What is required, as you say, is to raise the hydrogen to a suitably high temperature and pressure. The laser is used not only to increase the temperature of your target material, but also the pressure (the sudden increase in temperature causes the surface to ablate, which causes an increase in pressure on the remainder of the target). To achieve the same effect using a fission reaction, you would essentially have to create an explosion, which would be *far* harder to do safely. You're likely to just destory the reactor, as only a small fraction of the total output energy of the explosion could actually be directed at the target.
In an H-bomb, of course, the more destructive the reaction the better.
In a word, magnets. The idea behind fusion, essentially, is that you raise a (hydrogen) plasma to sufficient temperature and pressure, and it will undergo fusion. If you get the conditions just right, it'll then continue to fuse once you've ignited it, thus supplying you with energy.
What you may not appreciate is that a plasma is electrically charged, and can therefore be contained using a suitable magnetic field. Arguably the most promising containment setup at the moment is the tokamak (from the Russian for bottle, iirc), which is a torus-shaped machine. Electromagnets around the torus create a circular magnetic field, which keeps the plasma contained in a ring. (My apologies if my information is out of date, I quit my PhD in plasma physics 4 years ago...)
Despite what the article says, however, fusion is not entirely pollution-free. One of the byproducts is a fairly large supply of neutrons. These neutrons are absorbed by the reactor, which will slowly but surely become radioactive. Therefore, you will eventually be left with radioactive waste to dispose of. You won't get anything like the quantity you get with fission, though, and fusion certainly doesn't produce any "conventional" pollution.
I upgraded to 512mb ram half a year ago, and Linux hasn't done a disk write since
I very much doubt that, unless you're writing absolutely nothing to the disk at all - no data, no log files, nothing.
Otherwise, your machine most certainly is writing to the disk - it's just doing it gradually, as and when there's nothing else for it to do, so you don't notice it.
Other than that, I agree - I have recently upgraded my machine, which included putting 512MB of RAM in it, and neither OS (Linux & XP Pro) noticeably writes to the disk, other than "on demand" (ie when I hit "save").
the eye has so many patterns, its almost impossible to fake
Interesting use of the word "almost" there. So, what happens if/when a method to fake arbitrary retinal patterns is developed?
The great thing about usernames and passwords, PINs, etc, is that if you have reason to think that someone else knows it, you can change it. You can't change your retinal pattern, or any other biometric data.
If someone manages to fake your thumb print or retinal pattern, you're stuffed - you can't change them, and if that's the only official method of identifcation, you can't even revoke them and use a PIN.
Besides, the problem isn't people managing to copy retinal patterns, its people managing to fool detectors into accepting other patterns, or even a photograph of the correct eye...
Not if the libraries in question are licenced under the LGPL. The LGPL specifically allows linking with non-GPLed code; that's the reason it exists distinct from the GPL.
Have you tried using MPlayer? I admit that I've not sat and watched a DVD all the way through (I have a 15" monitor, and a 32" widescreen TV...), but when I bought a DVD drive a few months ago I thought I'd give it a go.
Playing DVDs under Windows proved to be impossible. The OS refused to allow it, as it couldn't disable the TV-out socket on my grpahics card (Creative GeForce3 Ti200). So, I rebooted to Linux and had a go with MPlayer. I only played a few minutes worth, but it worked fine.
I think the original poster's point was that with an open source product, one could simply remove the key checking stage entirely, rather than having to rely on acquiring a key.
Re:Refactoring does not depend on Eclipse: Emacs!
on
Eclipse 2.1 Released
·
· Score: 1
Alternatively, you could ask "what other editor is so complex that even after ten years of using it, you still don't know all its features?".
Some of us prefer to use powerful, but still relatively simple tools, so that we can get on with our jobs. I'm paid to design and develop software (amongst other related things), not to learn to use a tool.
for anyone in denial, try to log in as "Administrator" on a WinXP machine
I did so immediately after reading your comment. I'm somewhat confused - I was able to log in as "administrator" with no warnings, no fuss, and no loss of functionality. My normal user account, in fact, is a member of the administrators group (it's just easier that way).
Are you referring to the lack of an "administrator" icon on the welcome screen? If so, then you need to know 3 things:
1) the welcome screen can be switched off, leaving you with the "old-style" login prompt 2) from the welcome screen, if you hit ctrl-alt-delete a couple of times, the screen is replaced with the login prompt anyway 3) the welcome screen is not available on machines that are part of a domain
Perhaps before launching into your next anti-MS rant, you would do well to do a little research. There are plenty of reasons to bash them, but the supposed lack of ability to log into a machine running XP as the administrator account isn't one of them.
If someone uses GPL code, they have to make their mods to it available now, not after earning some supposed amount of money.
That's true, but that's not what this is about. The idea being discussed is that a company writes software and releases it under a proprietary licence. Then, once it has sold some certain number of units, they re-release it under an open source licence.
There's no question of using GPL code and releasing it closed-source - it's exactly the other way around. Taking an established, closed-source product and making it open source. Don't think "GPL violation", think Quake.
Shared Source has nothing to do with any new policy or some sudden change in behavior. Its just an attempt in keeping the remaining developers hooked onto MS and stop the massive stampeede onto linux and others.
What massive stampede? Linux is making strong headway in the server market, but there, it's mostly eating into Unix market share. I don't see any evidence of much more than a trickle of developers migrating as far as client development work is concerned. That will remain true as long as there aren't very many desktops running Linux, which will remain true as long as there aren't many client apps being developed...
Traditionally, the way to break into a market with an incumbent leader is to spend lots and lots of money, essentially buying support. Throw cash at development of your product (OS, console, whatever) and throw cash at developers to persuade them to develop for it. Without this, it takes a lot, lot longer to break into that market. Open source's only real advantage is that it doesn't have to make a profit any time soon (or at all, really), so it can afford to take the long route. That's not true of the companies supporting it, of course - they do need to make money.
I think developers will gradually make the switch, if only because Linux-based systems are cheaper, and offer greater freedom. Once sufficient developers switch that we start getting some high-quality desktop applications and a stable, consistent desktop, then users will start to switch. It's going to be a long, slow process, however - don't expect a migratory stampede any time soon.
As much as I love Linux, right now, I cannot make the switch entirely. There are still too many things that I can't use it for. I use it almost exclusively at work, with XP permanently running under VMWare, but very little at home, and it's the home users that you really need to convince, as they don't have on-tap support for when something breaks.
You also forgot.NET. I can't speak for any other programmers, but if I were to start coding Windows-based applications now, I'd do it in C# and.NET, rather than learning MFC.
I've looked at the MFC, and frankly, it looks horrible. Then again, I'm an OO type of guy; my primary language is currently Java, and I'm learning C#. I'm also learning DirectX rather than OpenGL, not out of any love for MS, but because OGL is C based. Sure, there are wrappers available, but that just adds another layer of complexity and more opportunity for bugs.
Anyway, the point was that I don't think that the remaining 30% would all be using MFC/WinAPI. Some would be doing Java on Windows, some.NET, some Delphi, some VB, etc. Even if you lump all the Windows-only ones under the "WinAPI" label, that still leaves Java and Delphi.
Once an easy to use, open document format is created, and the ability to read and write those documents is built into many programs, I think we will see an end of.DOC file attachements.
Easy to use, open document format, that includes support for rich media and formatting, with many programs able to read and write it?
Okay, fine, you make it illegal (with heavy penalties) for Americans to send spam. So, all the spam outfits move overseas, and you still get tons of it. You'll see a drop, of course, but hey, I'm in the UK and 99% of the spam I get is US in origin. The fact that I'm in a different country doesn't bother them now, why should it in the future?
Technology can't solve a social problem, but legislation can't slove a *global* social problem either.
Re:Right Direction, Wrong Implementation
on
Opencroquet
·
· Score: 1
So, are you saying that because the hardware isn't ready for this (or at least, the "average" user's hardware), that no-one should work on it at all?
Software advances often drive hardware ones. If it weren't for games like Quake, we'd have no need of 3d accelerators, and so may well not even have them.
Look at it this way - if people work on this sort of thing now, then we'll have some sort of (semi-)usable system ready for when we *do* have the hardware. It might also prompt hardware developers to start working out how to support this sort of thing - the two efforts could drive each other forwards.
At worst, someone other than myself is wasting their time; I for one have no problem with that.
Depending on what you're using your machine for, 20GB isn't huge. If you're a graphic artist, for instance, working on multi-megabyte images, you could quickly fill that space. Or scanning in your photo collection - at 10MB+ *per TIFF*, space disappears pretty quickly.
Hell, I don't do anything special, and I've filled about 40gig of my 50gig drive just with games, OS, prorams, and associated data.
On the other hand, drives are only getting bigger, so no, the space used by having several copies of each dll isn't a problem. I just wanted to point out that some people really *need* big drives, and not just for "mp3/pr0n collections".
"Weak and feeble" - and yet I for one cannot remember ever hearing about a company telling them where to go.
They may not have legal powers, but for now at least, the UK advertising industry listens to them. The alternative is for the government to take matters into their own hands and legislate, which the industry certainly won't want.
So, basically, you've not touched DirectX in what, 3 or 4 years?
I'm currently teaching myself DirectX 9 for C#, and so far I have no complaints. Well, one, I guess - the API documentation on MSDN is extremely terse, with a distinct lack of explanation. Still, Managed DirectX hasn't been out for very long (since late December), so I'm prepared to give it a little time. The C++ API docs look to be okay, although I've not really read them.
And what if you don't want to upgrade from version 2.0 to 13.0? Well, that's too bad, because 2.0 is no longer supported.
And so, of course, it magically stops working...
Okay, so perhaps v13 won't have the same API as v2. In a lot of cases, that's a good thing. That means that all the old, badly-thought-out stuff has been removed, and good riddance to it.
After all, you just said that DX3,4 and 5 were terrible - surely you *want* them to throw the API away and replace it with something better in a later release?
Not to mention into Earth orbit, or on to the asteriods or other planets (Mars, anyone?).
Mining extra-terrestrial bodies and bringing stuff back will only ever be economically viable if we can't (or won't) mine the stuff here. For building satellites and spacecraft, though, it makes far more sense to mine the Moon, rather than doing it down here at the bottom of the gravity well.
Go compare other professional IDE's at the street price, you'll realize that even then VS.NET is a great deal.
I can't comment on the price of VS.NET, but I can comment on the price of professional Java IDEs.
At my company, we use JBuilder Enterprise 7, which cost us about £1500 each as an upgrade from JB4 Pro (our previous IDE). We also have a single floating licence for TogetherJ's Control Centre, purchased a few years ago, which cost us about £10k. JProbe, a Java profiling tool, cost us something in that region.
They are all nothing, though.
LoadRunner, a Windows-based loadtesting suite, cost us a rumoured £45k about 3 years ago. I have no doubt that other "specialist" software is far more expensive.
So, yes, commercial IDEs are damn expensive - but compared to some software, they're effectively free.
Also, you have to consider this - they are aimed primarly at buisiness (especially any edition with "Professional", "Enterprise" or "Architect" in the title). For a business, if it saves each staff member a day or two's worth of time over its entire lifetime, it will have paid for itself. At the rate I'm usually charged out to clients at, my JB7 Enterprise cost my company about 1.5 day's worth of my time. It's saved me more than that, so it was money well spent.
All of those things can be bought singly or together, in varying quantities, for no more than a couple of dozen dollars/pounds/insert local currency here.
Now go see how much Word, XP, etc cost. They're a *different order of magnitude* in price.
I'm 28, own a house and a car, have a family, have a very well paid job, and *I* can't afford MS Office. Hell, I can't afford Visual C#.NET, which costs about 1/7th as much.
Slashdot Mirror
using fission power is not that popular
It would also be wholly impractical in this case.
What is required, as you say, is to raise the hydrogen to a suitably high temperature and pressure. The laser is used not only to increase the temperature of your target material, but also the pressure (the sudden increase in temperature causes the surface to ablate, which causes an increase in pressure on the remainder of the target). To achieve the same effect using a fission reaction, you would essentially have to create an explosion, which would be *far* harder to do safely. You're likely to just destory the reactor, as only a small fraction of the total output energy of the explosion could actually be directed at the target.
In an H-bomb, of course, the more destructive the reaction the better.
but how the hell can they harness that
In a word, magnets. The idea behind fusion, essentially, is that you raise a (hydrogen) plasma to sufficient temperature and pressure, and it will undergo fusion. If you get the conditions just right, it'll then continue to fuse once you've ignited it, thus supplying you with energy.
What you may not appreciate is that a plasma is electrically charged, and can therefore be contained using a suitable magnetic field. Arguably the most promising containment setup at the moment is the tokamak (from the Russian for bottle, iirc), which is a torus-shaped machine. Electromagnets around the torus create a circular magnetic field, which keeps the plasma contained in a ring. (My apologies if my information is out of date, I quit my PhD in plasma physics 4 years ago...)
Despite what the article says, however, fusion is not entirely pollution-free. One of the byproducts is a fairly large supply of neutrons. These neutrons are absorbed by the reactor, which will slowly but surely become radioactive. Therefore, you will eventually be left with radioactive waste to dispose of. You won't get anything like the quantity you get with fission, though, and fusion certainly doesn't produce any "conventional" pollution.
No one complains when ... the murdering bad guy in a movie gets killed
Perhaps that's because it's not real?
There's a hell of a difference between a demonstrably guilty movie bad guy being killed by the hero, and a real-life person being executed.
It's only availble to US citizens. What do you propose the rest of us do?
I upgraded to 512mb ram half a year ago, and Linux hasn't done a disk write since
I very much doubt that, unless you're writing absolutely nothing to the disk at all - no data, no log files, nothing.
Otherwise, your machine most certainly is writing to the disk - it's just doing it gradually, as and when there's nothing else for it to do, so you don't notice it.
Other than that, I agree - I have recently upgraded my machine, which included putting 512MB of RAM in it, and neither OS (Linux & XP Pro) noticeably writes to the disk, other than "on demand" (ie when I hit "save").
the eye has so many patterns, its almost impossible to fake
Interesting use of the word "almost" there. So, what happens if/when a method to fake arbitrary retinal patterns is developed?
The great thing about usernames and passwords, PINs, etc, is that if you have reason to think that someone else knows it, you can change it. You can't change your retinal pattern, or any other biometric data.
If someone manages to fake your thumb print or retinal pattern, you're stuffed - you can't change them, and if that's the only official method of identifcation, you can't even revoke them and use a PIN.
Besides, the problem isn't people managing to copy retinal patterns, its people managing to fool detectors into accepting other patterns, or even a photograph of the correct eye...
Not if the libraries in question are licenced under the LGPL. The LGPL specifically allows linking with non-GPLed code; that's the reason it exists distinct from the GPL.
Have you tried using MPlayer? I admit that I've not sat and watched a DVD all the way through (I have a 15" monitor, and a 32" widescreen TV...), but when I bought a DVD drive a few months ago I thought I'd give it a go.
Playing DVDs under Windows proved to be impossible. The OS refused to allow it, as it couldn't disable the TV-out socket on my grpahics card (Creative GeForce3 Ti200). So, I rebooted to Linux and had a go with MPlayer. I only played a few minutes worth, but it worked fine.
I think the original poster's point was that with an open source product, one could simply remove the key checking stage entirely, rather than having to rely on acquiring a key.
Alternatively, you could ask "what other editor is so complex that even after ten years of using it, you still don't know all its features?".
Some of us prefer to use powerful, but still relatively simple tools, so that we can get on with our jobs. I'm paid to design and develop software (amongst other related things), not to learn to use a tool.
for anyone in denial, try to log in as "Administrator" on a WinXP machine
I did so immediately after reading your comment. I'm somewhat confused - I was able to log in as "administrator" with no warnings, no fuss, and no loss of functionality. My normal user account, in fact, is a member of the administrators group (it's just easier that way).
Are you referring to the lack of an "administrator" icon on the welcome screen? If so, then you need to know 3 things:
1) the welcome screen can be switched off, leaving you with the "old-style" login prompt
2) from the welcome screen, if you hit ctrl-alt-delete a couple of times, the screen is replaced with the login prompt anyway
3) the welcome screen is not available on machines that are part of a domain
Perhaps before launching into your next anti-MS rant, you would do well to do a little research. There are plenty of reasons to bash them, but the supposed lack of ability to log into a machine running XP as the administrator account isn't one of them.
If someone uses GPL code, they have to make their mods to it available now, not after earning some supposed amount of money.
That's true, but that's not what this is about. The idea being discussed is that a company writes software and releases it under a proprietary licence. Then, once it has sold some certain number of units, they re-release it under an open source licence.
There's no question of using GPL code and releasing it closed-source - it's exactly the other way around. Taking an established, closed-source product and making it open source. Don't think "GPL violation", think Quake.
Shared Source has nothing to do with any new policy or some sudden change in behavior. Its just an attempt in keeping the remaining developers hooked onto MS and stop the massive stampeede onto linux and others.
What massive stampede? Linux is making strong headway in the server market, but there, it's mostly eating into Unix market share. I don't see any evidence of much more than a trickle of developers migrating as far as client development work is concerned. That will remain true as long as there aren't very many desktops running Linux, which will remain true as long as there aren't many client apps being developed...
Traditionally, the way to break into a market with an incumbent leader is to spend lots and lots of money, essentially buying support. Throw cash at development of your product (OS, console, whatever) and throw cash at developers to persuade them to develop for it. Without this, it takes a lot, lot longer to break into that market. Open source's only real advantage is that it doesn't have to make a profit any time soon (or at all, really), so it can afford to take the long route. That's not true of the companies supporting it, of course - they do need to make money.
I think developers will gradually make the switch, if only because Linux-based systems are cheaper, and offer greater freedom. Once sufficient developers switch that we start getting some high-quality desktop applications and a stable, consistent desktop, then users will start to switch. It's going to be a long, slow process, however - don't expect a migratory stampede any time soon.
As much as I love Linux, right now, I cannot make the switch entirely. There are still too many things that I can't use it for. I use it almost exclusively at work, with XP permanently running under VMWare, but very little at home, and it's the home users that you really need to convince, as they don't have on-tap support for when something breaks.
You also forgot .NET. I can't speak for any other programmers, but if I were to start coding Windows-based applications now, I'd do it in C# and .NET, rather than learning MFC.
.NET, some Delphi, some VB, etc. Even if you lump all the Windows-only ones under the "WinAPI" label, that still leaves Java and Delphi.
I've looked at the MFC, and frankly, it looks horrible. Then again, I'm an OO type of guy; my primary language is currently Java, and I'm learning C#. I'm also learning DirectX rather than OpenGL, not out of any love for MS, but because OGL is C based. Sure, there are wrappers available, but that just adds another layer of complexity and more opportunity for bugs.
Anyway, the point was that I don't think that the remaining 30% would all be using MFC/WinAPI. Some would be doing Java on Windows, some
Once an easy to use, open document format is created, and the ability to read and write those documents is built into many programs, I think we will see an end of .DOC file attachements.
Easy to use, open document format, that includes support for rich media and formatting, with many programs able to read and write it?
We have it already - it's called HTML.
What does legislation solve?
Okay, fine, you make it illegal (with heavy penalties) for Americans to send spam. So, all the spam outfits move overseas, and you still get tons of it. You'll see a drop, of course, but hey, I'm in the UK and 99% of the spam I get is US in origin. The fact that I'm in a different country doesn't bother them now, why should it in the future?
Technology can't solve a social problem, but legislation can't slove a *global* social problem either.
So, are you saying that because the hardware isn't ready for this (or at least, the "average" user's hardware), that no-one should work on it at all?
Software advances often drive hardware ones. If it weren't for games like Quake, we'd have no need of 3d accelerators, and so may well not even have them.
Look at it this way - if people work on this sort of thing now, then we'll have some sort of (semi-)usable system ready for when we *do* have the hardware. It might also prompt hardware developers to start working out how to support this sort of thing - the two efforts could drive each other forwards.
At worst, someone other than myself is wasting their time; I for one have no problem with that.
Depending on what you're using your machine for, 20GB isn't huge. If you're a graphic artist, for instance, working on multi-megabyte images, you could quickly fill that space. Or scanning in your photo collection - at 10MB+ *per TIFF*, space disappears pretty quickly.
Hell, I don't do anything special, and I've filled about 40gig of my 50gig drive just with games, OS, prorams, and associated data.
On the other hand, drives are only getting bigger, so no, the space used by having several copies of each dll isn't a problem. I just wanted to point out that some people really *need* big drives, and not just for "mp3/pr0n collections".
So Europe wants to go to the moon? Good for them. They can have second place
Second? Surely you mean third, at least as far as unmanned missions go. The Russians had unmanned probes on the Moon before the US did.
"Weak and feeble" - and yet I for one cannot remember ever hearing about a company telling them where to go.
They may not have legal powers, but for now at least, the UK advertising industry listens to them. The alternative is for the government to take matters into their own hands and legislate, which the industry certainly won't want.
So, basically, you've not touched DirectX in what, 3 or 4 years?
I'm currently teaching myself DirectX 9 for C#, and so far I have no complaints. Well, one, I guess - the API documentation on MSDN is extremely terse, with a distinct lack of explanation. Still, Managed DirectX hasn't been out for very long (since late December), so I'm prepared to give it a little time. The C++ API docs look to be okay, although I've not really read them.
And what if you don't want to upgrade from version 2.0 to 13.0? Well, that's too bad, because 2.0 is no longer supported.
And so, of course, it magically stops working...
Okay, so perhaps v13 won't have the same API as v2. In a lot of cases, that's a good thing. That means that all the old, badly-thought-out stuff has been removed, and good riddance to it.
After all, you just said that DX3,4 and 5 were terrible - surely you *want* them to throw the API away and replace it with something better in a later release?
Not to mention into Earth orbit, or on to the asteriods or other planets (Mars, anyone?).
Mining extra-terrestrial bodies and bringing stuff back will only ever be economically viable if we can't (or won't) mine the stuff here. For building satellites and spacecraft, though, it makes far more sense to mine the Moon, rather than doing it down here at the bottom of the gravity well.
Go compare other professional IDE's at the street price, you'll realize that even then VS.NET is a great deal.
I can't comment on the price of VS.NET, but I can comment on the price of professional Java IDEs.
At my company, we use JBuilder Enterprise 7, which cost us about £1500 each as an upgrade from JB4 Pro (our previous IDE). We also have a single floating licence for TogetherJ's Control Centre, purchased a few years ago, which cost us about £10k. JProbe, a Java profiling tool, cost us something in that region.
They are all nothing, though.
LoadRunner, a Windows-based loadtesting suite, cost us a rumoured £45k about 3 years ago. I have no doubt that other "specialist" software is far more expensive.
So, yes, commercial IDEs are damn expensive - but compared to some software, they're effectively free.
Also, you have to consider this - they are aimed primarly at buisiness (especially any edition with "Professional", "Enterprise" or "Architect" in the title). For a business, if it saves each staff member a day or two's worth of time over its entire lifetime, it will have paid for itself. At the rate I'm usually charged out to clients at, my JB7 Enterprise cost my company about 1.5 day's worth of my time. It's saved me more than that, so it was money well spent.
All of those things can be bought singly or together, in varying quantities, for no more than a couple of dozen dollars/pounds/insert local currency here.
Now go see how much Word, XP, etc cost. They're a *different order of magnitude* in price.
I'm 28, own a house and a car, have a family, have a very well paid job, and *I* can't afford MS Office. Hell, I can't afford Visual C#.NET, which costs about 1/7th as much.
Spell it?
I can't even pronounce it!
(And yes, I do listen to industrial stuff)