How do we find stars and planets? We make assumptions about stellar phenomena and then predict other phenomena using those assumptions as long as they seem to work.
We do the same thing with everything we use. We've done the same thing with other stuff, but most of the time, we can observe a lot more dimensions of the data than we can with stellar phenomenon to make our predictions.
So I suggest that there are any number of reasons that could indicate why this answer makes sense: the model for detecting mass may be wrong, or the model of the formation of black holse, or somethinig else that I haven't considered.
At any rate, we have a long way to go to learn to understand stellar phenomena.
All problems as far as we know can be placed into a few general categories based upon their complexity. It can be shown (though I won't be doing it) that both making a robot do what you want and making a human do whatever are decidability problems: both have the same degree of complexity in the general sense - one task may be harder, but only because its "bigger" - containing more dimensions or variables that must be tested. HOWEVER, if we can find a technique that will describe/model the robots, then it is extremely likely that we can use a morphologically equivalent technique to describe/model human behaviour (based upon the fact that both problems are decidability problems).
This is one of the basic ideas of modern artificial intelligence.
The general technique is to use an approximation of search (finding based upon luck) which tends to find good solution. Generally, it is based upon luck, and the only way to guarantee that its human-like is to restrict the search space to human-like possibilities. Whatever human-like means.
This is NOT the simulation of algorithms already created always, its often the creation of totally new algorithms, which can be extracted after development - a learning approach.
How? Straight statistics, support vector machines, decision trees, neural networks, fuzzy logic, and simulated annhealing are all common techniques to lead towards the goal. Who knows what they actually use.
No more stack overflows, now there are are engine overflows (flooded engines).
Boot up can be stopped by engine not starting.
And a crash could be a little worse than they are now...
I doubt this technology will actually be useful for small electronics. The power output is too unreliable, as are engines in general. Not to mention the volitility of hydrogen. Fuel cells have similar problems.
They'll really be good for portable applications that require massive amounts of power more than anything else.
The holodeck color scheme used on the "Enterprise" (black with yellow lines) is used by Nasa as the background for vector mapping when information is not known or to convey the axes. The engineers specifically requested it.
While its true that QT library can't be compiled,
there are some ways to get around that. And while I like the way the Windows shell works, occasionally I find advantage in using other visual shells. I'd use KDE.
HOWEVER, the big problems lie in the conversion.
I love cygwin and use it as my main POSIX environment. I use it much more than Linux, and have added most of the best tools to my version of it.
Icewm and twm have already been ported, and despite the simplicity of the two, both have problems in Windows, making a lot of other programs less workable (this is especially true of Icewm). In addition, configuration is based upon a series of workarounds, and is thoroughly unlike the elegant method used within Linux.
KDE already somewhat slow, buggy, and complex in Linux - while Icewm and TWM are quite simple. There's no way KDE will even be usable considering its complexity. The ability to change the configuration will make it far too difficult to change, and it would be too slow to run on all but the fastest machines since cygwin versions of apps are all slower than their linux counterparts.
In addition, having the Window manager is one thing, but having apps for it is something else entirely. That's what it'd really be good for. I'd love to use kdevelop under cygwin, but its not going to be joining Windows with KDE.
Simulations have been a large part of Military development for a LONG time; almost since we had computers. I used to go to a school which was next to a longstanding military research post for virtual reality installations. When I left, they where working on a way to make a compact way to distribute the tank simulation data to multiple systems for networking.
Mostly they focus on the visual data in all of the simulations because it helps the most. This is an interesting new thing, but its really only an incremental improvement, considering how advanced the military simulations already are.
Its worse if that happens? Hardly. Its different. Many sci-fi shows throughout the history of television have been action based - especially ones about the future of (mostly) humanity. And this show does focus upon the future of humanity more than other shows - consider that a few of the different characters are wayists (more like deist), Nietzschian (more like Darwinist), pragmatist, and idealist.
And basically, at the beginning of the show, it was a lot like the original star trek - take human philosophies such as these that clash to the extreme and have them fight or at least struggle in order to accomplish something. Of course, you could always develop them more as they have been after the beginning, but not necessarily.
V, Logan's run, and the original Star Trek all had SOME elements of fighting, but mostly it was just fighting. Not a lot of science, either, just a futuristic look. So what?
I'll enjoy the show the way I enjoy cheasy action sci-fi shows rather than the way I enjoy cheasy plot sci-fi shows.
Political suicide? Riiigghhhtt. The US has done FAR worse things than remove an ISP. We've brought economies to our knees because we suspected them of supporting our enemies. Which is worse, cutting of an ISP of a starving country or causing a country to starve (please note that I'm NOT saying we caused the conditions in Somalia - I'm referring to other countries)?
We've also broken the rules of war (yes, they do exist, albiet informally) by hurting the innocents without cause, and without a good enough pretext.
So...are they going to bring back all of the old video games as movies, then? And using actors who have very little to do with anything in the plot?
I don't think I'm going to go see that. I think I'll wait for the pong movie to come out. I hear they're casting Elvis as the lead opposite Marylin Monroe. Apparently there's gong to be a musical number about hitting the ball.
Methane is also a natural gas. And both methane and propane are odorless. The odor is ADDED so that we can tell when there's a gas leak. Did you forget your high school chemistry? This is just basics.
I can think of a couple of ways this could be useful. First, a DJ should have some idea of what they want to make the crowd feel. If a fuzzy genetic algorithm told him something about the emotions of the crowd based upon heart signals (for instance, everyone is roughly "excited"), he might have an idea of what to play next to make everyones experience more entertaining. As a sound technician (which is in no way a DJ, but still has similar attributes), I have noticed that I can do my job best with as much information as possible. At best, this has includes hearing the mix before effects are added, hearing it after, seeing the (amplitude of sound) levels for each channel being mixed, and seeing frequency responses from a real time analyzer. Walking around the room to listen to the sound in various locations also helps. I imagine if I had biofeedback, I could do an even better job of making everything what the audience wants to hear.
It might also help (though minimally) in the design of certain types of music. If you haven't heard, disco is usually set to a tempo which is roughly the speed of the human heartbeat based upon the theory that it makes it more exciting, and a lot of modern pop has encorporated elements of disco. In addition, consider that the Mozart's music is very similar to a mapping of neuronal firings in the brain, including (remarkably) the same fractal dimension, and it has been proven to increase the spacial and temporal reasoning skills temporarily (the Mozart effect).
This could be wrong, but biological theories are a starting point for many kinds of music. Good stuff could come out of this! The real problem that is not being considered is that music is something of a context sensitive language, with lingual rules. Generating music with just a computer must necessarily be a simple subset(because creation in the context sensitive case is NP-hard). Therefore, it will be a tool for people to use, rather than a solution of itself, for a long time.
Its not dumb or pointless, if you're willing to sacrifice slightly.
First, consider that LCDs are expensive and nonstandard, as already mentioned. TVs on the other hand, are cheap, mass-produced and rugged, even when portable. So...use a TV for the display. Of course, the bigger the TV, the more expensive it'll be...but then you can use it to watch TV too.
Then you want rugged and portable? Get a biscuit form factor PC. They're small, but they're still standard PCs (which can therefore run Linux), including all the usual connections. You can get them with TV outputs (for VGA drivers), and built in ethernet and serial connections. Check out some of them at this site.
Most don't come with a hard drive, and you'll have to "roll your own" there, too. You have a choice: buy a biscuit with an IDE connector, or buy one with another standard connection, such as Compact flash or PCMCIA. If you go with the IDE, a portable 500MB IDE isn't that expensive. I saw one for $40 (US dollars) recently.
Then of course, you have to worry about batteries/power supply. That's not as big of a problem as you might think - there are LOTS of battery manufacturers. You want long lasting? You'll have to pay for longer lasting. On the upside, the compact system shouldn't take as much power as a standard laptop would.
How rugged is the result? Well, you'd be creating the casing for everything, so that's pretty much up to you.
So, I figure these costs:
$400 CPU (with ethernet, etc...)
$400 portable Flat Screen TV (VGA video displayed)
$50 Hard Drive
$800 Batteries/Power supply (enough to last a week - a SERIOUS load, and still probably an underestimate of the price)
$80 Casing
Of course, for an extra $500, you could upgrade to having real SVGA video, but you don't need hardware acceleration of video for most applications with which this technology is normally used, so...you can't get it.
I suppose if $1730 is three times as much as you would have paid, then this is a bad deal...but since the Thinkpad X22 retails for around $3000, I think I'm talking about a better deal than a prebuilt machines. Plus this thing is actually upgradable.
Palm really has it figured out: apps that take very little memory or power, and a good handwriting recognition system.
WinCE and Pocket Windows just try to embed Windows, including the lack of speed and horrible handwriting recognition.
Now we give Linux a go. So far, it hasn't really been much of a PDA, just a port. If you have a full sized keyboard (not a little dinky one that's hard to use like this one), and a way to hook up a mouse, then you're fine. Barring that, the user interface is a pain.
What's my point? Use whatever codebase you want! Just give me an easy way to access and input my data - that's exactly what a personal digital assistant is for - NOT for general purpose computing, like a desktop.
This guy must be crazy. The main reason why companies use Windows is familiarity - people who are familiar with Unix don't switch to Windows. And ESPECIALLY not in order to port their Unix software.
The situation described by the article is one in which any salesman would absolutely have to admit defeat: the competitor's product is better at every feature.
Consider:
Cheaper: free vs not
Easier to use: already know Unix
Easier to port stuff: already have apps in Unix
More stable: Umm...duh?
What else does M$ have going for it? About the only thing I can think of is that IE is the best web browser (defining "best" as "having the most capability to parse webpages"), and I don't think that's a big enough selling point.
Windows is made to be sold to the secretaries who use whiteout instead of backspace, not to huge firms who are migrating to cheaper platforms.
I don't think we should ever talk about interesting things to do with cats without talking about the amazing Bonzai Kitties
This isn't a troll! Its just that the very thought of the sound that a cat would make as you tried to shove it into a blender would be...interesting. And it is therefore worth mentioning.
If you're going to make use of the big, powerful editors, the main reason is to make use of rapid application development - building the guis on the fly, and then adding the code (or visa versa). Its the reason that the big, powerful development tools are big. Otherwise, you might as well use a text editor.
Most of the applications have a huge problem: for the parts of the code that they write, you can't edit it, unless you don't want to use the RAD tool for creating gui's anymore (their parsers often only work with their templates - especially J++). The big exception to this is JBuilder. You can write code, change it with the RAD tool, and then change it more. Not only that, but the code that is produced by JBuilder isn't half bad, unlike most of the other GUI's I've tried, which produce horrible looking code.
Of course, there are a few other features I particularly like - it can keep track of the classes you've created so that you can easily figure out what methods to call, but those are just icing.
As to the speed - most of the GUIs are written in Java, which makes them very portable, but about half the speed they would otherwise be. However, its a constant speed slowdown, not an exponential one. Just use a fast enough computer, and you won't notice the difference.
I have a Compaq power supply, though not one of the ones listed. There's a fundamental design flaw in mine - the connection to the computer is made in such a way that the end that connects to computer can easily short (the process is caused by wire breakage due to inserting and removing the connector) causing the rubber to melt/burn, and then burning anything nearby.
Luckily, when mine shorted, I was there, so I disconnected it and soldered it back together with a MUCH more secure connection, and a good insulating supply of duct tape.
I'm pretty certain that all of the Compaq computers use the same cheap connector, making them all susceptable to this fire hazard - something which could be completely avoided if an extra $.50 had been spent on a higher gauge wire. Cheapskates...
You're all forgetting that Microsoft has realeased a converter that allow you to convert Office2000 documents into Office 97 documents. Tell me again why I should switch?
Features #1 and #3 are built into MSOffice, and with ActiveX and OLE servers you can do the Windows equivalent of piping to affect any of the office programs in a multitude of ways; the reason you've probably never heard of them is because you don't use them much. In fact, almost no one knows about them because they're not often mentioned. Also, you can write function extensions to Excel in any language you like, provided you can interface with MSDev (which most languages can), or OLE or ActiveX (I've used a few; they actually have a special facility for creating these).
Oh, and about the CVS history. Many of the document formats include the history of keystrokes as the storage format, allowing an unlimited number of "undos." While this isn't CVS, it does cause that functionality (to a certain degree), though this could CERTAINLY be improved.
So in summary, all these features you think aren't in MSOffice are in MSOffice.
My theory is that these features aren't as desired as you think, since you, and a lot of other people who have read this thread haven't mentioned them.
The experiments that we have had with chess programs have shown us that intelligence is about testing the most possibilities. In other words, humans, with all of our wisdom, couldn't come up with a better way than to let the machines try as many of the possibilities as it can (i.e. Deep Blue). While Deep Blue doesn't use neurons to solve its problem, it uses randomness more than heuristics (it has an extremely simple search heuristic). Also keep in mind the power constraint: computation takes electricity. Neurons are extremely conservative in the use of power, yet the brain uses up a LOT of electricity. Our bodies may not be as wonderful as they could be, but we have yet to produce anything close to as good.
This conversation as it stands so far is a little bit crazy. Humans ARE the most efficient machines at solving problems, and if computers where out to enslave us, we WOULD be the best for the job, not other machines. Of course, most your understandings of computers is that they can do what you cannot do, which is true, but only by virtue of the fact that they run so quickly, whereas the average biological neuron runs at about 100Hz, allowing them to solve strictly linear problems more quickly. There are a lot of problems that you take for granted that computers can't solve (such as navigational problems).
So lets really look at the issue: the SPEED of computers is doubling every 18 months, but at best estimate, the intelligence we've got is that of an 18 month old child at best, and that's probably hype.
One other thing: since when is ambition the same as intelligence? You assume that computers will want things! Why?
This is all just computerstition, a more commonly occuring phenomenon as time progresses that may eventually turn into a false religion.
A year spent in artificial intelligence is enough to make one believe in God.
Alan J. Perlis - Epigrams of Programming
Slashdot Mirror
...about extrapolated data.
How do we find stars and planets? We make assumptions about stellar phenomena and then predict other phenomena using those assumptions as long as they seem to work.
We do the same thing with everything we use. We've done the same thing with other stuff, but most of the time, we can observe a lot more dimensions of the data than we can with stellar phenomenon to make our predictions.
So I suggest that there are any number of reasons that could indicate why this answer makes sense: the model for detecting mass may be wrong, or the model of the formation of black holse, or somethinig else that I haven't considered.
At any rate, we have a long way to go to learn to understand stellar phenomena.
All problems as far as we know can be placed into a few general categories based upon their complexity. It can be shown (though I won't be doing it) that both making a robot do what you want and making a human do whatever are decidability problems: both have the same degree of complexity in the general sense - one task may be harder, but only because its "bigger" - containing more dimensions or variables that must be tested. HOWEVER, if we can find a technique that will describe/model the robots, then it is extremely likely that we can use a morphologically equivalent technique to describe/model human behaviour (based upon the fact that both problems are decidability problems).
This is one of the basic ideas of modern artificial intelligence.
The general technique is to use an approximation of search (finding based upon luck) which tends to find good solution. Generally, it is based upon luck, and the only way to guarantee that its human-like is to restrict the search space to human-like possibilities. Whatever human-like means.
This is NOT the simulation of algorithms already created always, its often the creation of totally new algorithms, which can be extracted after development - a learning approach.
How? Straight statistics, support vector machines, decision trees, neural networks, fuzzy logic, and simulated annhealing are all common techniques to lead towards the goal. Who knows what they actually use.
New windows errors!
No more stack overflows, now there are are engine overflows (flooded engines).
Boot up can be stopped by engine not starting.
And a crash could be a little worse than they are now...
I doubt this technology will actually be useful for small electronics. The power output is too unreliable, as are engines in general. Not to mention the volitility of hydrogen. Fuel cells have similar problems.
They'll really be good for portable applications that require massive amounts of power more than anything else.
The holodeck color scheme used on the "Enterprise" (black with yellow lines) is used by Nasa as the background for vector mapping when information is not known or to convey the axes. The engineers specifically requested it.
Its a small, but notable influence.
What if we could install operating systems on our brains?
This'll be a big area of expansion for Linux. Everyone will install it rather than Windows so that they don't crash their minds.
While its true that QT library can't be compiled,
there are some ways to get around that. And while I like the way the Windows shell works, occasionally I find advantage in using other visual shells. I'd use KDE.
HOWEVER, the big problems lie in the conversion.
I love cygwin and use it as my main POSIX environment. I use it much more than Linux, and have added most of the best tools to my version of it.
Icewm and twm have already been ported, and despite the simplicity of the two, both have problems in Windows, making a lot of other programs less workable (this is especially true of Icewm). In addition, configuration is based upon a series of workarounds, and is thoroughly unlike the elegant method used within Linux.
KDE already somewhat slow, buggy, and complex in Linux - while Icewm and TWM are quite simple. There's no way KDE will even be usable considering its complexity. The ability to change the configuration will make it far too difficult to change, and it would be too slow to run on all but the fastest machines since cygwin versions of apps are all slower than their linux counterparts.
In addition, having the Window manager is one thing, but having apps for it is something else entirely. That's what it'd really be good for. I'd love to use kdevelop under cygwin, but its not going to be joining Windows with KDE.
Simulations have been a large part of Military development for a LONG time; almost since we had computers. I used to go to a school which was next to a longstanding military research post for virtual reality installations. When I left, they where working on a way to make a compact way to distribute the tank simulation data to multiple systems for networking.
Mostly they focus on the visual data in all of the simulations because it helps the most. This is an interesting new thing, but its really only an incremental improvement, considering how advanced the military simulations already are.
Its worse if that happens? Hardly. Its different. Many sci-fi shows throughout the history of television have been action based - especially ones about the future of (mostly) humanity. And this show does focus upon the future of humanity more than other shows - consider that a few of the different characters are wayists (more like deist), Nietzschian (more like Darwinist), pragmatist, and idealist.
And basically, at the beginning of the show, it was a lot like the original star trek - take human philosophies such as these that clash to the extreme and have them fight or at least struggle in order to accomplish something. Of course, you could always develop them more as they have been after the beginning, but not necessarily.
V, Logan's run, and the original Star Trek all had SOME elements of fighting, but mostly it was just fighting. Not a lot of science, either, just a futuristic look. So what?
I'll enjoy the show the way I enjoy cheasy action sci-fi shows rather than the way I enjoy cheasy plot sci-fi shows.
Political suicide? Riiigghhhtt. The US has done FAR worse things than remove an ISP. We've brought economies to our knees because we suspected them of supporting our enemies. Which is worse, cutting of an ISP of a starving country or causing a country to starve (please note that I'm NOT saying we caused the conditions in Somalia - I'm referring to other countries)?
We've also broken the rules of war (yes, they do exist, albiet informally) by hurting the innocents without cause, and without a good enough pretext.
YEAH!! ANOTHER STANDARD WE CAN'T USE!
I'm going to put my copy with the CSS aural style sheets standard.
How about a standard cost, bandwidth and level of service (i.e. cost lead) for broadband first?
We must learn to walk before we can fly.
So...are they going to bring back all of the old video games as movies, then? And using actors who have very little to do with anything in the plot?
I don't think I'm going to go see that. I think I'll wait for the pong movie to come out. I hear they're casting Elvis as the lead opposite Marylin Monroe. Apparently there's gong to be a musical number about hitting the ball.
Dihydrogen monoxide is the name for water when used in inorganic chemistry (using that naming scheme).
Hydrogen hydroxide is the name for water when used in organic chemistry.
Water is the rare substance that seems to work its way into both very commonly so both names are actually used...
Methane is also a natural gas. And both methane and propane are odorless. The odor is ADDED so that we can tell when there's a gas leak. Did you forget your high school chemistry? This is just basics.
I can think of a couple of ways this could be useful. First, a DJ should have some idea of what they want to make the crowd feel. If a fuzzy genetic algorithm told him something about the emotions of the crowd based upon heart signals (for instance, everyone is roughly "excited"), he might have an idea of what to play next to make everyones experience more entertaining. As a sound technician (which is in no way a DJ, but still has similar attributes), I have noticed that I can do my job best with as much information as possible. At best, this has includes hearing the mix before effects are added, hearing it after, seeing the (amplitude of sound) levels for each channel being mixed, and seeing frequency responses from a real time analyzer. Walking around the room to listen to the sound in various locations also helps. I imagine if I had biofeedback, I could do an even better job of making everything what the audience wants to hear.
It might also help (though minimally) in the design of certain types of music. If you haven't heard, disco is usually set to a tempo which is roughly the speed of the human heartbeat based upon the theory that it makes it more exciting, and a lot of modern pop has encorporated elements of disco. In addition, consider that the Mozart's music is very similar to a mapping of neuronal firings in the brain, including (remarkably) the same fractal dimension, and it has been proven to increase the spacial and temporal reasoning skills temporarily (the Mozart effect).
This could be wrong, but biological theories are a starting point for many kinds of music. Good stuff could come out of this! The real problem that is not being considered is that music is something of a context sensitive language, with lingual rules. Generating music with just a computer must necessarily be a simple subset(because creation in the context sensitive case is NP-hard). Therefore, it will be a tool for people to use, rather than a solution of itself, for a long time.
First, consider that LCDs are expensive and nonstandard, as already mentioned. TVs on the other hand, are cheap, mass-produced and rugged, even when portable. So...use a TV for the display. Of course, the bigger the TV, the more expensive it'll be...but then you can use it to watch TV too.
Then you want rugged and portable? Get a biscuit form factor PC. They're small, but they're still standard PCs (which can therefore run Linux), including all the usual connections. You can get them with TV outputs (for VGA drivers), and built in ethernet and serial connections. Check out some of them at this site.
Most don't come with a hard drive, and you'll have to "roll your own" there, too. You have a choice: buy a biscuit with an IDE connector, or buy one with another standard connection, such as Compact flash or PCMCIA. If you go with the IDE, a portable 500MB IDE isn't that expensive. I saw one for $40 (US dollars) recently.
Then of course, you have to worry about batteries/power supply. That's not as big of a problem as you might think - there are LOTS of battery manufacturers. You want long lasting? You'll have to pay for longer lasting. On the upside, the compact system shouldn't take as much power as a standard laptop would.
How rugged is the result? Well, you'd be creating the casing for everything, so that's pretty much up to you.
So, I figure these costs:
$400 CPU (with ethernet, etc...)
$400 portable Flat Screen TV (VGA video displayed)
$50 Hard Drive
$800 Batteries/Power supply (enough to last a week - a SERIOUS load, and still probably an underestimate of the price)
$80 Casing
Of course, for an extra $500, you could upgrade to having real SVGA video, but you don't need hardware acceleration of video for most applications with which this technology is normally used, so...you can't get it.
I suppose if $1730 is three times as much as you would have paid, then this is a bad deal...but since the Thinkpad X22 retails for around $3000, I think I'm talking about a better deal than a prebuilt machines. Plus this thing is actually upgradable.
Its a PDA! I want it to do that!
Palm really has it figured out: apps that take very little memory or power, and a good handwriting recognition system.
WinCE and Pocket Windows just try to embed Windows, including the lack of speed and horrible handwriting recognition.
Now we give Linux a go. So far, it hasn't really been much of a PDA, just a port. If you have a full sized keyboard (not a little dinky one that's hard to use like this one), and a way to hook up a mouse, then you're fine. Barring that, the user interface is a pain.
What's my point? Use whatever codebase you want! Just give me an easy way to access and input my data - that's exactly what a personal digital assistant is for - NOT for general purpose computing, like a desktop.
This guy must be crazy. The main reason why companies use Windows is familiarity - people who are familiar with Unix don't switch to Windows. And ESPECIALLY not in order to port their Unix software.
The situation described by the article is one in which any salesman would absolutely have to admit defeat: the competitor's product is better at every feature.
Consider:
Cheaper: free vs not
Easier to use: already know Unix
Easier to port stuff: already have apps in Unix
More stable: Umm...duh?
What else does M$ have going for it? About the only thing I can think of is that IE is the best web browser (defining "best" as "having the most capability to parse webpages"), and I don't think that's a big enough selling point.
Windows is made to be sold to the secretaries who use whiteout instead of backspace, not to huge firms who are migrating to cheaper platforms.
I don't think we should ever talk about interesting things to do with cats without talking about the amazing Bonzai Kitties
This isn't a troll! Its just that the very thought of the sound that a cat would make as you tried to shove it into a blender would be...interesting. And it is therefore worth mentioning.
If you're going to make use of the big, powerful editors, the main reason is to make use of rapid application development - building the guis on the fly, and then adding the code (or visa versa). Its the reason that the big, powerful development tools are big. Otherwise, you might as well use a text editor.
Most of the applications have a huge problem: for the parts of the code that they write, you can't edit it, unless you don't want to use the RAD tool for creating gui's anymore (their parsers often only work with their templates - especially J++). The big exception to this is JBuilder. You can write code, change it with the RAD tool, and then change it more. Not only that, but the code that is produced by JBuilder isn't half bad, unlike most of the other GUI's I've tried, which produce horrible looking code.
Of course, there are a few other features I particularly like - it can keep track of the classes you've created so that you can easily figure out what methods to call, but those are just icing.
As to the speed - most of the GUIs are written in Java, which makes them very portable, but about half the speed they would otherwise be. However, its a constant speed slowdown, not an exponential one. Just use a fast enough computer, and you won't notice the difference.
I have a Compaq power supply, though not one of the ones listed. There's a fundamental design flaw in mine - the connection to the computer is made in such a way that the end that connects to computer can easily short (the process is caused by wire breakage due to inserting and removing the connector) causing the rubber to melt/burn, and then burning anything nearby.
Luckily, when mine shorted, I was there, so I disconnected it and soldered it back together with a MUCH more secure connection, and a good insulating supply of duct tape.
I'm pretty certain that all of the Compaq computers use the same cheap connector, making them all susceptable to this fire hazard - something which could be completely avoided if an extra $.50 had been spent on a higher gauge wire. Cheapskates...
You're all forgetting that Microsoft has realeased a converter that allow you to convert Office2000 documents into Office 97 documents. Tell me again why I should switch?
Oh, and about the CVS history. Many of the document formats include the history of keystrokes as the storage format, allowing an unlimited number of "undos." While this isn't CVS, it does cause that functionality (to a certain degree), though this could CERTAINLY be improved. So in summary, all these features you think aren't in MSOffice are in MSOffice. My theory is that these features aren't as desired as you think, since you, and a lot of other people who have read this thread haven't mentioned them.
The experiments that we have had with chess programs have shown us that intelligence is about testing the most possibilities. In other words, humans, with all of our wisdom, couldn't come up with a better way than to let the machines try as many of the possibilities as it can (i.e. Deep Blue). While Deep Blue doesn't use neurons to solve its problem, it uses randomness more than heuristics (it has an extremely simple search heuristic). Also keep in mind the power constraint: computation takes electricity. Neurons are extremely conservative in the use of power, yet the brain uses up a LOT of electricity. Our bodies may not be as wonderful as they could be, but we have yet to produce anything close to as good.
This conversation as it stands so far is a little bit crazy. Humans ARE the most efficient machines at solving problems, and if computers where out to enslave us, we WOULD be the best for the job, not other machines. Of course, most your understandings of computers is that they can do what you cannot do, which is true, but only by virtue of the fact that they run so quickly, whereas the average biological neuron runs at about 100Hz, allowing them to solve strictly linear problems more quickly. There are a lot of problems that you take for granted that computers can't solve (such as navigational problems).
So lets really look at the issue: the SPEED of computers is doubling every 18 months, but at best estimate, the intelligence we've got is that of an 18 month old child at best, and that's probably hype.
One other thing: since when is ambition the same as intelligence? You assume that computers will want things! Why?
This is all just computerstition, a more commonly occuring phenomenon as time progresses that may eventually turn into a false religion.
A year spent in artificial intelligence is enough to make one believe in God.
Alan J. Perlis - Epigrams of Programming