"Rheostat" would mean that you have a variable resistor in front of the load. Dimmer switches use a potentiometer connected to a triac, which controls the duty cycle of the voltage supplied to the load. Inane nitpicking I know, but it's slashdot... what do you expect? LEDs are a little more complicated because they don't run directly off line AC power (or, not usually). Neither do CFLs, by the way... most new ones have an electronic ballast, and the old ones have a magnetic ballast.
However, LED dimming is really easy. LED power supplies rectify the AC, then convert the resulting DC voltage into a controlled DC current output. (If your LED power supply has a voltage output, you're either inefficent or you're going to run into lots of problems). LED power supply controllers come on really cheap chips that have a connection for a dimmer control. Unfortunately, most of them dim the LEDs by producing PWM current output rather than simply reducing the current output, causing the LEDs to flicker when dimmed. This can be mitigated if the frequency is high enough, as the LEDs have a little bit of capacitance.
A block cipher uses a key to map plaintext blocks to ciphertext blocks. Given 2^N plaintext blocks and 2^N ciphertext blocks, the number of possible encryptions are 2^N!, not 2^N. This is a staggering difference in magnitude.
For example, the maximum keysize of a 128 bit block cipher is about 10^22 bits.
It just makes me cringe. What you really mean is "Produces 500W into an 8 ohm resistive load at 0.1% distortion given a sinusoidal waveform at 1kHz". Or you can just drop the RMS and say 500W because we know what you mean... you really mean average power.
Please leave RMS measurements to the electricians.
1. Yes, you can simply transform DC voltages to different voltages. They're called switching power supplies, and you find them EVERYWHERE. You get them off the shelf or build your own, they're cheap, they're light, and they're efficient (90% is not uncommon). Your computer does NOT step down AC to a low voltage then rectify it... it rectifies it to high voltage DC, then steps it down.
2. Losses have nothing to do with AC or DC, it's just a function of current.
Let's say you've got 12 AWG wire in your house (not uncommon). Resistance is.00187 ohm per foot. Let's further say you're running 5VDC across it, and the wire distance to transformer is 50 feet. A short circuit would suck:
(5 V)**2 / (.00187 ohm/ft * 50 ft) = 267 W
Divide this by two to get the maximum power draw from a device: 133W. Sounds like a lot of headroom, but at that point half your electric bill is going to heating the wires! This is why we have high voltage distribution systems.
On the other hand, I would like to see cool medium-high voltage DC distribution systems in the home. This would reduce the complexity of power supplies in our electronics: instead of having power drop out 60 times a second, they see 200VDC or something.
You ever seen a heat sink on a capacitor? No? That's because they don't draw actual power (not counting ESR, which isn't very much). They do increase the apparent power of the load, but domestic consumers (in the US at least) are NOT charged for this and it will not show up in the spec sheet (except some power supplies advertise high power factor, which is only useful to datacenters and the like).
You sure ain't no majestic mathematician... check those stats again. From Wikipedia: The per capita consumption of power in the US is 11.4 kW. Don't forget we sleep eight hours a day too. In colder climates, such a power supply could supplant what heating there is... I know I can turn the heater off when I'm computing in the winter.
This is what makes slashdot so amusing sometimes: the quantity of armchair philosophers who don't even bother to do thirty seconds of research before posting. (It happens plenty in Real Life, but out there it's no longer amusing.)
Organisms can die from diseases. A virus won't destroy a computer, the worst case scenario is a wipe and fresh install. This means that Microsoft can make their software bug-ridden.
Maybe if viruses were to fry hardware, we could see some improvements.
...a striking similarity to Mac OS X? I felt this after reading about the underlying concepts such as prebinding (used heavily in Mac OS X). Then I saw the screenshots. Can't they get some two bit rookie with the GIMP to make them some original icons, or at least correct the gamma in the icons they ripped off?
And get someone to explain basic linking concepts to the reviewer... programs are linked to libraries, not the other way around. Linking usually happens at compile time. Binding is the process of replacing function names with function addresses. Prebinding puts the libraries in a persistent location in memory so that the bindings can be retained across sessions.
I think the reason they don't seem popular now is just because they were over-popular a while ago, around the era of Myst. Kind of like MMORPGs right now.
Speaking of Cliff Johnson, I figured out how to easily solve the 3 in Three puzzle "Whale of a Time" (when I played it again this year). Man, if you don't figure it out, that puzzle will kick your ass. The ending to that game rocks.
It would need to be a really fucking strong magnet.
The DoD once built a degausser strong enough to wipe a hard drive, but it made the lights dim and bent the disk platters. Try an experiment yourself: take a floppy, stick it to the refrigerator with a magnet, leave it there a few days, then try reading the data. It will likely still work (unless it was broken in the first place).
When I was in 9th grade, I wrote a program which would make the screen go black and "The Matrix Has You..." would appear onscreen, just like the movie. Then I set up applescript on the machines in the lab so I could trigger it remotely.
...is to see this downsampled properly. Put a low-pass filter on each pixel in the time domain before changing the framerate. Hell, I'd write the program. but I'd need some kind of framework to stick the code in.
If I actually come up with anything I'll reply to this. Otherwise, assume I learned something but never felt a need to finish the project.
I'm for anything that discourages the production of movies based on books/short stories.
Movies are audiovisual. The "story" is told by the use of composition, lighting, sound effects, and music. When you bring the baggage of literary devices, such as omniscient narration, metaphor, or wordplay, you're taking power away from the director for no good reason whatsoever.
As games become more mature, game designers will develop more techniques to translate traditional linear stories into a non-linear interactive format. Saying that in a video game the story is defined by the user is dishonest -- at least when comparing games to traditional storytelling.
In a game, there are goals defined by the programmers. These goals are usually quite limited. Most games, even the good ones, don't give the player a chance to express motivation in-game. The player only expresses action. This is not a limitation of the medium as the players motivation is merely expressed outside the game, in the player's mind. The only part of the game's story that changes is the actions taken. Games give a more personal experience than anything else short of reality, in spite of the pre-determined stories.
Some day, games will likely exist where the story changes to fit the user. Today, and for the forseeable future, a game's story is a network of objectives and obstacles. Unlike a traditional story, the objectives and obstacles are limited by the sophistication of computers. Unlike a traditional story, goals are connected by a network rather than in a sequence. These differences don't seem major enough to warrant stories' total exclusivity to one medium.
I personally think that many years from now, games will be seen as the ultimate medium for serious expression and storytelling. However, the failure of movie adaptations to games is more a symptom of economics than problems with the adaptation process. When a game publisher has two titles with problems costing, say, an estimated additional $50k to fix (each), one title being licensed the other original, the publisher will often spend the money on the original but not on the licensed title... even if there is enough money for both. Why? Because licensed games have the marketing power of an entire franchise! Warner is trying to tip this offset by adding a new term to the equation.
This probably came of a little condescending, but that's just because I'm thinking out loud. Yah, it's also poorly written, but I'll save my effort for stuff people might actually read.
A photon isn't a particle, it isn't a wave. It's a photon. So many people don't understand that.
So when a bunch of photons show up as an interference pattern, they think of it as waves and the model produces accurate and useful results. When people knock electrons off atoms with gamma rays, they think of the photons as particles and the model produces accurate and useful results. When the two models come together, people have the hardest time understanding it because they forgot the most elementary rule of models:
ALL MODELS ARE WRONG.
As I understand it, under the standard model, we figure out if a photon interacts with another particle by integrating over the set of all possible paths the photon can take in the meantime, producing a probability. That seems like a pretty sound model to me. Does this model require more than one photon to explain diffraction? Nope. Does it talk about waves and interference? Nope. It doesn't mention parallel universes either. This is the model that scientists started using when they realized that both the particle and the wave models were not only wrong, but they didn't always produce useful results.
The problem I have with the claim that this is proof of parallel universes is that parallel universes doesn't add anything to the existing theory. Now, if the parallel universe theory were to predict something disagreeing with the standard model, anything at all whatsoever, it would be useful. However, as it stands, I see the theory as just a more complicated explanation of the standard model. It may be true, but it doesn't seem useful, and usefulness is the only desirable trait in a model.
No scientist understands the laws of the universe, scientists don't even agree on the laws. They don't agree on string theory, on the existance of black holes, on the fate of the universe, on the presense of dark matter, or interpretation of quantum mechanics. If anyone were to ask me about any of those, I would give a resounding maybe. Heck, there are scientists proposing revisions of Newton's law F = m*a to explain discrepancies in galactic rotation.
I just get sick of scientists peddling useless but imaginative models to the public like this. That's what philosophers / fiction writers are for.
I think 'clean' and 'optimized' are very different goals. I like for my code to work first, be clean and readable second, and be optimized third.
With a small app, I have a very big excuse for not optimizing: in the time it takes me to figure out the bottlenecks and rewrite them, I could have gotten some actual work done on a different project. Check the Portland Pattern Repository's entry on the rules of optimization (http://c2.com):
- First rule of optimization: don't
- Second rule of optimization: don't, yet
- Third rule of optimization: the bottleneck isn't where you think it is
Here's a suggestion for them programming folk: RUN the program first, if the speed is acceptible, then don't optimize. If it isn't acceptable, don't optimize yet, profile your code. If you don't profile your code but you do optimize, I hate to say this, but you're being a moron.
The cool thing about profiling is that it makes optimizing a large application much, much easier. Instead of staring down 100 klocs to optimize, you're looking at the 1 or 2 klocs that use 80% of the time.
Remember Final Fantasy? You may not save except on the world map. You may be 90% through a dungeon, then, randomly, you meet some dragon with the capability to wipe your party out in seconds. Try again, hope for better luck. Maybe next game you'll fight a pair of them?
I only saw previews for The Core, but I gathered that the core of the Earth had stopped spinning, and the good guys had to restart it with nukes.
Recall moment of inertia for a sphere, I = 2/5 mr^2. The mass of the Earth's core is 1.932e+24 kg, the radius, 3.488e+6 m. This gives a moment of 9.402e+36 kg m^2. The period of the core's rotation (one sidereal day) is 8.616e+3 s, giving [E= 1/2 Iw^2] rotational kinetic energy of 2.500e+28 J. Note that SI prefixes only go up to 10^24 (unless I'm mistaken).
Now, how many nukes would have to be used to supply this energy? One kiloton TNT is 4.184e+12 J, giving the Earth's core kinetic energy of 5.975e+15 kilotons TNT. Were we to actually use TNT, the diameter of the dynamite required would be 953 kilometers and surface gravity would be 4.5% that of Earth. But I digress.
So, back to nukes, the highest yield nuclear weapon that the US has ever produced (I think) is the triple-stage Mk-41, with up to 25 megatons TNT of explosive yield. 2.4e+11 of these would be required to provide sufficient energy to start the core's motion. To put this in perspective, each Mk-41 being 3.4 m long, the nuclear bombs required would span the average distance between the Earth and the sun five and a half times. (Hey, a lever! Never mind that the outside edge of this ridiculous construction would be moving at.3c.)
For the Star Trek crowd, the amount of antimatter required is half of [E=mc^2] 2.781e+11 kg. The amount of energy is the same amount that the sun releases [our nice big 4e+26 W bulb] in about an hour. Enough energy to boil all the oceans almost thirty million times over. I knew that the movie premise was absurd, but I had no idea how many orders of magnitude the absurdity was.
I leave it as an exercise for the reader to figure out such trivia as "Where the hell did the law of conservation of angular momentum go?"
Sources: http://www.strategic-air-command.com/weapons/nucle ar_bomb_chart.htm, CRC books, Wikipedia, and sites on the internet I forgot about =).
.DS_Store files have nothing to do with standard metadata or resource forks. I don't know exactly why they appear, but I think it has something to do with extra metadata that the Finder uses, and I think it changed in Panther (they seem to appear less often, but I may be mistaken). The files that appear have obvious names like resource.frk for the resource fork.
I think that in the best of all possible worlds (not counting ones where computers are our omniscient servants) that the metadata would still be just extra tags on files. For example, I could have a file named "Circuit Diagram" and it would be an SVG file, because the OS recorded a MIME type of image/svg+xml.
If you think about it, every networked platform already supports MIME types. On a Mac, you wouldn't have to worry about mapping creator codes to MIME types or vice versa, and on Windows you wouldn't have to worry about nasty things like extension conflicts. The only problem is that there isn't any OS-level support for this (i.e. there are filesystems that can handle this). You could add tags for file previews, source URLs, authors, file position.
Tags could be changed without regard to file write permissions, but these changes would only be visible to the user that made them. For example, I could specify that when I double click on the MP3s in your shared folder, they would open in Quicktime instead of iTunes, without changing with what application you open them.
The filesystem would be versioned, and metadata would persist across versions unless explicitly removed. This would eliminate the old problem of editing files on a Mac and having the comment / preview / resource fork / etc. disappear.
People with operating systems that didn't support this would simply not get the extra data. Too bad for them. Their mail program would probably have to add a filename extension too.
For the command-line users who want to be able to know the file's type, just add that feature to ls. You could add a feature to find as well: imagine 'find . -mime image/jpeg', or 'find ~/Downloads -srcurl \*gnu.org\*'.
Personally, I think bundles get a little overused at times. It's an awesome idea for applications and plugins, but I'm not fond of the idea for files. In Garage Band, the sound data is stored as AIFFs... if I want to edit them in a different program, I have to right click > show package contents.
The best part? Anyone could add this to GNU/Linux if he felt like it (and had enough free time & dedication). You'd take a filesystem that supports it, add some metadata code, and write in application support (like mod_metadata for figuring MIME types).
Having been raised on a Mac, I expect to be able to screw with the name all I want without the file magically changing type. The name of the file and its type are not the same information. Extensions are the lowest denominator solution, same reason people have to tar their sources before uploading -- to preserve the files' executable flags, which get stripped over HTTP.
Ideally, I would like my filesystem to know the MIME type of my files. Remember when you could look at a file you downloaded on a Mac, and you could get the URL it was from? Metadata is insanely useful.
Oh, and if you use a Mac to copy files to foreign file systems, the metadata and resource forks will be preserved (through extra hidden files).
The resource fork is not CFM-specific, and is not where metadata goes. Metadata, like the type and creator, are stored along with info like the filename. A file can have this metadata without having a resource fork.
A resource fork is used for extra data. Pre-OS X applications store dialogs, sounds, pictures, icons, strings, and even program code in the resource fork. All files on Mac OS X are capable of having resource forks, this is used by programs like BBEdit which store cursor & window position in the resource fork of text files you create.
Mac OS X is only capable of running one type of application binary, the Mach-O executable. When you run a CFM (Code Fragment Manager) application, launch services will run the 'LaunchCFMApp' program transparently. Normal CFM programs require a 'cfrg' resource in order to function, as well as a 'carb' resource to launch outside the Classic environment. CFM applications aren't necessarily Carbon, but that's by far the most common case.
The program isn't all that special. It has a custom icon, like every other application, but the icon looks like an MP3. If you transfer it without archiving it with Stuffit or MacBinary, the type & creator get killed (can't launch) and the resource fork goes away (no custom icon, can't launch). Since the data fork is a valid MP3 file, when you launch the stripped version it will open iTunes and play. You can also strip the file by going to the command line, and running 'cp virus.mp3 virus2.mp3'.
The 'cfrg' (Code FRaGment) resource is usually created automatically by development tools. It specifies where in the data fork the application code resides. So it's trivial to create an application that is also valid as a different kind of file.
I suspect it will catch the kind of people who put '.' in their $PATH, browse slashdot as root, and open email attachments in Microsoft Outlook.
Oh, and don't think that Mac users haven't had *problems* with viruses, as any Hypercard programmer will tell you (I hated MerryXmas virus).
All you got to do is run your DRM player under emulation / with a debugger so you can peek at the routines used to decrypt it, and grab the key from there. You see, unlike cracking encryption under normal circumstances where you are, e.g., eavesdropping on a conversation between two other parties, the music is a message to YOU, and the company that makes the DRM gives the key to you to unlock it.
DRM is all about hiding keys, it is NOT CRYPTOGRAPHICALLY SECURE. It's not really encryption, it's more like obfuscation that uses the same algorithm.
And by the way... I'd say AAC is more popular / portable than WMA in the first place, AAC is what format many (if not most) DVDs use for audio.
Applescript is a tool for sending messages to other applications, mostly. Hypercard is a database / presentation system, mostly.
I think it's clear you aren't very familiar with Hypercard. You could put Applescript code in a Hypercard stack. You use C/C++ code through XCMDs. When I was in elementary school I created a sequencer in hypercard... you clicked on a drawing of a piano keyboard and could record / play back songs you wrote. The songs appeared in a field that I could copy and paste into a different stack which was a computer game I was working on.
The only reason I gave it up was because it was so slow, not that it wasn't powerful enough. I had written my own XCMDs to add color pictures, get the state of the keyboard, anything a game normally does with the Mac APIs.
The interfaces that are created in Applescript programs are a pale shadow of Hypercard. Applescript also isn't that great with persistent data. If you really want power, stick to C. Actually, you can't do coroutines in C, so you should stick to assembler. Wait, you really should be programming in FPGAs. Er, just wire the logic gates by hand.
Slashdot Mirror
"Rheostat" would mean that you have a variable resistor in front of the load. Dimmer switches use a potentiometer connected to a triac, which controls the duty cycle of the voltage supplied to the load. Inane nitpicking I know, but it's slashdot... what do you expect? LEDs are a little more complicated because they don't run directly off line AC power (or, not usually). Neither do CFLs, by the way... most new ones have an electronic ballast, and the old ones have a magnetic ballast.
However, LED dimming is really easy. LED power supplies rectify the AC, then convert the resulting DC voltage into a controlled DC current output. (If your LED power supply has a voltage output, you're either inefficent or you're going to run into lots of problems). LED power supply controllers come on really cheap chips that have a connection for a dimmer control. Unfortunately, most of them dim the LEDs by producing PWM current output rather than simply reducing the current output, causing the LEDs to flicker when dimmed. This can be mitigated if the frequency is high enough, as the LEDs have a little bit of capacitance.
mix your metaphors before they hatch.
A block cipher uses a key to map plaintext blocks to ciphertext blocks. Given 2^N plaintext blocks and 2^N ciphertext blocks, the number of possible encryptions are 2^N!, not 2^N. This is a staggering difference in magnitude.
For example, the maximum keysize of a 128 bit block cipher is about 10^22 bits.
It just makes me cringe. What you really mean is "Produces 500W into an 8 ohm resistive load at 0.1% distortion given a sinusoidal waveform at 1kHz". Or you can just drop the RMS and say 500W because we know what you mean... you really mean average power.
Please leave RMS measurements to the electricians.
It's funny to try and serve the disabled with FOSS... the user interface is so bad most of the time that people without disabilities have trouble.
Get a user inteface that doesn't suck. Then we can talk about extending it for disabled users.
Wrong! Google HVDC... there's plenty of DC power distribution (read: across many km).
1. Yes, you can simply transform DC voltages to different voltages. They're called switching power supplies, and you find them EVERYWHERE. You get them off the shelf or build your own, they're cheap, they're light, and they're efficient (90% is not uncommon). Your computer does NOT step down AC to a low voltage then rectify it... it rectifies it to high voltage DC, then steps it down.
.00187 ohm per foot. Let's further say you're running 5VDC across it, and the wire distance to transformer is 50 feet. A short circuit would suck:
2. Losses have nothing to do with AC or DC, it's just a function of current.
Let's say you've got 12 AWG wire in your house (not uncommon). Resistance is
(5 V)**2 / (.00187 ohm/ft * 50 ft) = 267 W
Divide this by two to get the maximum power draw from a device: 133W. Sounds like a lot of headroom, but at that point half your electric bill is going to heating the wires! This is why we have high voltage distribution systems.
On the other hand, I would like to see cool medium-high voltage DC distribution systems in the home. This would reduce the complexity of power supplies in our electronics: instead of having power drop out 60 times a second, they see 200VDC or something.
You ever seen a heat sink on a capacitor? No? That's because they don't draw actual power (not counting ESR, which isn't very much). They do increase the apparent power of the load, but domestic consumers (in the US at least) are NOT charged for this and it will not show up in the spec sheet (except some power supplies advertise high power factor, which is only useful to datacenters and the like).
You sure ain't no majestic mathematician... check those stats again. From Wikipedia: The per capita consumption of power in the US is 11.4 kW. Don't forget we sleep eight hours a day too. In colder climates, such a power supply could supplant what heating there is... I know I can turn the heater off when I'm computing in the winter.
This is what makes slashdot so amusing sometimes: the quantity of armchair philosophers who don't even bother to do thirty seconds of research before posting. (It happens plenty in Real Life, but out there it's no longer amusing.)
Organisms can die from diseases. A virus won't destroy a computer, the worst case scenario is a wipe and fresh install. This means that Microsoft can make their software bug-ridden.
Maybe if viruses were to fry hardware, we could see some improvements.
...a striking similarity to Mac OS X? I felt this after reading about the underlying concepts such as prebinding (used heavily in Mac OS X). Then I saw the screenshots. Can't they get some two bit rookie with the GIMP to make them some original icons, or at least correct the gamma in the icons they ripped off?
And get someone to explain basic linking concepts to the reviewer... programs are linked to libraries, not the other way around. Linking usually happens at compile time. Binding is the process of replacing function names with function addresses. Prebinding puts the libraries in a persistent location in memory so that the bindings can be retained across sessions.
It's a step in the right direction, though.
I think the reason they don't seem popular now is just because they were over-popular a while ago, around the era of Myst. Kind of like MMORPGs right now.
Speaking of Cliff Johnson, I figured out how to easily solve the 3 in Three puzzle "Whale of a Time" (when I played it again this year). Man, if you don't figure it out, that puzzle will kick your ass. The ending to that game rocks.
It would need to be a really fucking strong magnet.
The DoD once built a degausser strong enough to wipe a hard drive, but it made the lights dim and bent the disk platters. Try an experiment yourself: take a floppy, stick it to the refrigerator with a magnet, leave it there a few days, then try reading the data. It will likely still work (unless it was broken in the first place).
When I was in 9th grade, I wrote a program which would make the screen go black and "The Matrix Has You..." would appear onscreen, just like the movie. Then I set up applescript on the machines in the lab so I could trigger it remotely.
...is to see this downsampled properly. Put a low-pass filter on each pixel in the time domain before changing the framerate. Hell, I'd write the program. but I'd need some kind of framework to stick the code in.
If I actually come up with anything I'll reply to this. Otherwise, assume I learned something but never felt a need to finish the project.
I'm for anything that discourages the production of movies based on books/short stories.
Movies are audiovisual. The "story" is told by the use of composition, lighting, sound effects, and music. When you bring the baggage of literary devices, such as omniscient narration, metaphor, or wordplay, you're taking power away from the director for no good reason whatsoever.
As games become more mature, game designers will develop more techniques to translate traditional linear stories into a non-linear interactive format. Saying that in a video game the story is defined by the user is dishonest -- at least when comparing games to traditional storytelling.
In a game, there are goals defined by the programmers. These goals are usually quite limited. Most games, even the good ones, don't give the player a chance to express motivation in-game. The player only expresses action. This is not a limitation of the medium as the players motivation is merely expressed outside the game, in the player's mind. The only part of the game's story that changes is the actions taken. Games give a more personal experience than anything else short of reality, in spite of the pre-determined stories.
Some day, games will likely exist where the story changes to fit the user. Today, and for the forseeable future, a game's story is a network of objectives and obstacles. Unlike a traditional story, the objectives and obstacles are limited by the sophistication of computers. Unlike a traditional story, goals are connected by a network rather than in a sequence. These differences don't seem major enough to warrant stories' total exclusivity to one medium.
I personally think that many years from now, games will be seen as the ultimate medium for serious expression and storytelling. However, the failure of movie adaptations to games is more a symptom of economics than problems with the adaptation process. When a game publisher has two titles with problems costing, say, an estimated additional $50k to fix (each), one title being licensed the other original, the publisher will often spend the money on the original but not on the licensed title... even if there is enough money for both. Why? Because licensed games have the marketing power of an entire franchise! Warner is trying to tip this offset by adding a new term to the equation.
This probably came of a little condescending, but that's just because I'm thinking out loud. Yah, it's also poorly written, but I'll save my effort for stuff people might actually read.
A photon isn't a particle, it isn't a wave. It's a photon. So many people don't understand that.
So when a bunch of photons show up as an interference pattern, they think of it as waves and the model produces accurate and useful results. When people knock electrons off atoms with gamma rays, they think of the photons as particles and the model produces accurate and useful results. When the two models come together, people have the hardest time understanding it because they forgot the most elementary rule of models:
ALL MODELS ARE WRONG.
As I understand it, under the standard model, we figure out if a photon interacts with another particle by integrating over the set of all possible paths the photon can take in the meantime, producing a probability. That seems like a pretty sound model to me. Does this model require more than one photon to explain diffraction? Nope. Does it talk about waves and interference? Nope. It doesn't mention parallel universes either. This is the model that scientists started using when they realized that both the particle and the wave models were not only wrong, but they didn't always produce useful results.
The problem I have with the claim that this is proof of parallel universes is that parallel universes doesn't add anything to the existing theory. Now, if the parallel universe theory were to predict something disagreeing with the standard model, anything at all whatsoever, it would be useful. However, as it stands, I see the theory as just a more complicated explanation of the standard model. It may be true, but it doesn't seem useful, and usefulness is the only desirable trait in a model.
No scientist understands the laws of the universe, scientists don't even agree on the laws. They don't agree on string theory, on the existance of black holes, on the fate of the universe, on the presense of dark matter, or interpretation of quantum mechanics. If anyone were to ask me about any of those, I would give a resounding maybe. Heck, there are scientists proposing revisions of Newton's law F = m*a to explain discrepancies in galactic rotation.
I just get sick of scientists peddling useless but imaginative models to the public like this. That's what philosophers / fiction writers are for.
I think 'clean' and 'optimized' are very different goals. I like for my code to work first, be clean and readable second, and be optimized third.
With a small app, I have a very big excuse for not optimizing: in the time it takes me to figure out the bottlenecks and rewrite them, I could have gotten some actual work done on a different project. Check the Portland Pattern Repository's entry on the rules of optimization (http://c2.com):
- First rule of optimization: don't
- Second rule of optimization: don't, yet
- Third rule of optimization: the bottleneck isn't where you think it is
Here's a suggestion for them programming folk: RUN the program first, if the speed is acceptible, then don't optimize. If it isn't acceptable, don't optimize yet, profile your code. If you don't profile your code but you do optimize, I hate to say this, but you're being a moron.
The cool thing about profiling is that it makes optimizing a large application much, much easier. Instead of staring down 100 klocs to optimize, you're looking at the 1 or 2 klocs that use 80% of the time.
Remember Final Fantasy? You may not save except on the world map. You may be 90% through a dungeon, then, randomly, you meet some dragon with the capability to wipe your party out in seconds. Try again, hope for better luck. Maybe next game you'll fight a pair of them?
I only saw previews for The Core, but I gathered that the core of the Earth had stopped spinning, and the good guys had to restart it with nukes.
.3c.)
e ar_bomb_chart.htm, CRC books, Wikipedia, and sites on the internet I forgot about =).
Recall moment of inertia for a sphere, I = 2/5 mr^2. The mass of the Earth's core is 1.932e+24 kg, the radius, 3.488e+6 m. This gives a moment of 9.402e+36 kg m^2. The period of the core's rotation (one sidereal day) is 8.616e+3 s, giving [E= 1/2 Iw^2] rotational kinetic energy of 2.500e+28 J. Note that SI prefixes only go up to 10^24 (unless I'm mistaken).
Now, how many nukes would have to be used to supply this energy? One kiloton TNT is 4.184e+12 J, giving the Earth's core kinetic energy of 5.975e+15 kilotons TNT. Were we to actually use TNT, the diameter of the dynamite required would be 953 kilometers and surface gravity would be 4.5% that of Earth. But I digress.
So, back to nukes, the highest yield nuclear weapon that the US has ever produced (I think) is the triple-stage Mk-41, with up to 25 megatons TNT of explosive yield. 2.4e+11 of these would be required to provide sufficient energy to start the core's motion. To put this in perspective, each Mk-41 being 3.4 m long, the nuclear bombs required would span the average distance between the Earth and the sun five and a half times. (Hey, a lever! Never mind that the outside edge of this ridiculous construction would be moving at
For the Star Trek crowd, the amount of antimatter required is half of [E=mc^2] 2.781e+11 kg. The amount of energy is the same amount that the sun releases [our nice big 4e+26 W bulb] in about an hour. Enough energy to boil all the oceans almost thirty million times over. I knew that the movie premise was absurd, but I had no idea how many orders of magnitude the absurdity was.
I leave it as an exercise for the reader to figure out such trivia as "Where the hell did the law of conservation of angular momentum go?"
Sources: http://www.strategic-air-command.com/weapons/nucl
.DS_Store files have nothing to do with standard metadata or resource forks. I don't know exactly why they appear, but I think it has something to do with extra metadata that the Finder uses, and I think it changed in Panther (they seem to appear less often, but I may be mistaken). The files that appear have obvious names like resource.frk for the resource fork.
I think that in the best of all possible worlds (not counting ones where computers are our omniscient servants) that the metadata would still be just extra tags on files. For example, I could have a file named "Circuit Diagram" and it would be an SVG file, because the OS recorded a MIME type of image/svg+xml.
If you think about it, every networked platform already supports MIME types. On a Mac, you wouldn't have to worry about mapping creator codes to MIME types or vice versa, and on Windows you wouldn't have to worry about nasty things like extension conflicts. The only problem is that there isn't any OS-level support for this (i.e. there are filesystems that can handle this). You could add tags for file previews, source URLs, authors, file position.
Tags could be changed without regard to file write permissions, but these changes would only be visible to the user that made them. For example, I could specify that when I double click on the MP3s in your shared folder, they would open in Quicktime instead of iTunes, without changing with what application you open them.
The filesystem would be versioned, and metadata would persist across versions unless explicitly removed. This would eliminate the old problem of editing files on a Mac and having the comment / preview / resource fork / etc. disappear.
People with operating systems that didn't support this would simply not get the extra data. Too bad for them. Their mail program would probably have to add a filename extension too.
For the command-line users who want to be able to know the file's type, just add that feature to ls. You could add a feature to find as well: imagine 'find . -mime image/jpeg', or 'find ~/Downloads -srcurl \*gnu.org\*'.
Personally, I think bundles get a little overused at times. It's an awesome idea for applications and plugins, but I'm not fond of the idea for files. In Garage Band, the sound data is stored as AIFFs... if I want to edit them in a different program, I have to right click > show package contents.
The best part? Anyone could add this to GNU/Linux if he felt like it (and had enough free time & dedication). You'd take a filesystem that supports it, add some metadata code, and write in application support (like mod_metadata for figuring MIME types).
Having been raised on a Mac, I expect to be able to screw with the name all I want without the file magically changing type. The name of the file and its type are not the same information. Extensions are the lowest denominator solution, same reason people have to tar their sources before uploading -- to preserve the files' executable flags, which get stripped over HTTP.
Ideally, I would like my filesystem to know the MIME type of my files. Remember when you could look at a file you downloaded on a Mac, and you could get the URL it was from? Metadata is insanely useful.
Oh, and if you use a Mac to copy files to foreign file systems, the metadata and resource forks will be preserved (through extra hidden files).
The resource fork is not CFM-specific, and is not where metadata goes. Metadata, like the type and creator, are stored along with info like the filename. A file can have this metadata without having a resource fork.
A resource fork is used for extra data. Pre-OS X applications store dialogs, sounds, pictures, icons, strings, and even program code in the resource fork. All files on Mac OS X are capable of having resource forks, this is used by programs like BBEdit which store cursor & window position in the resource fork of text files you create.
Mac OS X is only capable of running one type of application binary, the Mach-O executable. When you run a CFM (Code Fragment Manager) application, launch services will run the 'LaunchCFMApp' program transparently. Normal CFM programs require a 'cfrg' resource in order to function, as well as a 'carb' resource to launch outside the Classic environment. CFM applications aren't necessarily Carbon, but that's by far the most common case.
The program isn't all that special. It has a custom icon, like every other application, but the icon looks like an MP3. If you transfer it without archiving it with Stuffit or MacBinary, the type & creator get killed (can't launch) and the resource fork goes away (no custom icon, can't launch). Since the data fork is a valid MP3 file, when you launch the stripped version it will open iTunes and play. You can also strip the file by going to the command line, and running 'cp virus.mp3 virus2.mp3'.
The 'cfrg' (Code FRaGment) resource is usually created automatically by development tools. It specifies where in the data fork the application code resides. So it's trivial to create an application that is also valid as a different kind of file.
I suspect it will catch the kind of people who put '.' in their $PATH, browse slashdot as root, and open email attachments in Microsoft Outlook.
Oh, and don't think that Mac users haven't had *problems* with viruses, as any Hypercard programmer will tell you (I hated MerryXmas virus).
All you got to do is run your DRM player under emulation / with a debugger so you can peek at the routines used to decrypt it, and grab the key from there. You see, unlike cracking encryption under normal circumstances where you are, e.g., eavesdropping on a conversation between two other parties, the music is a message to YOU, and the company that makes the DRM gives the key to you to unlock it.
DRM is all about hiding keys, it is NOT CRYPTOGRAPHICALLY SECURE. It's not really encryption, it's more like obfuscation that uses the same algorithm.
And by the way... I'd say AAC is more popular / portable than WMA in the first place, AAC is what format many (if not most) DVDs use for audio.
Applescript is a tool for sending messages to other applications, mostly. Hypercard is a database / presentation system, mostly.
I think it's clear you aren't very familiar with Hypercard. You could put Applescript code in a Hypercard stack. You use C/C++ code through XCMDs. When I was in elementary school I created a sequencer in hypercard... you clicked on a drawing of a piano keyboard and could record / play back songs you wrote. The songs appeared in a field that I could copy and paste into a different stack which was a computer game I was working on.
The only reason I gave it up was because it was so slow, not that it wasn't powerful enough. I had written my own XCMDs to add color pictures, get the state of the keyboard, anything a game normally does with the Mac APIs.
The interfaces that are created in Applescript programs are a pale shadow of Hypercard. Applescript also isn't that great with persistent data. If you really want power, stick to C. Actually, you can't do coroutines in C, so you should stick to assembler. Wait, you really should be programming in FPGAs. Er, just wire the logic gates by hand.