Dark matter is more than a "cop out". If dark matter only explained one thing, it might be a cop-out, but it simultaneously explains observations in cosmology, observations of galactic rotation curves, and large-scale structure formation -- all independent phenomena. This is a nontrivial accomplishment.
It is a non-trivial accomplishment, but a math problem has been solved, not a physics problem. The numbers line up, but we have a big missing hole which we can't detect.
It is also not true that no particle theorists have come up with anything which could be dark matter. Quite the opposite -- they have too many theories. The leading candidates are axions (thought to be needed to solve problems with CP symmetry in QCD) and supersymmetric partners. Experiments are underway to try to detect these.
All of the theories fall into two catagories: not enough of them around to account for the dark matter, or they are theorized but undiscovered. Axions fall into the latter catagories.
I'm an experimentalist, and there just isn't enough ground to stand on here. We haven't seen dark matter. I'll believe it once we've seen it.
Dark matter is the "new aether"? You could have said the same to people who proposed that an unseen body (Neptune) was perturbing the orbit of Uranus. But people looked, and they eventually found that there really was something there.
Beautiful example! I couldn't have thought of a better one! Le Verrier was the one who did the calculation to discover Neptune. With that grand success, he did the same thing for the perturbations in Mercury's orbit and named the predicted planet "Vulcan". There were many false observations of Vulcan before Einstein explained Mercury's orbit with relativity. Le Verrier's calculations were about equally sketchy in both cases, and both yielded observations confirming the existence of a planet. The difference? He was right the first time, and wrong the second.
And this was based on only one kind of observation, the orbit of Uranus, not cosmological observations plus galactic rotation observations plus galaxy cluster observations, all of which are explained by dark matter.
The mathematical explanation is there, but we don't have the experimental evidence, and it's embarrassing. It bugs me when it's presented as "Nothing to see here, it's Dark Matter, we've got the problem solved, 100%", which is how the press presents it. At my school, amongst the physicists, we grin sheepishly over not having a better or more complete answer on this one.
There is still the possibility of an alternate explanation to come out. In fact, what I was saying, since the dark matter correction has been so successful, this points to the fact that all these phenomena will likely be explained using the same mechanism whether it ends up being dark matter or not.
Another case in point, Copernicus' heliocentric system still used epicycles. He believed firmly in perfect circular motion, and put all the planets in perfectly circular orbits around the sun. Then, to correct for the deviations which we now know come from the fact that orbits are elliptical, he put each planet (except earth) on a small epicycle. Both Copernicus' and Ptolemy's systems had a 17 year glitch, where every 17 years, the whole system would significantly mispredict planetary locations. However, had Copernicus been "more Copernican" and put an epicycle on earth's orbit, the glitch would have gone away. The point being, do you believe in elliptical orbits, or perfect circular motion with two cycles, a great circle and an epicycle? In the 17th and 18th centuries, you wouldn't have been able to tell the difference had Copernicus gotten earth right.
To get to my comment about dark matter being the "new æther": at the end of the 19th centur
By the way, if you're unable to put together a keyboard layout from memory, I suggest taking a couple of quick pictures of your keyboard with a digital camera - at least that way you won't be left wondering which key goes where.
IAAP (I am a physicist), and the gravitationally stable spheres is a problem. That's what this study was looking at.
The bright, visible, normal matter forms into a disk in every galaxy we see. This cannot be explained with Newtonian gravity (or Einsteinian, for that matter). You see, when you just stick the normal matter in a simulation to check the evolution of a galaxy, it doesn't stay in the disk shape. To get the simulations to work (meaning, predict disk galaxies), you have to put a spherical halo of dark matter around the galaxy. With the dark matter there, it works perfectly.
The other option is that we're completely screwed up in terms of our beliefs about gravity. However, we haven't gotten any clean results using alternate formulations of gravity.
So the answer is, to the best of my knowledge, galaxies are disks because they either have (a) halos of dark matter or (b) our formulation of gravity is extremely flawed. Both of these answers are the current big embarrassments in physics. Both are bad answers. I haven't seen an alternate to gravity that works, and dark matter is a cop out.
From looking at the article, it looks like they've done a huge simulation taking the dark matter halo bit to the extreme and finding that it correctly predicts a flat arrangement for the satelite galaxies.
This is not proof of dark matter, it merely shows that it is the appropriate adjustment to gravity to explain some of the phenomena in the sky. What this actually means is that the correct answer for "why are galaxies disks?" will likely be the same answer for the pancake mystery. So, the authors did not actually solve the problem, they just showed that the current, in vogue kludge works for the pancake mystery too.
I'm, personally, quite sick of hearing it held up as an explanation as if it is the end all be all of fixing cosmology. It is a stupid cop-out with nothing to back it up. There is no experimental evidence for it. We have yet to find anything that could be dark matter (and there are people looking very hard to find it). Other theories don't have anything in them which could be dark matter (no particle theorist has come up with anything, and certainly not the high energy experimentalists). It is a big ol' strap of duct tape to fix a gaping hole in cosmology. There is nothing backing it up. Quite frankly, it's starting to sound like the Æther to me. And we all remember where that got us.
P.S. I see this study was done at my alma-matter, the University of Washington. I wonder if my old roommate Jim Oliver might have been affected, since he did handstands from our 7th floor balcony railing - maybe he should have been wearing a tin-foil hat?;-)
Ever tried measuring the mass of a neutrino? Yeah, it's pretty damn hard. The easiest way to do that is to watch the beta decay of the tritium nucleus and make precision measurements of the products' and reactants' momentum. The only one you can't really measure is the neutrino itself, but a certain percentage of the decays give you data that you can reduce such that you can plug the results into an equation for the neutrino mass.
There's a problem with this equation. The mass of the tritium nucleus and the helium-3 nucleus need to be subtracted off to get the neutrino mass. The upper bound on the neutrino mass is 2eV. Both those nuclei are approximately 3GeV. To get the error to work out, you need the nuclear masses of the tritium and helium accurate to the parts per trillion range.
In other words, before you can measure the neutrino mass, someone needs to push as hard as possible for perfect measurements of the tritium and helium-3 nuclear masses.
I like it when you define hbar = c = 1 (Plank's Constant over 2*pi, instead of being in energy/Hz, it's in energy*sec/radian). Once you've done this, because of E = mc^2, mass has units of energy, and then time and distance both have units of 1/energy. Neat, eh? It's the natural unit system of the universe. It's pretty inconvenient unless you do particle physics, however.
Basically, the idea is once you've got c, hbar, and the three basic unit types (mass, length, time), you've got a complete system of units. c and hbar then turn into conversion factors which can be set arbitrarily to get a system of units--this is what actually makes the Plank's constant elegant, by nailing it down as an exact number, we've completely defined our system of units (c is already a defined quantity). Then, more accurate measurements of hbar are not measurements, so much as refinements of the definitions of kilograms, seconds and meters.
There is. I'm running Slackware 10.0 (can't wait to get my 10.1 disks! w00t!). I use vim+LaTeX for all my document prep needs on the command line (use an xterm if you so choose). As others have mentioned, LyX isn't bad.
I'm a physics major at UW, so I do a decent bit of scientific work on my computer. I use GNUPlot, XFig and the Gimp to generate drawings for lab reports and whatnot.
I'm typing this from Firefox running on the Xorg6.7.0 server+WindowMaker 0.91. The key here is to use a lightweight window manager. Blackbox and fluxbox are other good choices (light, usable, not fugly (cough, fvwm, cough)). If you have to have that desktop environment, go with Xfce.
The only gap that I occasionally feel in my user experience is a good spreadsheet. I haven't found one. KSpread, OpenOffice Calc and Gnumeric either are or require the use of heavy GUI software which we are trying to avoid (KDE and Gnome are not as big as XP, but far too big to run comfortably on my system). I've glanced at Siag, but haven't really tried it out (I don't know scheme and don't have the time to figure it out right now--see physics undergraduate work).
I use mutt or pine, depending on which email address I'm checking. Thunderbird looks promising for being light and good, if you want a GUI based email client.
Recompile your kernel to match your hardware (trim the fat and optimize for your processors), and turn off any extra servers that you don't need (don't need telnetd, ftpd, &c. running? Turn off inetd--it's also more secure). Customize your boot sequence to only start and load that which your system needs and those things which you use.
I also boot to the command line and don't run xdm or the like. I do a lot of work from the command line, and X+light WM doesn't take long to start. It is, again, one less thing wasting clock cycles on my machine.
For reference, I'm running my Slack 10 system on an Abit BP6 with two PIII 866MHz processors underclocked to 650MHz (long story... Has to do with the fact that the BP6 doesn't technically support the PIII). I've got 384MB of RAM and a GF4 video card. It is lightning fast. The only exception to this is when I'm running X with the closed nVidia drivers (damn thing has a 3MB kernel module... grrr...), but that only adds a hang of a couple seconds when switching between X and the consoles, and that's it. If I'm not playing Quake or dealing with 3D visualization stuff, I can use the OSS driver (2D accel only), and get rid of even that performance problem.
So, yes, the middle ground is there, and it rocks. My computing experience is awesome, my slightly dated hardware is rock solid and perfectly responsive. Take a good, customizable Linux distribution, run light weight software, turn off stuff in the background and run a lean, mean, customized kernel, and you'll reclaim those lost cycles as interface responciveness. I suggest Slackware for this. FreeBSD, Debian, and any other Linux distro which is aimed at power-users will be good for setting up a configuration like this.
Mandrake, RHAT (RHEL & Fedora), SuSE and any other user-friendly type distro is ill-suited to this, IMO. Not that you can't, but my experience with these distros and their high-level admin tools is that if you try to do something too different from the default, it gets extra hard. So, Slackware and the like just end up being simpler, and now you know what Slack users mean when they say "it's simple." So stop giving us funny looks when we say it.
Jeff
Re:Graphics card's driver must support the game?!?
on
Does Linux Have Game?
·
· Score: 1
The ATI drivers for linux suck. ATI doesn't release the spec for the FOSS community to write our own drivers, and ATI's binary drivers are terrible.
I've been looking at upgrading my video card (I want to run tenebrae on my box, but don't have the horsepower in my video card. I've decided against going with ATI because fo this article and this comparison of the available ATI drivers (the FOSS dri drivers, the ATI binary drivers and the XiG Xserver). All three fare pretty poorly in different ways, and it's just a mess. My impression is that things have not gotten any better.
Basically, the ATI drivers don't present a problem for testing under Windows, but under Linux it is a mess best left alone, simply because all three implementations are incomplete or underperform in their own special ways. It's an intractable problem for the game vendors, and I think the best route for them is to make everything work in standard OpenGL, and check it with nVidia hardware. If it works there, they've got it. Gamers using ATI cards with linux are out of luck because of ATI's stance on specs and linux support.
So, your problem is that you assume ATI has written complete drivers for linux.
Firstly, I'd have to assume that should the batteries have no charge, the gasoline engine would kick in, defeating your "no impact" hypothesis. Similarly, I'd imagine that when you're out of gasoline, the electric would act as a failsafe. I haven't had the opportunity to test these out on a hybrid, but it certainly would make sense.
Nope, this doesn't defeat the argument, because when the charge drops below a certain threshhold, the gas engine kicks in to charge them no matter what---independent of whether or not the car is stopped. So, the fact that the car is at a stoplight doesn't matter. (The engine only runs at the optimum power output to charge the batteries, so if does it at the stoplight or while the car is moving, it doesn't change the total amount of gas converted to electricity converted to kinetic energy efficiency for the trip.)
Anything that can be done to a normal car to make it more efficient can be done to a hybrid to make it more efficient. Toyota make the some of the most efficient gasoline cars in the world---their Prius is right in line with their high standard of engineering (IIRC, Toyota has the highest number of cars/trucks with greater than 30mpg).
Comparing a Diesel car to a hybrid is slightly apples to oranges. (Diesel engies are more efficient that spark engines for various reasons, so yes, you're buddy's Jetta does fare reasonably well against a Prius, but that's only because the Prius is still running a spark engine). If VW went and designed a Diesel hybrid, it would smoke your buddy's Jetta for mileage. Here's why: a normal car with a reciprocating engine can only get at most 20% efficiency. A hybrid scheme allows you to design the engine to a lower power output and only run it at its optimum power output (to get more you run engine+motor, to get less you just run motor, you design the charger such that the engine runs at optimum power output). So, you can get 30% efficiency out of the reciprocating engine or a little more. So, a Diesel hybrid against a normal Diesel would get the better mileage.
The parent is right that the cars only get their energy from gasoline. However, there is a deeper efficiency story here that just isn't as quick and easy a sound bite as "getting your car's energy from two sources---gas and electricity"---which is an awful marketroid half-truth.
Anyway, to clarify on hybrids:
The efficiency story goes like this: your normally car engine sucks as far as efficiency is concerned. This is because they have to operate over a wide range of speed and power requirements. Eg. from just taking off from a dead stop to running up a hill at 70mph or more. A spark engine can get to be about 30% efficient (this is from memory, it might even be up to 40%, I'd have to go look it up, and I'm feeling lazy;). But anyway, the point is, that because of the requirements put on the spark engine in a car, it has to be designed for maximum power output, and this means for most driving the highest efficiency a car engine can attain is 20% (this would be for a tiny Japanese car which is engineered for fuel efficiency, like a Toyota Tercel or Honda Civic, other cars, like most SUVs, perform worse).
The trick with the hybrids like the Prius is that they have the battery+electric motor to supplement the gas engine. So, the designers can do something important: they can pick a median power output (much below maximum required power output), and design the spark engine to be maximally efficient for that power output. This allows them to get the 30-40% efficiency out of the gas engine mentioned above. The hybrid only ever runs the gas engine at this power output. If this is too much, the electric motor run the wheels. If this is too little, the motor and the engine drive the wheels. If the batteries are getting low, the gas engine drives the electric motor to charge the batteries. When braking, part of the axel motion is used to drive the electric motor and charge the batteries (reclaiming some of the energy already expended to be reused---this is the regenerative braking that others have mentioned). Note: the designers at Toyota and Honda have taken advantage of the fact that an electric motor and generator are merely the same device, which one it functions as depends on which end the energy comes in, so there is no separate generator. (And if it occurs to you that the clutching system would be complicated because of this, you're right.)
As far as being able to charge up your hybrid, there are some experimental models with that feature. You might eventually be able to do that; so if you just drive around town, you'd only rarely have to fill your tank (however, this feature requires that the bank of batteries is bigger, and 50% of the electricity in the US comes from coal, so the pollution/energy expenditure could end up being worse off the wall charge, depending on where your power comes from;).
By your own admission you have no commercial software on your machine, so this is quite obvious. There really isn't any commercial software for Linux (at least, I haven't seen any), and the GPL, or rather the philosophy that created it, is at least partly to blame.
Yes, I don't use commercial software. About the closest thing to commercial software that I have on my machine is some commercial content in the form of the Quake 1 data files run on the Quakeforge engine. I, personally, have no need whatsoever of commercial software. I have never seen a commercial software package that is particularly enticing, particularly nothing that is used for productivity. You haven't seen any open source code you would use, and I haven't seen any closed source code I would use.
The philosophy that you yourself promote by believing in "open source world and the closed source world". The philosophy that all software should be free, in both meanings of the world, and that the only way to make money from software should be to babysit people who don't have enough brains to use it.
I never said I think software should be free in all senses of the word. In fact, I believe quite the opposite. It is perfectly fine to make money off of software, you just need to create a well engineered product that actually works for a fair price. I'm running Slackware v10.0, and I have a subscription to their distro (I've signed up to buy a copy of every release of their system for about $25 a pop, it costs them less than $10 to actually press and ship the CDs). I will pay for good software. I will not pay for buggy software which I cannot modify. The GPL protects me as the end user, and the quality of the programs available for linux is excellent by my standards.
The problem with this is obviously that if nannies make money and programmers make none, eventually there will be no more programmers. Sure, there will always be a few hobbyists, but you have to realize that they can exist only as long as there is free time. By eliminating programmer jobs you'll force them into nanny jobs instead, which have much longer hours and much lower pay. Eventually nobody will have the time to write free software; they'll be too busy earning enough to eat and pay rent. It is already happening, mind you, except that we still have the reservoirs of cheap labour in India and China.
This is an economics argument which involves a runaway process. The open source movement is doing two things. First, it is growing, and I have watched it do so since about mid 1998. Second, it has a critical mass of programmers already---there are fully usuable (I believe more usable) systems out there. My computer has never been touched by a closed system.
As far as your runaway process is concerned, the argument may well be meaningless. Economics is a chaotic system where the particles (the people) have very strong interactions and, in today's world, long distance communication. We have only a very crude understanding of this system, and their are a lot of complex feedback loops in the system which we don't know about, and you certainly aren't taking even potential known ones into account. The system will come to a ballance somewhere, with people figuring out how to take advantage of the most efficient production systems. In other words, I'm using the hated argument of "market forces will make everything work out ok" to shoot you down, or rather, I'm saying that over a long time scale, permanent runaway scenarios are silly doom and gloom arguments, and that we'll end up producing software in the most efficient manner possible---some of us happen to think that is open source, and for it to maintain itself as a force, we need licenses like the GPL. We don't need commercial software to get by, because if we're right, our own software will eventually outdo any commercial offerings.
But if I am selling the word processor which links to the spellchecker, I am selling my own work (remember, the spellchecker is freely distributable, has the source code in the public domain, and its price is obviously not included in that of the word processor), and I don't see how I can be legally required to do anything at all for the spellchecker authors. Which is why I call this "theft", perhaps figuratively, because the "public domain" "steals" my code. My code, which contains no other code but mine.
The source code is not public domain. The source code is licensed under the GPL. The source code may be publicly available, but it is not public domain. I will assume from here on out that you really meant publicly available.
This is, admittedly, an ugly borderline case, but the way I see it, is by linking to the spellchecker code---even dynamicall---you are incorporating that functionality into your program, and from the user's perspective, once your wordprocessor and the spellchecker have been downloaded and installed (possibly compiled in the case of the checker), the experience of the end user is that the spellchecker is a part of your program. The underlying technical bits aside (which I know quite well, thankyou, my three most used programs are vim, pdfLaTeX and gcc), on the outside, you've incorporated the functionality of their program into yours---you may not have touched any of their code, but from the users perspective, you've incorporated it. The spellchecker authors are licensing (again, copyright is still reserved, they have not given anything up to the public domain) their code for use with other open source programs only. They are not licensing it out so that you can use it with your closed source wordprocessor. They are, probably very willfully, pushing an agenda.
Yes it does. It forces me to license my code under GPL. Yes, I agree that I am not forced to make use of any GPL code, but that does not change the immorality of the action. I don't have to go wandering in a bad neighborhood in the middle of the night, and by not doing it I can avoid getting robbed. But is the robbery any more virtuous because I voluntarily placed myself in an environment where I was certain to be robbed?
That is a terrible analogy. There is nothing immoral about the GPL and its requirements. The GPL is no less immoral than your closed source program. If I closed sourced the spell-checker you wanted to use, you couldn't use it without negotiating with me---and I'd probably make you pay for it. By your analogy, that would also be tantamount to robbery, because your underlying assumption is that if you want to use my spell-checker for free (monetary) you should be able to. So, by your logic, your closed source wordprocessor is immoral, because I can't use it for free (monetary) (I assume this, because you talk of making money off of closed source software).
By using the GPL, I force nothing on you. You've already admitted to that. However, I have not declared that you can use my code. Instead, I've declared the terms on which you may use my code---much like you would for your wordprocessor. But even beyond that, I've defined my terms in a clear, standard manner, and you know two things. One, what exactly the conditions for using my code are, and two, that I'm not likely to budge. Compared to the conditions set out in most closed source EULA's, I think by using the GPL I'm being much more fair. If you're a programmer, you've gotten a much larger advantage---the code itself. It is just defining terms. It's legal and just as moral as closed source---if not more so. Because remember, I'm not saying everyone has the rights to my code (that would be public domain or a BSD like license), I'm saying on what terms you can use my code. The price may not be money, but it's still there---much like in closed source. Quit whining that you can't play the game and make up your own rules.
Exactly my point. And what it means is that no one wishing to make money from his code will ever touch GPL code.
RedHat just announced a profit for third quarter 2004. If you want to play the game, you play by all the rules, or none at all. There are some projects---like PNG---who picked their license such that it could be used in closed source software. In the case of PNG, I would say this was a good idea, because that allowed it to become standard across all platforms, and thus usable on the web as a replacement for the GIF format. The types of licenses you say are better are good for projects which interface between the open source world and the closed source world.
They'll just write their own. It's not that hard. Hello 12 different libpng versions! Hello 24 variants of mozilla! Hello "out of memory" and "insufficient disk space"! Hello "incompatible library version detected"! And how did you say this benefit free software?
However, there aren't 24 variants of the libraries I use. This isn't a problem, and that is why your argument is silly. GPL components belong in open source, and they form the core of the Free Sofware/Open Source ecosystem. The LGLP type code only belongs at the interface between this ecosystem and the closed source ecosystem (while it lasts). Again, you want to use open source? You should go all the way. It works better. All the components of my system are Free Software, and it works great.
Play the game, or don't. There isn't a middle ground, and I see no reason for one.
Instead of squawking away out on the internet and probably getting ignored, the FSFE gets to squawk directly at them. I see it as a question of how ignorable we are.
On those grades... I'm a University of Washington student (physics major). UW admissions policy will deduct or add to your HS GPA based on the quality of your HS. I went to Olympia HS (in Olympia, WA), because of this, I got +.2 added to my HS GPA for admissions purposes. The cross town rival, Capitol HS, has a similar problem to your HS -- they have an extremely high number of students who graduate with a 4.0. While their academics are fine, their easy grading practices cause UW to deduct points from the GPA's of Capitol HS's students for admission purposes. The high schools are really doing a disservice to students when they grade like that. Colleges often take it into account, and all the students from those high schools get hurt. I would even say that my high school was too easy with its grading policies, but then again, I fully agree with putting a 1 hour time limit on a 2 hour test, and putting the scores on a bell curve.
A quicker effect is to take a pair of tinted ski goggles and put them on. After a while your eyes will adjust, and everything will look normal instead of tinted. It's really dramatic with orange tinted goggles (those are what I wear skiing). Everything looks blue afterwards. I would think that this is a similar effect. It seems our brains are very good at making the world around us look the way it is ``supposed'' to.
Bought a harddrive lately? This is one of the bigger reasons. Mibibytes are bigger than Megabytes, so if a Harddrive manufacturer is advertising, they advertise in MB = 10^6 bytes, as opposed to the 2^10 bytes, which is the common accepted definition of MB, and since they are technically correct, you get a smaller drive than you think you are buying. *shrug* Not a major sin, but it seems cheap and sleezy to me. I'd like a clearer definition, personally. Seems like an ok fix for pure, proper technical applications. It's a pretty safe bet MiB won't make it into mainstream use, except as a reference to a movie.
rot13Char c
| isUpper c = chr ((((ord c) - 52) `mod` 26) + 65)
| isLower c = chr ((((ord c) - 84) `mod` 26) + 97)
| otherwise = c
It's in Haskell, and it successfully decodes the above post. Along with anything else using any form of rot13 encryption (eg, double rot13 and quadruple rot13 are common. All that is required is to run the cyphertext through the above program 2 or 4 times).
In this case, he was in the country when Adobe decided to press charges. That actually has a long history in this country. There is a law (I wish I could remember the name of it) which states that if a crime is committed, and the criminal and the victim step into an area under US jurisdiction, we can arrest the criminal and run them through our court system. This has been on the books for just about the entire time the US has been a country. It fell into dissuse about 150 years ago, and only got dredged up in the 80's when a man who was tortured in a foriegn country immigrated to the US. His torturer visited the US, was arrested and tried. I don't remember the outcome of this trial.
I'm really kicking myself for not having my sources (nor the time to go look it up), but it's there. I actually found out about this while helping Amnesty International gather signatures for an Internation Right to Know act (which would require companies to report about their labor and environmental practices abroad, like they have to do in the US). Basically, with the reporting could potentially come prosecution of companies breaking US law abroad. So, this law can be a Good Thing.
What to do about the DMCA is another thing entirely. Personally, I'm thinking it might be high time to abuse the law in the opposite direction. Say create a PDF, post it with a licence and sue Adobe for creating a program which breaks your copy protection (PDF being the copy protection). Now, I'd want to go in on this with a better understanding of the DMCA (to properly abuse it) and a couple friendly lawyers (there are cool lawyers out there, I've met a few). Just something inspired by another thread under this article.
Slashdot Mirror
Say it with me! Santorum is a FLIP-FLOPPER!
Jeff
It is a non-trivial accomplishment, but a math problem has been solved, not a physics problem. The numbers line up, but we have a big missing hole which we can't detect.
All of the theories fall into two catagories: not enough of them around to account for the dark matter, or they are theorized but undiscovered. Axions fall into the latter catagories.
I'm an experimentalist, and there just isn't enough ground to stand on here. We haven't seen dark matter. I'll believe it once we've seen it.
Beautiful example! I couldn't have thought of a better one! Le Verrier was the one who did the calculation to discover Neptune. With that grand success, he did the same thing for the perturbations in Mercury's orbit and named the predicted planet "Vulcan". There were many false observations of Vulcan before Einstein explained Mercury's orbit with relativity. Le Verrier's calculations were about equally sketchy in both cases, and both yielded observations confirming the existence of a planet. The difference? He was right the first time, and wrong the second.
The mathematical explanation is there, but we don't have the experimental evidence, and it's embarrassing. It bugs me when it's presented as "Nothing to see here, it's Dark Matter, we've got the problem solved, 100%", which is how the press presents it. At my school, amongst the physicists, we grin sheepishly over not having a better or more complete answer on this one.
There is still the possibility of an alternate explanation to come out. In fact, what I was saying, since the dark matter correction has been so successful, this points to the fact that all these phenomena will likely be explained using the same mechanism whether it ends up being dark matter or not.
Another case in point, Copernicus' heliocentric system still used epicycles. He believed firmly in perfect circular motion, and put all the planets in perfectly circular orbits around the sun. Then, to correct for the deviations which we now know come from the fact that orbits are elliptical, he put each planet (except earth) on a small epicycle. Both Copernicus' and Ptolemy's systems had a 17 year glitch, where every 17 years, the whole system would significantly mispredict planetary locations. However, had Copernicus been "more Copernican" and put an epicycle on earth's orbit, the glitch would have gone away. The point being, do you believe in elliptical orbits, or perfect circular motion with two cycles, a great circle and an epicycle? In the 17th and 18th centuries, you wouldn't have been able to tell the difference had Copernicus gotten earth right.
To get to my comment about dark matter being the "new æther": at the end of the 19th centur
Or don't.
(And yes, I'm typing on that keyboard right now. I don't even notice the difference.)
Jeff
IAAP (I am a physicist), and the gravitationally stable spheres is a problem. That's what this study was looking at.
The bright, visible, normal matter forms into a disk in every galaxy we see. This cannot be explained with Newtonian gravity (or Einsteinian, for that matter). You see, when you just stick the normal matter in a simulation to check the evolution of a galaxy, it doesn't stay in the disk shape. To get the simulations to work (meaning, predict disk galaxies), you have to put a spherical halo of dark matter around the galaxy. With the dark matter there, it works perfectly.
The other option is that we're completely screwed up in terms of our beliefs about gravity. However, we haven't gotten any clean results using alternate formulations of gravity.
So the answer is, to the best of my knowledge, galaxies are disks because they either have (a) halos of dark matter or (b) our formulation of gravity is extremely flawed. Both of these answers are the current big embarrassments in physics. Both are bad answers. I haven't seen an alternate to gravity that works, and dark matter is a cop out.
From looking at the article, it looks like they've done a huge simulation taking the dark matter halo bit to the extreme and finding that it correctly predicts a flat arrangement for the satelite galaxies.
This is not proof of dark matter, it merely shows that it is the appropriate adjustment to gravity to explain some of the phenomena in the sky. What this actually means is that the correct answer for "why are galaxies disks?" will likely be the same answer for the pancake mystery. So, the authors did not actually solve the problem, they just showed that the current, in vogue kludge works for the pancake mystery too.
I'm, personally, quite sick of hearing it held up as an explanation as if it is the end all be all of fixing cosmology. It is a stupid cop-out with nothing to back it up. There is no experimental evidence for it. We have yet to find anything that could be dark matter (and there are people looking very hard to find it). Other theories don't have anything in them which could be dark matter (no particle theorist has come up with anything, and certainly not the high energy experimentalists). It is a big ol' strap of duct tape to fix a gaping hole in cosmology. There is nothing backing it up. Quite frankly, it's starting to sound like the Æther to me. And we all remember where that got us.
Jeff
Doesn't pass the railgun test. I'd never get an accurate shot with that thing. The optical mouse stands.
Besides, stretching your thumbs like that looks terrible.
Jeff
Well, at least he didn't fall.
7th floor though, that takes some balls.
Jeff (soon to be UW alum)
Ever tried measuring the mass of a neutrino? Yeah, it's pretty damn hard. The easiest way to do that is to watch the beta decay of the tritium nucleus and make precision measurements of the products' and reactants' momentum. The only one you can't really measure is the neutrino itself, but a certain percentage of the decays give you data that you can reduce such that you can plug the results into an equation for the neutrino mass.
There's a problem with this equation. The mass of the tritium nucleus and the helium-3 nucleus need to be subtracted off to get the neutrino mass. The upper bound on the neutrino mass is 2eV. Both those nuclei are approximately 3GeV. To get the error to work out, you need the nuclear masses of the tritium and helium accurate to the parts per trillion range.
In other words, before you can measure the neutrino mass, someone needs to push as hard as possible for perfect measurements of the tritium and helium-3 nuclear masses.
Jeff
I like it when you define hbar = c = 1 (Plank's Constant over 2*pi, instead of being in energy/Hz, it's in energy*sec/radian). Once you've done this, because of E = mc^2, mass has units of energy, and then time and distance both have units of 1/energy. Neat, eh? It's the natural unit system of the universe. It's pretty inconvenient unless you do particle physics, however.
Basically, the idea is once you've got c, hbar, and the three basic unit types (mass, length, time), you've got a complete system of units. c and hbar then turn into conversion factors which can be set arbitrarily to get a system of units--this is what actually makes the Plank's constant elegant, by nailing it down as an exact number, we've completely defined our system of units (c is already a defined quantity). Then, more accurate measurements of hbar are not measurements, so much as refinements of the definitions of kilograms, seconds and meters.
Jeff
There is. I'm running Slackware 10.0 (can't wait to get my 10.1 disks! w00t!). I use vim+LaTeX for all my document prep needs on the command line (use an xterm if you so choose). As others have mentioned, LyX isn't bad.
I'm a physics major at UW, so I do a decent bit of scientific work on my computer. I use GNUPlot, XFig and the Gimp to generate drawings for lab reports and whatnot.
I'm typing this from Firefox running on the Xorg6.7.0 server+WindowMaker 0.91. The key here is to use a lightweight window manager. Blackbox and fluxbox are other good choices (light, usable, not fugly (cough, fvwm, cough)). If you have to have that desktop environment, go with Xfce.
The only gap that I occasionally feel in my user experience is a good spreadsheet. I haven't found one. KSpread, OpenOffice Calc and Gnumeric either are or require the use of heavy GUI software which we are trying to avoid (KDE and Gnome are not as big as XP, but far too big to run comfortably on my system). I've glanced at Siag, but haven't really tried it out (I don't know scheme and don't have the time to figure it out right now--see physics undergraduate work).
I use mutt or pine, depending on which email address I'm checking. Thunderbird looks promising for being light and good, if you want a GUI based email client.
Recompile your kernel to match your hardware (trim the fat and optimize for your processors), and turn off any extra servers that you don't need (don't need telnetd, ftpd, &c. running? Turn off inetd--it's also more secure). Customize your boot sequence to only start and load that which your system needs and those things which you use.
I also boot to the command line and don't run xdm or the like. I do a lot of work from the command line, and X+light WM doesn't take long to start. It is, again, one less thing wasting clock cycles on my machine.
For reference, I'm running my Slack 10 system on an Abit BP6 with two PIII 866MHz processors underclocked to 650MHz (long story... Has to do with the fact that the BP6 doesn't technically support the PIII). I've got 384MB of RAM and a GF4 video card. It is lightning fast. The only exception to this is when I'm running X with the closed nVidia drivers (damn thing has a 3MB kernel module... grrr...), but that only adds a hang of a couple seconds when switching between X and the consoles, and that's it. If I'm not playing Quake or dealing with 3D visualization stuff, I can use the OSS driver (2D accel only), and get rid of even that performance problem.
So, yes, the middle ground is there, and it rocks. My computing experience is awesome, my slightly dated hardware is rock solid and perfectly responsive. Take a good, customizable Linux distribution, run light weight software, turn off stuff in the background and run a lean, mean, customized kernel, and you'll reclaim those lost cycles as interface responciveness. I suggest Slackware for this. FreeBSD, Debian, and any other Linux distro which is aimed at power-users will be good for setting up a configuration like this.
Mandrake, RHAT (RHEL & Fedora), SuSE and any other user-friendly type distro is ill-suited to this, IMO. Not that you can't, but my experience with these distros and their high-level admin tools is that if you try to do something too different from the default, it gets extra hard. So, Slackware and the like just end up being simpler, and now you know what Slack users mean when they say "it's simple." So stop giving us funny looks when we say it.
Jeff
The ATI drivers for linux suck. ATI doesn't release the spec for the FOSS community to write our own drivers, and ATI's binary drivers are terrible.
I've been looking at upgrading my video card (I want to run tenebrae on my box, but don't have the horsepower in my video card. I've decided against going with ATI because fo this article and this comparison of the available ATI drivers (the FOSS dri drivers, the ATI binary drivers and the XiG Xserver). All three fare pretty poorly in different ways, and it's just a mess. My impression is that things have not gotten any better.
Basically, the ATI drivers don't present a problem for testing under Windows, but under Linux it is a mess best left alone, simply because all three implementations are incomplete or underperform in their own special ways. It's an intractable problem for the game vendors, and I think the best route for them is to make everything work in standard OpenGL, and check it with nVidia hardware. If it works there, they've got it. Gamers using ATI cards with linux are out of luck because of ATI's stance on specs and linux support.
So, your problem is that you assume ATI has written complete drivers for linux.
Jeff
Nope, this doesn't defeat the argument, because when the charge drops below a certain threshhold, the gas engine kicks in to charge them no matter what---independent of whether or not the car is stopped. So, the fact that the car is at a stoplight doesn't matter. (The engine only runs at the optimum power output to charge the batteries, so if does it at the stoplight or while the car is moving, it doesn't change the total amount of gas converted to electricity converted to kinetic energy efficiency for the trip.)
Jeff
Anything that can be done to a normal car to make it more efficient can be done to a hybrid to make it more efficient. Toyota make the some of the most efficient gasoline cars in the world---their Prius is right in line with their high standard of engineering (IIRC, Toyota has the highest number of cars/trucks with greater than 30mpg).
Comparing a Diesel car to a hybrid is slightly apples to oranges. (Diesel engies are more efficient that spark engines for various reasons, so yes, you're buddy's Jetta does fare reasonably well against a Prius, but that's only because the Prius is still running a spark engine). If VW went and designed a Diesel hybrid, it would smoke your buddy's Jetta for mileage. Here's why: a normal car with a reciprocating engine can only get at most 20% efficiency. A hybrid scheme allows you to design the engine to a lower power output and only run it at its optimum power output (to get more you run engine+motor, to get less you just run motor, you design the charger such that the engine runs at optimum power output). So, you can get 30% efficiency out of the reciprocating engine or a little more. So, a Diesel hybrid against a normal Diesel would get the better mileage.
Hybrid technology is not stupid. QED.
Jeff
The parent is right that the cars only get their energy from gasoline. However, there is a deeper efficiency story here that just isn't as quick and easy a sound bite as "getting your car's energy from two sources---gas and electricity"---which is an awful marketroid half-truth.
;). But anyway, the point is, that because of the requirements put on the spark engine in a car, it has to be designed for maximum power output, and this means for most driving the highest efficiency a car engine can attain is 20% (this would be for a tiny Japanese car which is engineered for fuel efficiency, like a Toyota Tercel or Honda Civic, other cars, like most SUVs, perform worse).
;).
Anyway, to clarify on hybrids:
The efficiency story goes like this: your normally car engine sucks as far as efficiency is concerned. This is because they have to operate over a wide range of speed and power requirements. Eg. from just taking off from a dead stop to running up a hill at 70mph or more. A spark engine can get to be about 30% efficient (this is from memory, it might even be up to 40%, I'd have to go look it up, and I'm feeling lazy
The trick with the hybrids like the Prius is that they have the battery+electric motor to supplement the gas engine. So, the designers can do something important: they can pick a median power output (much below maximum required power output), and design the spark engine to be maximally efficient for that power output. This allows them to get the 30-40% efficiency out of the gas engine mentioned above. The hybrid only ever runs the gas engine at this power output. If this is too much, the electric motor run the wheels. If this is too little, the motor and the engine drive the wheels. If the batteries are getting low, the gas engine drives the electric motor to charge the batteries. When braking, part of the axel motion is used to drive the electric motor and charge the batteries (reclaiming some of the energy already expended to be reused---this is the regenerative braking that others have mentioned). Note: the designers at Toyota and Honda have taken advantage of the fact that an electric motor and generator are merely the same device, which one it functions as depends on which end the energy comes in, so there is no separate generator. (And if it occurs to you that the clutching system would be complicated because of this, you're right.)
As far as being able to charge up your hybrid, there are some experimental models with that feature. You might eventually be able to do that; so if you just drive around town, you'd only rarely have to fill your tank (however, this feature requires that the bank of batteries is bigger, and 50% of the electricity in the US comes from coal, so the pollution/energy expenditure could end up being worse off the wall charge, depending on where your power comes from
Jeff
As far as your runaway process is concerned, the argument may well be meaningless. Economics is a chaotic system where the particles (the people) have very strong interactions and, in today's world, long distance communication. We have only a very crude understanding of this system, and their are a lot of complex feedback loops in the system which we don't know about, and you certainly aren't taking even potential known ones into account. The system will come to a ballance somewhere, with people figuring out how to take advantage of the most efficient production systems. In other words, I'm using the hated argument of "market forces will make everything work out ok" to shoot you down, or rather, I'm saying that over a long time scale, permanent runaway scenarios are silly doom and gloom arguments, and that we'll end up producing software in the most efficient manner possible---some of us happen to think that is open source, and for it to maintain itself as a force, we need licenses like the GPL. We don't need commercial software to get by, because if we're right, our own software will eventually outdo any commercial offerings.
This is, admittedly, an ugly borderline case, but the way I see it, is by linking to the spellchecker code---even dynamicall---you are incorporating that functionality into your program, and from the user's perspective, once your wordprocessor and the spellchecker have been downloaded and installed (possibly compiled in the case of the checker), the experience of the end user is that the spellchecker is a part of your program. The underlying technical bits aside (which I know quite well, thankyou, my three most used programs are vim, pdfLaTeX and gcc), on the outside, you've incorporated the functionality of their program into yours---you may not have touched any of their code, but from the users perspective, you've incorporated it. The spellchecker authors are licensing (again, copyright is still reserved, they have not given anything up to the public domain) their code for use with other open source programs only. They are not licensing it out so that you can use it with your closed source wordprocessor. They are, probably very willfully, pushing an agenda.
By using the GPL, I force nothing on you. You've already admitted to that. However, I have not declared that you can use my code. Instead, I've declared the terms on which you may use my code---much like you would for your wordprocessor. But even beyond that, I've defined my terms in a clear, standard manner, and you know two things. One, what exactly the conditions for using my code are, and two, that I'm not likely to budge. Compared to the conditions set out in most closed source EULA's, I think by using the GPL I'm being much more fair. If you're a programmer, you've gotten a much larger advantage---the code itself. It is just defining terms. It's legal and just as moral as closed source---if not more so. Because remember, I'm not saying everyone has the rights to my code (that would be public domain or a BSD like license), I'm saying on what terms you can use my code. The price may not be money, but it's still there---much like in closed source. Quit whining that you can't play the game and make up your own rules.
RedHat just announced a profit for third quarter 2004. If you want to play the game, you play by all the rules, or none at all. There are some projects---like PNG---who picked their license such that it could be used in closed source software. In the case of PNG, I would say this was a good idea, because that allowed it to become standard across all platforms, and thus usable on the web as a replacement for the GIF format. The types of licenses you say are better are good for projects which interface between the open source world and the closed source world. However, there aren't 24 variants of the libraries I use. This isn't a problem, and that is why your argument is silly. GPL components belong in open source, and they form the core of the Free Sofware/Open Source ecosystem. The LGLP type code only belongs at the interface between this ecosystem and the closed source ecosystem (while it lasts). Again, you want to use open source? You should go all the way. It works better. All the components of my system are Free Software, and it works great.Play the game, or don't. There isn't a middle ground, and I see no reason for one.
Instead of squawking away out on the internet and probably getting ignored, the FSFE gets to squawk directly at them. I see it as a question of how ignorable we are.
Jeff
On those grades... I'm a University of Washington student (physics major). UW admissions policy will deduct or add to your HS GPA based on the quality of your HS. I went to Olympia HS (in Olympia, WA), because of this, I got +.2 added to my HS GPA for admissions purposes. The cross town rival, Capitol HS, has a similar problem to your HS -- they have an extremely high number of students who graduate with a 4.0. While their academics are fine, their easy grading practices cause UW to deduct points from the GPA's of Capitol HS's students for admission purposes. The high schools are really doing a disservice to students when they grade like that. Colleges often take it into account, and all the students from those high schools get hurt. I would even say that my high school was too easy with its grading policies, but then again, I fully agree with putting a 1 hour time limit on a 2 hour test, and putting the scores on a bell curve.
Jeff
A quicker effect is to take a pair of tinted ski goggles and put them on. After a while your eyes will adjust, and everything will look normal instead of tinted. It's really dramatic with orange tinted goggles (those are what I wear skiing). Everything looks blue afterwards. I would think that this is a similar effect. It seems our brains are very good at making the world around us look the way it is ``supposed'' to.
Jeff
Bought a harddrive lately? This is one of the bigger reasons. Mibibytes are bigger than Megabytes, so if a Harddrive manufacturer is advertising, they advertise in MB = 10^6 bytes, as opposed to the 2^10 bytes, which is the common accepted definition of MB, and since they are technically correct, you get a smaller drive than you think you are buying. *shrug* Not a major sin, but it seems cheap and sleezy to me. I'd like a clearer definition, personally. Seems like an ok fix for pure, proper technical applications. It's a pretty safe bet MiB won't make it into mainstream use, except as a reference to a movie.
Jeff
That would be a ton of computing power... sorry... he started it...
Jeff
the first line is supposed to be:
rot13 str = [rot13Char c | c <- str]
sorry about that.
Jeff
It's in Haskell, and it successfully decodes the above post. Along with anything else using any form of rot13 encryption (eg, double rot13 and quadruple rot13 are common. All that is required is to run the cyphertext through the above program 2 or 4 times).
Jeff
In this case, he was in the country when Adobe decided to press charges. That actually has a long history in this country. There is a law (I wish I could remember the name of it) which states that if a crime is committed, and the criminal and the victim step into an area under US jurisdiction, we can arrest the criminal and run them through our court system. This has been on the books for just about the entire time the US has been a country. It fell into dissuse about 150 years ago, and only got dredged up in the 80's when a man who was tortured in a foriegn country immigrated to the US. His torturer visited the US, was arrested and tried. I don't remember the outcome of this trial.
I'm really kicking myself for not having my sources (nor the time to go look it up), but it's there. I actually found out about this while helping Amnesty International gather signatures for an Internation Right to Know act (which would require companies to report about their labor and environmental practices abroad, like they have to do in the US). Basically, with the reporting could potentially come prosecution of companies breaking US law abroad. So, this law can be a Good Thing.
What to do about the DMCA is another thing entirely. Personally, I'm thinking it might be high time to abuse the law in the opposite direction. Say create a PDF, post it with a licence and sue Adobe for creating a program which breaks your copy protection (PDF being the copy protection). Now, I'd want to go in on this with a better understanding of the DMCA (to properly abuse it) and a couple friendly lawyers (there are cool lawyers out there, I've met a few). Just something inspired by another thread under this article.
Jeff
*pulls out bottle of Dave's Insanity Sauce*
To quote the label:
``Great Cooking ingredient for sauces, soups and stews. Also, strips waxed floors and removes driveway grease stains.''
There you have it.
Jeff