreases food for bug predators, so decreases bug predators, which do not preferentially eat bugs attracted to light. The decrease in predators could increase the prevalence of certain bugs.
The advantage of listening to music at a music venue, e.g., dance venue, is that someone else, a DJ or music director, picks the music for you. A sophisticated DJ could program on the fly, adjusting the playlist by observing the audience. A less sophisticated DJ could incorporate feedback offline for the next iteration of the event.
The task of observing and incorporating feedback into music selection seems like a very human task; I don't see machines (e.g., machine learning) ever doing a good job at it. People rarely give obvious and useful feedback about how they feel about the overall music selection, unless it is really bad. There are even sociological and political decisions to be made: some people may be happy about the music, some people may be sad: which group is more important?
Desired generic features for account access (authentication):
Multiple keys, probably challenge-response with public-key cryptography. Making it look like password authentication can be delegated to a trusted third party.
Set and run arbitrary policies regarding how keys must be used to access the account. Perhaps several keys are needed simultaneously, some controlled by 2 factor authentication mechanisms using a trusted third party.
Set and run policies regarding how keys must be used to modify the account access policies, for example, revoking a key. Perhaps more keys are needed than mere account access.
Logs for key usage, sent off site to a trusted third party.
Devil in the details of these trusted third parties.
We don't seem to have easy to use tools like grep or regular expression libraries for the next step up in the Chomsky hierarchy, namely context free grammars. (Bison and yacc, happy, parsec, etc., exist, but they are rather cumbersome compared to regular expressions.)
Two cubic time parsing algorithms: the Earley algorithm and the Cocke Younger Kasami algorithm, with plenty of opportunities for optimization.
It might be lack of demand. Regular expressions might be enough, especially with non-regular extensions like matching backreferences in perlre.
Regular expression matching by default is unanchored: look for a match anywhere in the string. Context free grammars are by default anchored at the beginning of a document.
Regular expressions typically process one line at a time. Context free grammars typically process the entire document at once.
In response to the advancement of technology in recorded music, live music, or live entertainment in general, should push what live can offer but recorded cannot.
Generically, one of those things is interaction between the performer and audience, though the devil is in the details.
Sometimes, perhaps often, people psychologically want to passively consume entertainment without interacting. The performer could still passively observe the audience, but this may also be unsettling to people not wanting to be observed.
Create a T shirt with a molecule depicted on it (of course, not an original idea). There are many, many different molecules to suit different people's different preferences. Probably custom one-off printing.
We need a searchable database of chemicals: these already exist, though not targeted to the general public seeking images for a T-shirt or other art.
Consider a hypothetical society in which no one explicitly says no to sex: the lack of desire for (further) sex is communicated by lack of enthusiasm and bad chemistry during sex.
Could such a society exist? If so, what allows it to be possible? Does it exist? If this is the accepted custom of both sides of sexual interactions, it will interact poorly with some expecting consent culture.
It exists in dance culture: people tolerate one bad dance, then the nonverbal message is don't dance again with someone you have bad dance chemistry with.
Let's not be racist, we say, and we enact civil rights reform. Then we're racist again with the War on Drugs.
Distill this human fallibility of making this same mistake over and over again, despite the best of intentions, into a game, to illustrate it and highlight it and maybe prevent it.
I strongly suspect it is repeating again in "safe space" and "consent culture".
We consider large square board chess variants which are formed by simply increasing the number of rooks and bishops (and pawns). For random positions (leaving unspecified what is "random"), what percentage of squares are attacked? How does the percentage change as the board size increases?
The worry is that as the board size gets larger, because the relative portion of the board a single ranged piece can attack shrinks (curse of dimensionality), so the game will devolve into a boring game with relatively weak pieces.
What percentage of the board can be attacked in 1 move (the above was considering attacking in 0 moves)? On an empty board, a rook can attack any square in 1 move, no matter the size of the board. For large boards, the relativity sparsity of pieces might aid the ranged pieces. Is being able to attack in 1 move useful? The defender can often escape in 1 move as well.
A flying chess variant piece may move to any square in a region, perhaps the whole board, perhaps a portion. Modify it so pieces in the way can cast shadows, blocking flight. A piece at the same rational slope but closer blocks flight to a further square.
Nuclear weapons are seemingly only defensive weapons: if a country uses one offensively, they will suffer a devastating nuclear counterattack from either the country attacked or its nuclear allies. Iran, say, cannot offensively launch nuclear strikes against Israel or Saudi Arabia without incurring its own immediate obliteration.
That said, devise a means for a country to surreptitiously use nuclear weapons offensively against an enemy country.
First idea: give the weapon to a terrorist which cannot be traced back to the offensive country.
Second idea is more dramatic: stage a "fake" invasion by the enemy country, then use nuclear weapons against the invading enemy in the guise of defense and counterattack. The "fake" invasion is militarily real; it's fakeness derives from the surreptitious political engineering done beforehand by the offensive country to induce the enemy country to launch a real military invasion. This "fake" invasion could even be a nuclear attack, somehow convincing the enemy that incurring nuclear counterattack is "worth it". This is the kind of intrigue I could imagine the Mossad or CIA doing: both supporting Iran's nuclear program enough to build one bomb (survivable for Israel or Saudi Arabia or the United States), as well as supporting the growth and popularity of religious extremism ("worth it") in Iran, all for engineering an excuse to literally wipe Iran off the map. Perhaps ultimately to control the oil below.
Inducing a country to become extremist is not difficult: people become nasty when there is only a small piece of pie to fight over: economic sanctions should work.
American political correctness (in arguments such as cultural appropriation) likes to place non-white actors in films, but have them act in a way that white audiences can relate to (because making money off white Americans is the bottom line). Inspired by Western remake of Ghost in the Shell.
However, more interesting from an acting and screenwriting standpoint would be the opposite: white actors accurately acting non-white. Someone from another culture watches the film and thinks, "That's me / my culture! I've never seen me accurately depicted on screen (in an American film)! Even though of course the racial features are of course not accurate." This is, of course, not a good way to make money off of white Americans, who will not appreciate it, and may even find it unpleasant. Even foreign viewers unable to suspend disbelief might find the cognitive dissonance unpleasant.
Reversing again, we might see it in foreign films aimed at the American audience, those that do well with Americans. Though we do need to distinguish between a foreign film doing well universally (because it appeals to something universal in humans) or specifically for Americans.
For some people, their sexual orientation matters only for the type of porn they prefer to watch (and is just one of the many preferences affecting which of the zillion categories of porn one can watch). For others, it means much, much more, strongly affecting major life decisions. What causes this difference in people? Probably (among other things) differing social structures around courtship and marriage.
The stickerless versions of the YJ Yulong speedcube come in several different shades: original colored version, fluorescent colored version, pink version, and transparent version. Can they be disassembled and pieces from different shades combined?
Keep track of how many times each element of a set has already been selected. To select the next item, first uniformly select a subset of N items, then choose the one among the N which has been selected the least. Increment the counter of that one. Replace to do the next selection.
What kind of probability distribution does this induce? Parameter N.
Through various times in history, Greek, Latin, and English have been the lingua franca of science. Possibly others: Arabic, French, German. What are desirable features for a lingua franca for science, assuming we could switch away from, or modify, English?
Ability to coin new words, with guessable meanings, for new concepts. English reaches for Greek and Latin roots: not great because it requires knowledge of Greek and Latin, but it works.
Easy to write. Modern printed English has this; logographic languages like Japanese and Chinese are much harder to write, both for human input and for typesetting.
Machine readable with OCR. Again, modern printed English is pretty good, though mathematics is difficult. This is a vague vision of the future in which computers may be doing a lot of knowledge mining.
Ability to be precise in meaning. I'm not sure any language is any better than any other language on this point, especially because languages are living, so can be modified as necessary for when precision is demanded.
Easy to learn, at least the subset used to communicate science. English, with its tricky grammar and spelling, gets criticized on this point, though all living languages accrue complexity making them difficult to learn.
Easy to translate, especially machine translation. Of course, this depends of the distance between the two languages.
Easy to speak well enough for others to understand. Most languages do not have the unvoiced and voiced th sounds of English, but substituting s and z is still intelligible. Other difficulties: r/l, r-colored vowels. In contrast, I suspect tonal languages would not be easy for nontonal native speakers to learn.
A Rubik's cube limited only to slice moves is a relatively easy, though not trivial, puzzle. Having a nice speedcube helps, not an original Rubik's cube.
Can every state in which the corners have not moved but the edges have be reached only though slice moves? I'm guessing not: superflip is one such position. How many distinct orbits are there?
Create a communication device with which the speaker can speak extremely softly, e.g., whisper or mumble, and a listener on the other end can hear it intelligibly. However, if the speaker were to speak loudly, the listener's can also comfortably hear it without getting eardrums blown out.
Can the speaker speak softer than a whisper? Perhaps some lip reading sensor as the speaker simply mouths the words.
Consider modifying soccer (association football) so that touching the ball with one's hands is permitted if one is not in contact with the ground, i.e., only when one is jumping.
Partially inspired by reports injuries and brain damage from heading the ball.
If we are to believe George Lucas's story, the midichlorians saw or foresaw that Palpatine was, or was going to become, too powerful, so they engineered the conception of Anakin Skywalker to assassinate him, to bring balance to The Force. Their plan did eventually work, though through tremendous convolutions: Anakin had to become the Emperor's confidante, beget a son (Luke), and only though his love of his son would he find motive and opportunity to turn and kill the Emperor. Killing the Emperor came at extremely high cost: the deaths of almost all the Jedi as well as a galactic civil war. The midichlorians are either extremely cold hearted in how they achieve their objectives, or the Emperor must have been extremely important to kill. (One wonders how much worse things would be, or could be, if their plans failed and The Force became extremely unbalanced. Maybe the universe would have ended.)
And through all this, the Emperor, both possessing prescience and knowledge of The Prophecy, was still unable to foil the midichlorians' plan, and in fact walked right into their fatal trap.
Create a game based on the mental challenge of visualizing compositions of rotations in 3 dimensions. Rotations in 3D do not commute. Such visualization is something people can get better at with practice.
Two people, one cube with distinctly marked faces. They take turns rotating the cube according to some rules about what rotations are permitted for a given player on a given turn. Whoever achieves the goal orientation state wins. Or dodecahedron.
Rotations of a sphere (a Lie group). Perhaps some mechanical device that limits the rotations permitted from the current state.
Everyone has sexual boundaries (this is knowingly an incorrect assumption) that vary case-by-case, depending on some characteristics of the person being interacted with. Then, consider a person's most permissive boundary, the best case scenario, for a class of interactions, for example, hook-ups with people one has just met.
Do people differ significantly among where their most permissive boundary is? Or is everyone boringly the same, anything goes, in the best case scenario? If people differ, what causes the difference? Presumably some social indoctrination mechanism, the details of which we seek.
"I pledge to be monogamous to you, except for the following list of celebrities..."
Of course, the other critical question is, what characteristics of the other person cause someone's boundaries to be more restrictive than their best-case-scenario boundary?
In the Lebensraum arguments that convinced Nazi Germany to invade the Soviet Union, did they correctly predict, and give the correct mechanisms for, the economic stagnation and demographic problems now seen in Japan? Did anyone in Japan correctly predict the current economic state of Japan? (Predicting the low birth rate would have been quite a radical prediction.)
Germany itself so far does not seem to have suffered gloom and doom from not having room to live.
A rough idea of judging whether a scramble, perhaps hand scramble, of a Rubik's cube has been done long enough is that "things" have gone through a state that is as far from the solved state as possible.
For example, every cubie has an antipode: during the scramble, every cubie must have passed through its antipode position at least once. Or, every cubie must have passed through its original position, but in an incorrect orientation. Or both of the two conditions, in either order.
Other variations possible, including the very difficult diameter of the full Rubik's cube graph (God's number). Or, always start scrambles from Superflip.
An online news article can provide options to hide people's names, genders, locations, and other markers that people use to form preconceptions based on stereotypes. Hidden is the default; the reader has to actively choose to see those things, actively admitting that he or she wants to judge by those features.
The next iteration of chess opening play out will probably not play until end of game; instead, concentrate on just the opening phase. Instead, it will explore a much wider set of openings. Rough idea is that the outcome of the game is sometimes obvious when the middlegame rolls around, and computers of the future will easily be able to play out the remainder of the game.
Put the unevaluated positions in a priority queue. Components to calculate priority: total weight of position using chess piece values, move number, distance from main line (number of calculated moves).
For what software is it important to translate a (say) English UI into the local language? For what software can the user figure it out, either by translating from English themselves, or figuring it out by intuition or similarity to other interfaces?
Proportion of the software's users who have learned English as a second language. How many different things does the program do? Is what it does obvious?
Hypothesize that close economic ties prevent war. Because war is a negative sum game, we should obviously get on implementing this as soon as possible. But we aren't.
Is the hypothesis wrong? The devil is in the details of the implementing the economic ties: are the details too difficult?
Create a container in which to keep a running computer so that it will do an action if it detects that it is being tampered, e.g., in a law enforcement raid, perhaps wipe cryptographic keys.
The container could be the size of a computer case, or the size of a room or house.
Previously, for data centers.
Optimistically, one can describe the development of chess opening theory as a vast distributed computing project spanning centuries, seeking ground truth. Computers can continue this computation, and they do it well.
Pessimistically, and realistically, theory develops in the direction of offering the best "chances" against an opponent unfamiliar with the position, while also minimizing risks if the opponent is familiar with the position. It is a psychological struggle about the familiar and unfamiliar, human fallibility, a risk-benefit analysis with probabilities. This is very different from how computers play the game, at least currently.
The original idea was, chess at the highest level, with paid competitors and widely published and studied games, should pursue the noble optimistic goal above, advancing the state of knowledge about ground truth. The game at the highest level should be modified to best further the search for truth: probably incorporate computers and databases into at least the opening phase of the game, e.g., Advanced Chess.
However, chess at the highest level (and every other level) is actually mostly played according to the pessimistic interpretation above: how can I trick my opponent, for one game only? I mostly cannot see how that can be changed, except maybe for matches.
Incidentally, correspondence chess, despite computers, is also closer to the pessimistic interpretation: how can I avoid the weaknesses in my chess engines and exploit the weaknesses in my opponent's engines?
Actual ground truth in chess is probably that it is a draw.
Look at someone, for example, how they dress. Make predictions of how they will behave based on how they dress. (Continues the idea that a person is concretely their actions.) When are the predictions accurate? When they are, why are the predictions accurate? That is, what common cause directs both how a person dresses and how that person behaves, yielding correlation?
Inspiration was gender conventions of dress: masculine and feminine dress. If you see someone in feminine dress, what can be accurately predicted about their behavior? If you see someone else otherwise similar but not in feminine dress, how is their behavior predicted to be different? Many people have an intensely strong compulsion to dress according to their gender convention: not doing so is intensely uncomfortable or distasteful. Whatever mechanism programmed that compulsion, what other behaviors did it program?
The markers of masculinity or femininity are complicated social constructions, but they are just the surface. Looking deeper: Why does, say, femininity matter? Why do the behaviors correlated with femininity matter? Wild guess: imperfect information signaling mechanism to convey "I will be a good parent".
Often dress communicates social class. However, masculinity and femininity seem separate concepts from social class.
A reputable web crawler creates and makes available an index of (URL,hash of content) tuples. Maybe also add size.
Later, a user, wanting the content of a URL that is dead, looks up the URL in this index and gets the hash. The user then offers a prize to anyone on the internet who can supply content that hashes to the given hash. This can be verified automatically, though the details might be tricky. Probably need reputable broker or escrow.
The people you are attracted to represent your hopes. The people you can get along with in a relationship, people you are compatible with, represent reality. Conflict and turmoil abound if hopes do not match up with reality.
Examine the mechanisms by which attraction and hopes form and reality occurs, and why they diverge.
The canonical example is people hope to marry up in social class, but habits and behavior makes them compatible with only their social class.
Given a vector valued function F(t), derive its arc length parameterization, that is a function T(s) such that F(T(s)) moves along the arc at constant speed as s increases at a constant rate. We need automatic or numerical differentiation, quadrature, and root finding to invert the arc length integral. There might be a trick of inverting before integrating (implicit differentiation) to avoid needing root finding.
The motivation was for a color wheel (future post), needing to have a path that moves in CIELab color space at constant speed along a path that was specified in RGB space.
Ajinomoto (of course)
Sazon de goya
Hydrolyzed vegetable protein
Cured meat, aged meat
Cheese (e.g., Parmesan)
It's not that these foods merely contain MSG -- they do -- rather, the entire point for the existance of these foods -- why they got processed from their original state -- was to break down their proteins to release glutamate, the source of the savory or umami flavor.
(OK, for some things, preservation may have been a higher priority.)
The Salsa20 cipher core is reversible, so it almost looks like a weird block cipher running in counter mode to yield a stream cipher. It takes 128 bits of plaintext (the nonce and block counter) and yields 512 bits of ciphertext.
(Exercise, implement Salsa20 "decryption" by reversing its steps.)
Unfortunately, running the Salsa20 core backwards from the 512-bit ciphertext can be done without knowing the key. And doing so yields not just the plaintext but also the key.
One handed puzzles are surprisingly gratifying because they leave the other hand free to do something else, e.g., eat or operate a phone.
What is the best one-handed mechanical puzzle? The classic is a small, loosely tensioned Rubik's cube, but that is a considerable dexterity challenge. 2x2 is easier.
Restricting oneself only to half-turns (no quarter turns) on a Rubik's cube results in each face having at most two colors, the colors of opposite faces.
Is every reachable state (allowing quarter turns) that has such a coloring reachable with half turns only?
Consider the task of generating random pretty images like the Electric Sheep project, but on the surface of a sphere. Enjoy the presentation on a spherical display, or with virtual reality from the inside of the sphere.
With virtual reality one can also add depth, perhaps wisps of colored translucent gas or kaleidoscopic polyhedra flying around.
There remains the details of encoding images and video for the surface of a sphere, though astronomers have already solved this with HEALPix.
Table cells are too small to hold large amounts of text. The standard technique is instead to put in the cell a numeric or symbolic footnote reference, and then have footnotes below the table.
We could do a little better than just a number: a several letter code or abbreviation can give a hint or summary of the contents of the footnote. The abbreviations are given in alphabetical order below the table.
Inspired by Wikipedia tables, which make numbered footnotes easy, perhaps too easy.
An endorheic lake, or one with very slow outflow, stratifies, most obviously by temperature. Consider installing a screw-like pump to mix the layers. The screw can easily be powered by wind, or possibly even by the temperature gradient.
This may of course drastically affect the pond's ecosystem. A solution in search of a problem. What flux thoroughly mixes the pond? Very slow might still be effective.
Stuff from the bottom of the pond getting caught in the pump is of course a problem.
"Clean" and "unclean" are incredibly powerful psychological concepts, so easily abusable to manipulate people. When has it been used on you?
It is very difficult for a person to change their ideas of what is dirty.
Kosher. "Dirty" politician. "Dirty" people, or races. Cleanliness is next to godliness. Purity.
Greedo finds Han Solo before Obi-wan does, shoots first, and kills him.
Luke and Obi-wan then find a different freighter pilot at Mos Eisley cantina (frequented by "many of the best freighter pilots") and the whole saga unfolds differently.
Vaguely inspired by Harrison Ford wanting to leave the franchise.
30 games of blitz chess, spread out over at most 3 days. First to 15.5 points wins. Each player has one "time out", which adjourns play for the day. On day 2, adjournment can occur only after game 15, so at most 15 games are required on the third day.
Three day schedule inspired by single-elimination chess world cup (or women's chess championship).
Perhaps Chess960 to prevent games from repeating.
Tiebreak idea: advantage goes to the one who called timeout later. If neither player called timeout, i.e., all 30 games completed on the first day ending 15-15, then there remains 2 days for some other tiebreaker procedure.
The user specifies which letters are needed, the relative sizes of them (maybe spacebar and backspace are larger), and the dimensions of the soft keyboard. The computer packs the discs efficiently into the region then computes a weighted Voronoi partition to assign the areas between the discs.
Let a soft keyboard provide 24 buttons. 24 is a nice number because it can be laid out in many rectangular areas, depending on (say) orientation: 1x24, 2x12, 3x8, 4x6.
We need to vote some letters off the island, relegating them to a second screen. More than just two, if we want spacebar and maybe some common punctuation on the main keyboard. Maybe spacebar is double sized or more, because it is important, triggering autocorrect.
30 is the next number with lots of factors. We gain divisibility by 5 but lose divisibility by 4, so probably not worth it.
Feed partial input to the compressor, yielding compressed output blocks. Feed the compressed blocks to a decompressor, yielding a prefix of the partial input. The missing suffix represents the unfinished last block.
The decompressor should operate in an aggressive mode that outputs every byte as soon as it is available, not holding on to bytes for buffering, not waiting for the final "flush" block which will never come. Perhaps some sort of recovery mode.
This was a half-baked idea, continuing thoughts on appendable compression without loss of efficiency (asymptotically).
Pack a given collection of discs of different sizes into a given region without overlap, scaling the region larger if necessary. Minimize the scaling factor.
Start by placing the discs randomly. They will overlap as well as extend outside the region. Pick two discs that overlap and nudge one or both so that their centers move diametrically away from each other, so that there is locally less overlap. This can also be done with a disc extending beyond an edge. Accept if this improves a scoring function of total overlap. If after enough steps and nothing seems to improve things, then scale up the size of the region. Apply simulated annealing, including large jumps.
Improvement: for a selected circle, examine what it overlaps. Compute a vector sum of pushes from the overlaps and nudge in that direction.
One can generalize a convex polygon to have circular arcs as edges. The area is the intersection of the interiors or exteriors of a collection of circles. A straight edge is a circle of infinite radius.
If we permit unions, then non-convex is possible. Reminiscent of constructive solid geometry.
It is easy to test if a point is inside or outside an edge (circle), and how far inside or outside.
Given noisy empirical data about an event, estimate the probability or frequency of its future occurrence. A difficult task for natural or artificial intelligence. The nature of the noise could be extremely complicated. It takes intelligence to understand the nature of the noise and compensate for it.
But a predictive algorithm can later have its accuracy evaluated as the future rolls around.
Inspired by rare side effects of osteoporosis medication: do the benefits outweigh the risks? One needs to know probabilities.
Things would have been interesting if, when we peered out into the universe with telescopes and more powerful instruments, we saw glaring signs of extraterrestrial intelligent life, civilizations with far more capability than our own, perhaps astroengineering. But we don't, unless we've missed the obvious.
The cosmological principle predicts an "End of Greatness" at large scales. Imagine mapping superclusters of galaxies to discover a large void shaped like a perfect cube cut out of the universe.
There are as yet unexplained deviations from the cosmological principle, e.g., large quasar groups.
A type signature to a function creates the requirement that an argument passed to the function have a specified type. However, operationally, the function only needs for the argument to have certain operations (other functions) defined on it.
Type classes in Haskell provide a means for a function to be polymorphic, accepting many different types, with the type class defining the set of fundamental functions (methods) available. However, even type classes may be too much: a function might require only a subset of the fundamental functions.
Duck typing is a much "looser" way of addressing this same problem. Previous thoughts on this problem.
Sometimes, even though all the operations a function needs might be defined on a type, we want to use the type signature of the function to restrict what it can be called on. The "explode" method might be defined for both firecrackers and nuclear weapons, but we might never want to accidentally pass a nuclear weapon to a certain function.
There is also a software engineering, and somewhat social, aspect to this problem. Even though today's implementation of a function might not require some method of a type class, tomorrow's improved implementation might. Declaring the type signature to be more than minimal is like making a reservation for future demands of functionality in the arguments. Making such a reservation forces all other software calling that function to provide that functionality, even though it isn't currently used. A type signature therefore represents a battle over a property line: who "owns" the right to decide what functionality an object should have?
Construct a go 囲碁 set for which it is easy to separate the stones into white and black after the game.
Marbles of slightly different sizes: sort through a sieve which only lets the smaller size pass through. Board would need indentations to prevent placed marbles from moving around. However then stones would not slide easily for the scoring phase of post game.
The stones are on opposite sides of the density of water. Separate by dumping the board contents into a container of water. Messy and impractical.
Mechanical/electronic device which sorts stones dumped into its hopper by color using (very easy) computer vision. Maybe hand cranked.
Thinking outside the box: every stone has opposite colors on both sides, like Othello. Players draw from a common container during the game; there is no need to sort. This causes problems for scoring systems which assume both players start with exactly the same number of stones, counting remaining stones at the end.
Even more lazy: same as current stones, but place them all in a common container. Choosing the right color to play during the game is easy.
What material to make a projectile out of best survives the sudden acceleration and heat of being fired out of a gun? Shock resistance. Avoid inelastic deformation, though even elastic deformation into vibrational motion also wastes launch energy.
Previously: densest material is best against air resistance. Suppose we ignore air resistance, maybe shooting in the vacuum of space.
Store a compressed partial file in two pieces: one containing all the completed output blocks and one containing the compressed incomplete last block.
Devilish details, partially explored:
There does not seem to be enough rope in zlib to do this efficiently. We either need a way to snapshot and restore the internal compression state to roll it back, or for the compression function to leave unconsumed input in the input buffer when it produces output.
If the the first feature, then we do some binary search at the next to last block for the minimum amount of input that produces the next to last output block. I'm not sure where the output of the flush then goes.
We could serialize the final state and write it out as the uncompressed final block, but the unconsumed input serves the same purpose.
When displaying a collection of objects in random order, e.g., slideshow or musical playlist shuffle, let the randomization be keyed on the time of day. This allows recovering what was shown / played at a point in the past by just recording the time.
The existence of chess symbols in Unicode simplifies the task of writing chess software: no need to provide piece images. It can also allow the user to customize what character to use for each piece on the board, perhaps even permitting letters (of a native language) instead of figurines, which could helpfully matches move notation.
One slight difficulty is that the chess symbols are available in both black and white (as is necessary because they are drawn slightly differently in order to render properly on a white background), but letters are always just black, so need to be rendered in outline for a white letter on a white background. Also need an outline for black letters on a black background, though maybe avoid coloring dark squares black, instead some lighter color.
Scribal abbreviations made the orthography of Latin character based writing systems as complicated as Chinese characters. Inspired by the thousands of Tironian notes.
Assume the two examples demonstrate that written language naturally wants to become as complicated as Chinese characters. A modern example might be the explosion of emoji, including proposals that new emoji be constructed by joining them with zero width joiners. Can we deduce something deep about how the brain processes language by this tendency toward complexity? Possibly useful for programming language design.
Tell a story of the "joke" mutant from X-Men 2 with the ability to change TV channels by blinking. Perhaps he is revealed to be incredibly powerful, able to control all electronics, or media-related or communication-related electronics, or the media itself: all incredibly useful in the modern world.
Create a device, probably an app, which has 4 buttons (perhaps labeled 0 1 2 3) and records the button presses. The user is in charge of developing a code assigning meaning to numbers.
Not providing backspace requires less editing features. Inspired by a purely mechanical typewriter, with no white-out.
Whereas 2 buttons (binary) is sufficient to encode information efficiently, 3 is better for humans because it provides an "escape" value. Then, 4 to compensate for not being able to delete.
With just two buttons, 4 possible chords: left single press, right single press, hold down left then right (release in either order), hold down right then left (release in either order).
De Bruijn sequences are convenient for exploring kerning: //./*/\..*.\**\\ With this character set, the asterisk is hopefully elevated.
Nantucket Cottage Hospital is classified as the only rural hospital in Massachusetts in order to obtain significant federal Medicare benefits for all of Massachusetts. Could a journalist have discovered this sleaziness through transparency mechanisms in government before the filing mistake which caused it to make the news?
Find other similar things. A challenging task for an artificial intelligence agent.
A user delegates managing a personal server to a trusted authority. How can the user trust that the authority has not been compromised, e.g., by the government?
Open source and peer-to-peer attestations that software, especially software updates, can be trusted: web of trust.
The Fahrenheit and Kelvin temperature scales intersect at 574.5875, which is not that unreasonable a cooking temperature, maybe for grilling or for cooking a pizza. (About 300 degrees Celsius.)
"The cookbook told me to cook at 575, but everything came out wrong because I assumed it meant Fahrenheit when it was actually a metric cookbook and temperatures were in Kelvin!"
Painting, or generically 2D art, used to be the only way to record the appearance of a scene or person. It has become obsoleted by photography.
However, digital vector art has an advantage over photography in that it can be rendered at arbitrarily high resolution and not look "wrong" (for some aesthetic measure of "wrong"). Edges remain sharp. At worst, it only looks boring at high magnification.
Photographs look pixelated or weirdly interpolated at high magnification. Fancy interpolation between pixels of a raster photograph could get good results, but it requires computer vision and AI to figure out where the sharp edges should be.
Perhaps someday a hybrid: spline surfaces constructed among rasterized samples in color space on broad regions with sharp vector clipping.
Inspired by, why would anyone commission a portrait these days? Only if it is digital vector art.
A person has sexual boundaries they they don't like having violated, but they also recognize that their boundaries line up with great evils of society, e.g., racism, classism, religious intolerance, bigotry about sexual orientation. While having one's boundaries violated is unpleasant, also unpleasant is exercising the power of complaining about the consent violation, exercising a power that perpetuates and enforces such social injustices: one has become a conduit for evil. Discomfort with exercising such power is magnified if there are institutions which automatically trigger punishment once a consent violation has been determined to have occurred. A person feeling uncomfortable with such institutions would rather not think about sex in terms of boundaries and consent, so perhaps therefore they do not think about sex in those terms.
This makes consent culture tricky: it tries to force people into a box of thinking along the lines of consent and boundaries when some would rather not, and do not, for legitimate reasons.
Another person, believing that their partner would prefer not to think about sex in terms of consent and boundaries, does not ask about consent and boundaries in order to be polite, to respect their values.
Let a cloud data storage service provide a database interface. The user's client software could put a layer over it making it appear like a filesystem, e.g., Dropbox. This is in contrast to cloud services nowadays which provide only a filesystem interface.
Curiously, a filesystem backed by a database is an inversion of the typical implementation on a database server.
Tools like grep and ls could use the database interface to more efficiently process queries. A database could provide useful sophisticated operations for a filesystem, e.g., transactions or multiple versions of a file.
A puzzle program with an element of randomization can encourage people to repeatedly solve the puzzle quickly, recording and comparing times. A chess endgame training program often has many moves which provide the most resistance, or many positions which have the same distance to mate.
Inspired by speedcubing the Rubik's cube.
How much power is there in getting to decide what gets recorded as the official cause of death? It is often a single person who makes the call. Is this power being abused? The power of "the crown". Cause of death statistics are often used to formulate policy, though surreptitiously manipulating such statistics to affect policy will probably eventually become discovered in the ineffectiveness of the policy.
Consider humanity's purpose to be to enrich the knowledge of the universe: we should broadcast information that other alien civilizations might find useful. What should we broadcast?
Probably things that are universal. Science. Things that all civilizations might want to know, but perhaps due to the differences between their world and ours, they might not be able to easily learn or discover directly. Physics: transmit the Standard Model, theories of relativity. Chemistry: certain reactions we might be finding relatively easy to do because of elemental distribution on earth, and earth's ambient temperature and pressure. Astronomy: the WMAP cosmic microwave background anistropy picture looks the same everywhere in the universe. ICRF isn't quite universal, but close.
Probably not so much the intricacies of life and society on earth (though that could possibly be useful for comparative studies with other civilizations).
Non-rewriteable media seems more likely to preserve information for a long time, as people can't reuse the media to store new information. For example, printing on paper or other materials, (modulo bleaching like with the Archimedes palimpsest), CD-ROM, DVD-ROM.
Modify magnetic and solid state storage devices so that they cannot be overwritten. This is probably easiest done with a hardware embedded interlock. Maybe a cryptographic key can unlock it back to read-write mode. This should be fairly easy as these devices have sophisticated hardware in them already.
Inspired by a "Little Library" of a few books someone put in front of their house. People are encouraged to borrow and leave books by the honor system. The honor system works because there isn't much value in stealing a book and (say) attempting to sell the used book. This could also be done with CD-ROMs, but it could not be done with, say, flash drives, because people would just steal them and overwrite the data.
Consider a public policy mindset shift that treats acts of terrorism like graffiti. While efforts can be made to directly thwart it, it is understood that such direct efforts are only merely treating the symptoms of a deeper underlying problem.
In contrast, currently the mindset against terrorism is to ignore the underlying problem, perhaps rationalized by, if we pay attention to it, then the terrorists have won because attention is what they wanted.
Doubly rules surfaces are neat. Consider approximating arbitrary shapes by stitching together segments of planes, hyperbolic paraboloids, and hyperboloids of one sheet.
What happens at the boundaries? Of course, the straight rulings offer stitching boundaries. Can there be a curved boundary between such surfaces that avoids a crease?
In a future with roads populated only by fully automated cars, we imagine cars traveling both directions on all lanes to maximize bandwidth, rapidly and precisely switching lanes to avoid head on collisions.
Depict such a future using computer graphics.
Inspired by a traffic jam going only one direction on a highway, but the other direction nearly empty. Also inspired by the zipper merge, which maximizes bandwidth by using all available area.
Create a calculator powered by human power. Set up a computation, perhaps mechanically, then turn the crank. How sophisticated a calculation can it do, and how many turns of the crank will it need? Probably done by having the crank power an electric generator which powers an electronic computer, though it'll take some engineering for the computer to be able to work on the "dirty" power generated from a crank. Booting up an operating system at the start of a computation is a no go.
Depict superheroes and supervillains fighting among realistic depictions of hugely dangerous astronomical objects, e.g., black holes.
It is hard to imagine men and man-made objects of realistic materials surviving encounters with such objects (inspired by Interstellar), but indestructible superheroes, sure why not?
Science fiction of course invokes spacecraft protected by nearly indestructible "shields".
Musical wind instruments demonstrate that a large amount of information can "naturally" be transmitted through muscles in the mouth. This is especially pronounced in instruments like trumpet, in which the hands and fingers transmit only 3 bits (3 pistons): everything else is done through the embouchure and breath.
Create a digital input device exploiting this capability. Obviously hygiene issues, though that has always been present with wind instruments.
Hypothesize that the severity of PTSD depends on the social structures around a person (probably both before and after the trauma). Inspired vaguely by the Rat Park experiment which linked social effects with psycholgical effects.
Unknown are details of exactly what features of social structures around a person.
PTSD is most famously associated with soldiers returning from the battlefield. It is surprising that the military does not have a detailed model of who gets PTSD more than others, even though there is obviously variation between soldiers, and they have lots of data.
It might be that the people most prone to PTSD unfortunately get recruited into the military: military is often a last resort for those who cannot succeed elsewhere, but success elsewhere is often a result of social structures around a person.
How would baseball be different if the pitcher is permitted a running start before pitching the ball?
Running up the mound might be awkward, though there could be an entire elevated catwalk from second base. Base runners are probably allowed to steal during the run.
Inspired by cricket.
Create software to give the user control of what happens when a battery powered device is unplugged from wall power.
Most aggressive is to immediately turn off the device, essentially disabling the battery. Such a feature may be useful if the device is seized in a raid, and law enforcement attempts to move the device while keeping it powered on. The device powers off in response to the attempt to move it, then requiring, for example, a password on boot to unlock full disk encryption on boot.
These could be defeated by cold boot attacks against memory or devices like the Hotplug, but that requires more effort by the raiders.
Less aggressively, the screen can lock when switching from AC power to battery, again requiring a password to unlock.
There exist USB devices that simulate mouse movement and keyboard activity. They are famously used by law enforcement on computers seized in a raid to prevent them from locking and requiring a password to unlock.
Create software that detects that such a device is being used and locks the screen, or more aggressively, wipes the device.
Objects which have names but could by referred to by number:
Months of the year
Planets. Though theoretically planets could switch orbits or get destroyed.
Chemical elements (many were named before their proton count was known)
Astronomical spectrographic categories
States by statehood date, picking an order for North and South Dakota. There will be gaps if there is ever a successful secession. What should be done if a state splits into two? Or mergers? Probably the older one is the one which has the original state capitol, though the new border could go through the building.
Stages of grief
Inspired by how Japanese month names are just numbers.
Create a filesystem simulator. The motivation was to stress test the integrity of a real filesystem by having a model of ideal behavior to compare it against.
The simulator stores directory structure but not the contents of files. Instead of files, it stores the key or seed used to generate the data (probably random data) in each file. Therefore a large simulated filesystem can be stored in memory.
This should be easy, though there is an unlimited amount of sophistication possible.
There is a key moment in the evolution of life, in particular intelligent life, when it has discovered enough science to understand that it is doomed.
For humans, it is some combination of when we learned that the sun will get hot enough to boil away the oceans, expand to envelop the earth, that protons will (probably) decay, and that the heat death of the universe will occur.
Superman fights (say) General Zod and they continually up their game until they are doing astronomical amounts of damage to the surrounding landscape: e.g., black holes form from the energy of the colliding punches.
Needless to say, the earth is destroyed.
Perhaps fun to depict accurate special effects, inspired by Interstellar.
Permuting the RGB channels, and exploring all combinations of inverting individual channels: Normal=0, inverted=1.
Used the recolor.sh script from this project.
Possibly useful for a deck of cards.
The images look negative for the middle 4 columns, with the second component inverted.
Some scripts that take very large astronomical images, e.g., Veil Nebula, Carina Nebula, Orion Nebula, Tarantula Nebula, and cut them up into smaller pieces. Inspired by: has anyone ever looked at the entirety of these images very closely?
Also, all 90 degree rotations (because astronomy has no notion of which way is up) and permutations and inversions of color channels. These may enhance less noticeable detail.
(Work in progress.)
Cut the image in half orthogonal to the longer side. If a square image, prefer a horizontal cut, inspired by the commonness of images processed in horizontal raster lines top to bottom. If the length is odd, then make the right or bottom half the bigger half: coordinates therefore run 0 <= c < size/2 and ceil(size/2) <= c < size.
Recursively process the two halves, left right or top bottom. Stop at a single pixel.
0 1 2
4 6 7
5 8 9
This is a kd tree with k=2.
Choices made: what to do if square, odd, left first, top first. Create a tool to explore alternate choices, also including the possibilities of the dimensions (factorization) if one does not know it.
The conventional narrative goes, the bourgeoisie put into place barriers against upward mobility of the proletariat. This artificially keeps wages high among the bourgeoisie because of low labor supply. It keeps wages artificially low among the proletariat because of high labor supply, and therefore keeps the cost of servant labor low for the bourgeoisie.
What is the nature of these barriers? They are causing market failure in the labor market.
The barriers must be very secure because the proletariat are constantly probing and attacking it, constantly adapting, seeking to improve their lot in life. Analogies: sea wall, or microorganisms against immune system. I suspect inculcating identity is part of the system: it is something the proletariat (and people in general) cannot easily change.
Another problem with the conventional narrative is that the bourgeoisie are not a monolithic block. They also employ other bourgeoisie. The employer bourgeoisie compete against each other, seeking to minimize labor costs, so if they can expand the labor supply for a task formerly limited only to bourgeoisie, their labor costs will go down. Thus the employer bourgeoisie will want to provide upwards mobility to the proletariat, so will work to dismantle barriers. How do the barriers remain in place and not erode?
Given two Rubik's move sequences, determine whether they are the same. Different ways "the same" can be defined: 24 different orientations of the cube. Or also add reflection. Or discount trivial equivalencies, like RRR=R', opposite faces turning sequentially like RL=LR. TTBBRLRLRLRLTTBB. Or the moves don't matter, only the final effect, the permutation accomplished.
Assuming the Rat Park experiment to be true, then whether a person remains addicted is highly indicative of the social support structures around them. Social support structures are highly indicative of social class. Addiction becomes a visible marker to invisible social class. The observation of discrimination against people with addiction, often of contempt, suggests that Rat Park is true, as confirmed by society.
In chess, a reasonably skilled player (or fan) can instantaneously locate a square specified in algebraic notation, without having to refer to coordinates on the side of the board (or counting coordinates if they aren't written) and finding the intersection. The board seems small enough that each square has its own identity and the coordinates are just names rather than instructions on how to find it. (Perhaps beginners have such a hard time at this that it is a barrier to entry.)
In 19x19 go 囲碁, the board is too large for each intersection to have such an instantly locatable identity, so we consider ways to refer to a specific intersection, especially for things like commentators discussing a game in progress on a demonstration board.
The most obvious, and already commonly done method, is to have coordinates marked along the side of the board. However, because the board is so large, following a coordinate line all the way across the board takes time and is prone to error. (Geometrically, this is an instance of area growing faster than perimeter.) Only the corner regions are quick.
A simple elaboration would for coordinates to be marked on every intersection, similar to beginner's chess boards. Assuming letter plus number, we win because the highest number is 19, and the 1 digit does not take much space in a proportional font. The board does potentially get noisy with a lot of text, especially if each played stone is also annotated with its move number.
Another already commonly done method is for commentators to add marks, e.g., letters, to certain intersections (which may or may not be occupied with a played stone) and refer to the marks in discussion. The marks are landmarks.
With landmarks, we can ambitiously also consider a standardized language for referring to a point a given offset from a landmark. The area reachable by N wazir moves is a diamond shaped region given by OEIS A001844. Exploit approximate 8-way symmetry (understanding that orthogonal and 45 degree diagonal points will have redundant names). 24 letters, perhaps NATO alphabet plus an octant specifier can cover up to 9 wazir moves (A002620). An offset such as a knight's move will always have the same name (but different octant specifiers) in all 8 possible directions.
Inspiration was two commentators commenting on an online go board. Though this could be solved better with software: each commentator could have a different colored pointer.
Play around with overlaying the celestial sphere onto the surface of the earth. 1 meter on earth = 32 milliarcseconds in the sky, which is finer than the resolving power of the Hubble Space Telescope. Depict asterisms "actual size": it'll usually take huge amounts of area.
Inspired by the hoax of an ancient civilization aligning their cities with the stars.
Previously similar. Also xkcd 1276 "Angular size".
If you cite a work in an academic paper, you have a moral obligation to make sure the cited work is easily available, or make it so if not. Your paper depends on the previous paper; without it your paper loses value and meaning. If it weren't for copyright, copy it verbatim and include it as an appendix. Copyright interferes with the moral obligation.
Even with copyright, rewriting is possible.
Let this idea be the basis for modification of copyright law for academic publishing. Frequently cited works will become frequently copied, so preserved.
The eye is most sensitive to photons emitted by stars of the F spectral class. Unsurprisingly, they appear white. F is one category higher than the yellow G class of our Sun.
Not green, which lies between yellow and blue in the rainbow and is the pure frequency to which our eyes are most sensitive. Stars, approximately black body radiators, do not emit pure frequencies but a broad spectrum.
If we can put a spacecraft in orbit around a star with unusually high velocity relative to the center of the galaxy, then it can travel long distances "for free", and have a long-lived source of power. The time scale will still be at least millions of years, so it will have to be a very long lived spacecraft.
Need gravitational slingshots to insert into orbit, or else accelerating (and decelerating) will be just as difficult as achieving the high speed without a star.
Kapteyn's Star is heading away from us, so we would have independently achieve velocity fast enough to catch it. This seems like a bad idea. Barnard's Star is coming towards us.
Tagging along with a hypervelocity star allows exploring beyond the galaxy.
Put a space probe in orbit around a red dwarf star and get "solar" power for a trillion years or more. Watch out for flares and collisions with other orbiting bodies.
For a mission with such a long time scale, it's not so important to get there quickly. Nevertheless, there are quite a few nearby red dwarf stars.
One of the criticisms of the Foldable/Traversable Proposal (FTP) in Haskell is that error messages get more confusing and documentation gets harder to understand. Both of these problems could be addressed with improvements to tools.
Errors when calling a polymorphic function with a Foldable or Traversable context could have additional text repeating what the error message would be if the function were specialized to lists.
Haddock could generate additional documentation for a polymorphic function with Foldable or Traversable context: generate, as documentation, what the type signature would be if the function were specialized to lists. Or, the type variable could be named (renamed) "list":
mapM :: (Traversable list, Monad m) => (a -> m b) -> list a -> m (list a)