Public Service Announcement: Although the pores in most surgical facemasks are sufficiently large that swine flu can easily get through, having such a mask on may still prevent you from contracting it because it hinders you, at least psychologically, from touching your mouth and nose with your hands.
You may wish to consider eye goggles for the same reason.
One popular imagination of the future is "The Future Is Shiny". Create a film of such a future. Let there be shiny clothing and lots of reflective interior and exterior surfaces.
The challenge is of course these shiny surfaces will show lots stuff off the set, including for example the camera. The special effects will have to be ray traced. It could be an animated film -- a very computationally expensive fully ray traced film.
Animation is one solution, though the animators must create a fully immersive detailed world. Render at an incredible high resolution so that the home viewer may do a "Blade Runner" style zoom and video analysis of reflections of a scene to uncover interesting other things going on in the background.
To add on to the long list of "things I don't understand". How can forces be transmitted by a particle? When a magnet attracts another magnet, what are the photons?
Being familiar with only the macroscopic forces of infinite extent, gravity and electromagnetism, how do forces with limited extent (weak and strong) work? How does a force carried by a particle which decays (W and Z particles) work?
Bicycle-powered elevator.
An intriguing possible implementation is a glass elevator on the outside of an "exercise tower". The more you pedal, the better your view.
Poor Fermat. Two of the most famous things named after him turned out to be very wrong. His conjecture that all Fermat numbers are prime is spectacularly wrong: in fact, none of them beyond F4 seem to be prime, and disproving the primality of, and factoring, F5 should have been well within the mathematical techniques he himself discovered.
It seems unlikely that he found an elementary proof of Fermat's Last Theorem, so whatever he thought he had was probably dreadfully wrong.
Child birth seems to have gotten a lot safer for the mother over the past millennium. By how much? Exactly what has changed technologically?
Monochromatic light is always coherent or else a bunch of waves all at the same frequency but different phases would cancel out.
LEDs are therefore not monochromatic (but certainly narrow bandwidth) though I don't understand how that works: it's a fixed electron drop.
Unfortunately, the ability to lie well appears to be a useful life skill. How can we measure and experimentally prove such a hypothesis?
Complicating matters is it's not just about the ability to lie, but knowing when to lie and when not to.
Certainly learning this skill is not explicitly part of the standard primary education curriculum, though the "fringe" does learn it (getting through school by lying and cheating). Should it be?
Of course, detecting whether someone is lying is also a useful life skill.
If the universe were "closed", they say if you looked through a telescope powerful enough, you would see the back of your head. Is this really true?
Geodesics on a sphere certainly loop back on themselves, but is this true for all geodesics on all closed manifolds in all dimensions? I'm guessing not; in fact I would not be surprised if "most" geodesics on "most" manifolds do not loop back on themselves: the sphere is special.
Consider a two-dimensional manifold of a unit square with corresponding opposite edges glued together into a torus. A beam of light fired with an irrational slope ("most" slopes) will not loop back on itself.
For any given point on a closed manifold, does there exist some geodesic which loops back to the point?
Update: There is at least one closed geodesic (Lyusternik and Fet). It's probably not through your given point.
Create a simple animation of a analog clock with continuously sweeping clock hands. Continuously sweeping hands are seen in AC powered analog clocks in the real world: the second hand (and the minute and hour hands) moves continuously as opposed to ticking once a second as seen in other mechanical or battery powered watches and clocks.
The animation should update fast enough for the eye to perceive as continuous.
Use antialiasing. As the hands move, the antialiasing "gray" pixels slowly change in brightness.
An idea for a modern screensaver.
This is a proposal to drastically overhaul the chess title system, which has seen inflation in recent years.
The International Master title is awarded to a player who is within of 120 rating points of the number 5th ranked player. This title approximately corresponds to what is informally called a Super GM today. The International Master title is not a permanent title; it may be gained and lost as ratings change. In fact it is less a title as an abbreviation for a rating.
The number 5 comes from the apocryphal story of the 5 original grandmasters. Also, the number 5 ranking should be more stable than number 1. If the current number 5 dramatically changes in rating, another player with a rating likely similar to the old number 5 rating will become number 5. The value 120 is actually 120.4119982655924780854955579 which when plugged into the Elo formula represents a one-third probability of the lower-rated player defeating the higher rated player. As of April 2009, the threshold rating to be an IM is 2636, or ranked 96 or better. It is understood that the rating system is not perfect, that a player might inflate his or her rating by choosing opponents carefully and playing very conservatively, accepting quick draws against higher or similarly rated opponents. Thus we have the 120-point margin instead of awarding IM to just the top 5.
The title of Grandmaster is bestowed on a player who won 50 games against International Masters, counting at most two against any particular IM, once for each color. The GM title rewards wins only, not draws, and does not penalize losses, so encourages players to play exciting chess and go for the win. We are trying to help the perceived "draw problem" in top-level chess, where many players will take draws or play conservatively because rating points are on the line. The GM title is independent of rating. There may be IMs who do not achieve GM, and there may be GMs who are not IMs. The GM title is permanent. The GM title awards longevity, a lifetime achievement award signifying 50 moments of brilliance against strong opposition. The GM title is prestigious. By tradition, defeating an IM can be called acquiring a GM "norm", but we may also informally call it "collecting a scalp". The number 50 is chosen arbitrarily, and research should be done to tune the number to a reasonable number.
The FIDE Master title (FM) shall have the same requirements of the grandmaster title now.
This ends the "world class" titles. National and regional federations may establish their own titles calibrated to designate high level play relative their local level of play.
These titles can be calculated retrospectively.
A chess match structured that only the loser of the previous game may deviate. It's a theoretical discussion, possibly with plenty of time between games for analysis. It is also "in search of the perfect game".
We describe a state transition system to explain how the result of the previous games affects play in the next.
We refer to a certain previous game as the Deviation Base (DB). The DB may or may not be the immediately previous game. In the first game, the DB is a phantom zeroth game where draw was agreed before the first move.
A game proceeds along the DB until the player with the right to deviate exercises that right; until that point, the other player's moves are forced to be the same as the DB. (Or more practically, the starting position is the point where deviation happens.)
We operate under the dogma that white should win; that is, if a game ends in a draw, white will have the privilege of deviating in the following game. We will also use the shorthand "Black wins" to mean "Black wins or it is a draw".
If the DB result was Black won, and in the current game Black wins, then the DB remains unchanged.
If the DB result was Black won, and in the current game White wins, then the current game becomes the new DB.
If the DB result was White won, and in the current game Black wins, then the current game becomes the new DB.
If the DB result was White won, and in the current game White wins, then the DB remains unchanged.
When the DB remains unchanged, one can imagine it was a failed experiment to improve on the DB in the quest for the perfect game. Since the experiment failed, the game is "discarded", not affecting the DB.
The player with the right to deviate is solely a function of the DB. If the DB result was White won, then Black has the right to deviate in the current game. If the DB result was Black won, then White has the right to deviate.
The same player plays the same color through the whole match. However, in order to make the conditions fair, one can bid for the right to play White. Or one can actually play two parallel matches of this form, with each player playing one color for one match and the other color for the other match. For the purposes of scoring, draws are half a point and are not counted as Black wins, giving Black incentive to play for the win. The parallel match games are alternating interspersed: this additionally gives the back office team more time to analyze.
Team is important. Kind of like NASCAR, the team which does preparation and analysis of the DB is critically important, but so is the "driver" or player who takes over after the preparation is exhausted. This may be a good format for chess team competition.
This might also be a good format for a human (team) versus computer chess match.
If you could ask one question of God, what would you ask?
One sneaky idea is simply the above question itself, to have God look into you and provide you with the question-answer pair that will give you the most satisfaction of having spoken with God.
But let's try to get at the question directly. We want a question that is interesting, and for which the answer is not known. We want some assurance that the answer actually came from a deity, not just a mere mortal. Such an assurance should be encoded within the answer itself; that is, the answer should be self-certifying that it came from a deity. One way to achieve this is if the question is such that the answer is unknown to mankind, but once known, the answer may be verified by mankind. Examples of such questions are in the computational complexity class NP. Not every question in NP is difficult, in fact, "most" are not. We seek a specific NP question for which mankind has expended a great deal of effort, yet in vain, to discover its answer, yet thus far it has been in vain.
We want a question whose answer is significant on its own, not just for proving the existence of a deity.
We want a question for which we are sure exists an answer.
We avoid questions of prophecy because it seems our Universe is designed so that prophecy is in general impossible. Systems for which a classical approximation may be made may have their future predicted by mere mortals. We also wish to avoid prophecies that the a mere mortal can effect its occurrence by force. Prophecy is logically inconsistent with free will.
We want a question that is unambiguous and compact, to provide little room for the presumed deity to weasel a cop-out answer, and if you are suddenly put in a position to ask the question, you can have it memorized. We want the answer to be compact as well, in case there is only limited writing space on the stone tablets or you can only remember so much from your "vision".
Thus, here is my question: What is the second largest factor of the 12th Fermat number? We ask for the second largest instead of the largest to make the answer more compact, without affecting the difficulty of the question.
The answer is less than 593 digits, somewhat of a challenge for a human to memorize in a messianic vision, though you could ask for the answer in the form of a meaningful poem in rhyming iambic pentameter, whose each sentence MD5 hashes to 128 bits of the answer. God, and Bruce Schneier, can invert MD5 to use as a mnemonic.
This question has the bug that it is answerable by mere mortals with quantum computers.
I expect computer chess programmers to find these rules controversial or onerous, or to cry "not fair". But recent results have demonstrated that any traditional "fair" match between a human and a computer to be pretty much a certain victory for the latter. So we need a new computer chess challenge, one which might seem as insurmountable now as the Fredkin prize seemed when it was proposed decades ago, but eventually won by Deep Blue against Kasparov.
The CAPITALIZED WORDS are "free variables" whose values I chose arbitrarily. The match organizer may adjust the values to something more convenient or appropriate.
1. The match shall consist of an unlimited number of games in a SEVEN-day period.
2. The human may not receive any computer assistance during the entire duration of the match, including between games.
3. The human needs to win only one game to win the match.
4. The human may abandon a game at any point with no penalty.
5. The computer must remain unmodified for the duration of the match; that is, the computer must not receive any human assistance during the match.
6. The computer must play deterministically; that is, it must not deviate from any move played in a previous game of the match. In order to keep things moving along, the computer should replay previously played moves immediately.
7. The computer is limited to THIRTY SECONDS per move. There is no accumulation of time from previous moves. (Rationale: A computer may virtually get more time by using more processors, so THIRTY SECONDS should be more enough. A constant time limit per move also avoids tricks like the human trying to induce the computer into time trouble, which is not an interesting way to demonstrate human superiority at chess.)
8. The computer may not ponder. (Rationale: Every move, the human should should face an opponent of constant strength. This avoids the human trying to make the computer play weaker by steering to positions it has not pondered.)
9. The human faces no time control other than the length of the match.
10. The human may propose any position as the starting position of a game, with white to move. The computer shall have ONE MINUTE to decide which color it wishes to play, and if choosing white, to give its first move.
(Rationale: computers are often called upon to analyze positions they did not play into. These days, computers are thought to play certain positions worse than humans, which makes muddy the question of who is ultimately the best chess player. Human-computer games in the past have typically seen the human making suboptimal moves in order to steer the computer to a position it will play poorly. We allow the human to "fast-forward" directly to such positions. If the computer can outplay the human from any position of the human's choosing, then it will conclusively lay to rest who is the better player.
No doubt the arbitrary starting position will result in the human proposing many unusual "chess problem" positions or positions with deep home preparation. The idea is for any problem that has a winning strategy simple enough for a human to memorize, the computer should be able to rediscover it within ONE MINUTE.)
11. If the computer believes a starting position to be equal, it must defer the choice of side to the human.
12. If a game ends in a draw, the human may return to any previous position of the game and switch sides deviating from what the computer played, thus attempting to win from a position that the computer could only manage a draw. (Rationale: because the human must win a game to win the match, the computer might seek to always play for a draw, making suboptimal moves. This rule penalizes such a strategy.)
13. Embargo. This rule is awkward, and a different rule may be substituted for the same effect if one can be found:
The human should be discouraged from having the computer play against itself. Because the human may switch colors and deviate after a draw as explained above, and because the computer may not deviate from any previous game, it may be possible to play a series of games following a draw to achieve the effect of the computer playing against itself. Maybe this can be achieved by the same position achieved two different ways through transposition. If a game reaches a position with the computer to move that the computer has previously never faced, but the same position has been reached in a previous game with the human to move, the computer's move is embargoed (not divulged) for ONE HOUR. The computation time remains THIRTY SECONDS.
14. If the computer discovers it is in a losing position (forced mate), it must resign the game (and consequently the match). (Rationale: Lessen the importance of the human memorizing long endgame sequences. This rule also stresses the computer's mate-detection code which, if incorrect, will cause it to resign non-lost positions.)
15. The human is permitted a separate analysis board.
16. The human player is actually a TWO-human player team who are free to discuss with each other. (Rationale: Teams of people are often much more capable than a single person alone, and the resulting creation, for example, a chess win over a computer, is clearly still a product of human ingenuity.)
For every truncation of the Legendre spherical harmonic expansion of the shape of the earth, what are the most radially positive and negative deviations?
For EGM96, what is the point of greatest deviation? Is it Everest? Or do the Himalayas and Tibetan plateau induce enough of a term that Everest does not deviate too much from it?
You can build a beige box desktop computer out of commodity components, but we have yet to reach that point with smaller devices, even cheap cell phones. If your phone breaks, you cannot replace the broken component and be on your way.
Recharge your various mobile devices with a battery charger powered by an exercise bike. This will give you incentive to actually use the bike and stay in shape.
Given the tremendous popularity of e-mail, text messaging, twitter, and instant messaging, we might consider, or re-consider (in the olden days, it was called a pager) a mobile device that only does text: no voice, no graphics. People seem to prefer reading over listening, and the asynchronicity of texting, not disturbing others by talking on the phone, privacy. Devices and plans should be cheap.
Another old form factor, for a non-mobile device, that could be brought back is the ticker tape machine.
A very new form factor to try is the head's up display Visor. You have short text messages overlaid on your field of view. A remote control -- a Bluetooth keyboard, for example -- lets you type.
Let's say the hero outsmarts the villain in a story. But you can't get out of your head the nagging thought that maybe the hero's victory was a matter of luck rather than cunning, and if the villain had done this instead of that, things would have turned out differently. The way around it is to to allow the audience to explore different options, especially different options by the villain. For realistic movies, the standard way is a game tie-in, though it's a bit weird that the game is played from the antagonist's point of view, while the story was from the protagonist's.
Can one write a generally accessible exciting story based on an abstract game, for example, chess? Certainly very interesting chess problems have been constructed. The protagonist may explore different sidelines (and "resurrect" after failing, Lola Rennt) in the quest for the solution.
The chess game tree is a tree. A filesystem is a tree. It would be an interesting, though probably useless, exercise to create a FUSE module for chess or other extensive form game.
Mix and match:
Syntax: lisp like, C like, Python like
Evaluation model: eager, lazy
Namespace management: C++ namespaces, C files and includes, packages
Higher order functions: nil, Object-oriented, Functional
Type system: simple, hindley-milner, object-oriented
and automatically create your own language and compiler.
The end of the Dresden Chess Olympiad inspired speculation about what would have to transpire for one country or another to win a medal, depending the tiebreaks. This is something like a satisfaction or constraint solving problem. Create a general tool to do this, also trying the find the "most likely" scenario that the tiebreaks could come out in a certain way.
While playing around with Conway's game of life, I got frustrated that gliders only travel in certain orientations (diagonally): in particular the problem that if I fire a gun at a pile a debris, the debris will probably shoot back glider destroying the gun.
But cellular automata is all about playing God, so how about simply designing a universe that does not have this limitation? Then, an interesting series of deductions occurs.
In order for gliders to travel in any direction, we need a continuous universe, not a discrete gridded universe of cells. Then, proceeding naturally based on objects that travel, we make the fundamental objects in our universe point particles. Although it ultimately turns out that there are many things one can do with point particles (to be described later), my initial thought is that point particles can never collide in general position, just always "near miss", so are kind of boring. Which leads to the other formulation of the universe: everything is waves. Interestingly, this is the same wave particle duality of our actual universe.
Waves interact: superpositions of waves generate other waves.
Simulating waves everywhere requires a great amount more computational power than point particles floating around; we need to simulate every point in space: maybe some sort of finite element calcuation. Much like the countless hours wasted on Conway's game of life in the early days of digital computing, we might someday see countless hours spent on quantum computers precisely for this game of playing God in a quantum simulation universe.
Back to particles: particles emit a field depending on their state; particles' state may change depending on what fields they feel, as well as affect their velocity.
Yet another method to try to determine the greatest chess player in history. What is the greatest rating margin the number 1 player has opened up over the number N (N=2,3...10)? We assume that the number N player's ability has remained roughly constant.
The answer is probably Fischer.
What are some features we want in DJ software?
Do not cut the current playing song unless you have lots of confirmation: hold down left-Control, left-Shift, F12 and keypad slash
Allow DJ to attach and actually automatically use useful metadata to songs: start playing from time point, and end at a middle time point (fading out). Entire song volume rescale.
Add Tempo metadata, automatically (Fourier Transform), or manually, and search by tempo.
Ability to handle two sound outputs: (1)headphone out and (2) can be any of PC speaker, additional USB sound device, bluetooth headset.
Ability to crossfade between songs.
Assume cold dark matter does exist in the form of WIMPs: Weakly Interacting Massive Particles.
WIMPs feel the force of gravity, so they should become trapped in gravity wells such as the earth or other massive bodies.
How much of the earth's mass is composed of dark matter?
One way to experimentally confirm the existence of dark matter is to measure the gravitational mass of the earth using any of the standard methods, and compare it with the mass obtained by an alternate method assuming the earth is completely composed of regular matter.
Measuring the seismic propagation of earthquakes gives us a density profile of the earth. Integrating the density gives us the total amount of regular matter in the Earth. WIMP matter does not affect the seismic propagation of earthquakes.
The earth's active internal structure might be too complicated to determine its absolute density profile, in which case other astronomical bodies are available. We could go to the moon, which is geologically dead, and set off explosives. For a small asteroid, we could drill a core sample all the way to its core. Neutron stars undergo starquakes which might reveal its internal structure. Observing and modeling the Sun may determine how much of the Sun is made of regular matter.
The dark matter surface of the earth need not coincide with the regular matter surface. If the dark matter surface is higher than the ground we see, then satellites would orbit with unexpected periods depending on how much dark matter was below the satellite. This probably would have been observed already (elliptical orbits would behave very weird) so the dark matter surface is underground.
It might be too difficult to decouple gravitational effects from the seismic density profile to get an absolute density profile. If so, we can obtain a pure gravitational density profile to compare it against.
One can directly measure the gravitational density profile of the earth by shooting artificial or natural neutrinos through the earth and measuring how much they deflect based on how much earth they passed through. Supernova neutrinos are conveniently all parallel. Of course, neutrinos are horrendously difficult to detect, and measuring their angle is even more difficult.
There may be other ways of measuring the earth's internal gravitational mass distribution: satellites perhaps, or somehow use the Earth-Moon dipole.
A test mass at the center of the earth is weightless. Does a clock at the center of the earth experience no gravitational time dilation due to general relativity? If so, what layer of the earth, at what depth, experiences the greatest time dilation? What if the planet's density is not uniform (but still spherically symmetric) like Jupiter?
A Newtonian approximation will probably do fine. Given the radial density profile, calculate the strength of gravity at each point. For uniform density, mass increases by the cube, but gravity decreases by the inverse square, so the force increases linearly, so the maximum time dilation is at the surface.
After making a large number of untrue simplifying assumptions, if you were to feed a 100 minute film directly into a film scanner, and scan it at 1080 lines per frame, you would end up with a very tall and skinny 1,920 by 155,520,000 pixel image. 1080*24*60*100. About 300 gigapixels total. 0.9 terabyte uncompressed.
High-speed, high-resolution, low-light photography or video is difficult because there are very few photons. But astronomers have found the solution to be to use a large primary lens or (better yet) mirror.
Construct a camera with a very large light-gathering surface. No doubt it will be awkward and expensive. I don't know if salvaged astronomical mirrors will work because they might have been designed to focus on infinity. I haven't calculated how large a mirror must be to (say) take bright-as-day photographs or video illuminated by only starlight. It might be a billion dollar camera.
What will these images look like? The world has probably never seen high-resolution low-light video and as such we might intuitively (but wrongly) believe it to be impossible.
Similar in idea to Strobe Party.
Two knights versus pawn.
Troitzky Line
Write the smallest program that will play this endgame correctly. The size of "small" includes external data files and RAM allocated.
An exercise in knowledge compression or data compression.
3 measures of life expectancy.
Mean life expectancy at birth. USA males 74.83. (2004 Social Security Life Table)
Twice the age at which the remaining mean life expectancy equals the years you have lived. 2*38=76
Median age of death. 78
Examine carefully the causes of death for infants, children, and teenagers, for these are the deaths that evolution cares about: their DNA has (probably) failed to be passed on to the next generation.
Infant deaths account for the majority of them, but I don't know much about them.
The top causes of deaths among teenagers are accidents, suicide, and homicide. Certainly there is a genetic component to depression and subsequent suicidal behavior. Many accidents are car accidents caused by drunken driving -- there is probably a genetic component of how much alcohol impairs one's ability to drive. I can only imagine outlandish scenarios where one's genes affect whether one will survive or avoid a homicide attempt. Even if we did evolve to survive handgun gunshot wounds, technology would probably keep pace to develop more deadly weapons.
Certainly the causes of teenage deaths paint an interesting prediction of future human evolution.
HIV may be transmitted through birth and breast feeding, and so AIDS might be (I don't know) a leading cause of death before the child grows up. Also even an uninfected child is less likely to survive if both parents die from AIDS. Sub-Saharan Africa might be evolving HIV-resistant people.
Or the AIDS virus might evolve to remain dormant in children until after puberty.
A picture, or a short film, of a person chowing down on a bowl of glucose seemingly oblivious of the smorgasbord of other very tasty foods around him, until the glucose is thoroughly licked clean.
The joke is that yeast and E. coli exhibit this behavior.
Inspired by "Base Delta Zero", consider the task of even more thoroughly destroying life on a planet including "boiling" away the oceans so they don't come back via precipitation. The key is hydrogen (aptly Greek for "source of water"). After breaking water down to hydrogen and oxygen, the hydrogen will naturally evaporate into space never to return because its thermodynamic velocity exceeds the Earth's escape velocity (I think).
Electrolysis is the straightforward way. Some gut bacteria can produce hydrogen.
Energy computation.
Most quartz clocks and watches have an accuracy or stability of plus or minus 15 seconds a month. (About 1 part in 200000). In order to maintain the correct time to the second or tenth of a second, the watch needs to be synchronized every day or so.
I want a clock that needs to be synchronized less than once a year: the goal is to set your clock with the time ball (the way it's supposed to be done!) at midnight on New Year's and have the clock still be accurate enough to do the countdown one year later. 1 in a billion. Not quite atomic clock accuracy? Can it be done with quartz?
You also want the clock not to lose power over the year.
It's already been done to compare chess players by amount of mistakes they make using a computer as a (questionable) gold standard.
Now that we have a baseline level of mistakes for each player, we can next ask, how many additional or fewer mistakes does a player induce in his or her opponent compared to the opponent's baseline, averaged over all opponents of a player?
These days one can learn an awful lot about a person simply by searching the web for their name. A person with a very common name can hide among thousands of others with the same name; someone with an uncommon name has nowhere to hide.
It's probably inevitable that everyone has done something at some point in their lives that they wish to hide from someone. Consequently, we hypothesize that those who are more successful at hiding undesirable history are more successful in life.
Two experiments: 1. Study a large group of people over many years to see how starting salary, income, lifetime earning, or some other measure of quality of life correlate with commonness of their name. 2. Do a more directed study of selected people who actually do have something to hide and again correlate name commonness versus success in life.
By giving your baby boy the gift of the name Michael Johnson, just how many thousands of dollars of lifetime income are you giving him?
Or one can choose a name whose web search results will be dominated by someone famous.
Flat screens these days don't suffer burn in. We want a different kind of screensaver. Like a clock. Live information.
Do food allergies go away if your immune system is damaged or destroyed due to disease or cancer therapy?
So the doctor says to you, "I've got some good news and some bad news. The good news is, you can eat peanuts safely again. The bad news is, you've got AIDS."
Due to the time dilating effects of the gravity of the Sun (as predicted by General Relativity), how many seconds (or years?) slower is a clock on Mercury than on Earth, accumulated over the age of the Solar System?
How much time has the surface of the Earth lost compared to a clock in deep space far from any galaxies due to the gravitational time dilation of the Earth itself, the Sun, and the rest of the mass of the Milky Way galaxy? How much does each account for?
HT: Meg
Every once in a while, a crackpot idea ends up being right. How frequently is "every once in a while"? What is the probability?
I'm guessing about one in a billion. Copernicus.
An art project.
Your standard dance club party, except with attendees who have consented to be photographed.
The strobe light goes off every once in a while (it's probably too annoying for the party to continuously be lit only by the strobe, though maybe not). Cameras are synchronized with the strobe, and take a photo for each flash of the strobe.
With a strobe rate of 5 Hz, this is 300 photographs per minute, or 18000 photographs per hour. Specialized equipment might be necessary to handle the rate.
Strobes are bright, so they allow the camera to collect lots of photons and consequently photograph in high resolution.
The end result is a millisecond-by-millisecond photographic record of an event that normally takes place in the dark.
