You would be right if this weren't a prototype (it's not like they're selling it). What they demonstrated is that it's possible to do fusion outside of a tokamak (or similar device). From there, you can always try making the thing work at a higher scale, with less energy.
What??? The pgcc project basically died, though maybe some features were merged back to gcc. What you're probably refering to is the egcs project that forked gcc and proved so much better than it eventually replaced gcc. pgcc is (AFAIK) a version of gcc that was modified by Intel engineers to produce better code on the original Pentium.
OK, let's see who else is steeling out there... So OpenOffice is steeling from Microsoft because they're trying to read Office files. Of course, Microsoft also stole Windows from Apple by having a windowing system. Perhaps Apple also stole OS X by using BSD, which in turn stole from AT&T UNIX. Damn, the would be so much better if everyone stopped stealing!
Actually, I would say that there are types of organizations that can benefit from patents: 1) Big guy, as you mentioned, using patents to crush the innovative small guy; or 2) Small guy, as long as he doesn't produce anything useful. The example of this would companies like Eolas and many others that sue over stupid patents. The fact that they don't produce anything is important here, since it's the only way not to get countersued over patents by the big company you're suing.
I just read the definition of the definition of the CDDL. Outside of the fact that it's not compatible with the GPL (which can always be solved by dual-licensing), I find it reasonable. Of course, I may have missed some details, so I'd be happy if someone could point out what the problem is (outside of "it's from Sun" or "it's another license"). Also, I see that they distinguish between "Initial Developer" and "Contributor", but I don't see any difference between the rights given to one and the other. Can anyone enlighten me on this too?
Optical computers would transport information ten times faster than traditional electronic devices
That's simply not true. In "traditional electronic devices" signals travel faster than just 1/10 the speed of light (I think it's around 60%), so there's no way anything can transport information 10 times faster (unless they claim to go faster than light).
I think if the data is really sensitive, then there no way any deletion program would work. The reason is that sometimes the drive can detect bad sectors before they fail completely and copy it to a previously "hidden" sector. At this point, there no way you can access data on the "bad sector" to delete it.
The answer is that there would be no way to get a refund if it weren't for the EULA. When you buy a car, does your spare tire tell you "If you want to use me, you have to agree on this and that. Do you accept?". That's what Windows does, it lets you accept or refuse. If you chose to refuse (they can't force you to accept), then the product is completely useless, so you should be given a refund. Originally, I think the EULA even said something like "If you refuse, bring this back to your vendor" or something like that, but they removed it.
First, the DMCA is not (yet) law in all countries. Second, it's not clear whether this is illegal under the DMCA. Third, the tool was not made in the US. Fourth, use of the tool for fair use *is* legal under the DMCA.
When you're distributing GPL software without the source, you're violating copyright law. When you're listening some song you purchased on your Linux box (despite Apples attempts to remove your fair use rights) you are *not* violating copyright law. You're just working around (De-DRM) a workaround (DRM) on the copyright law.
Depends on the load. For most of my applications, I would say that the 64 KB L1 (32I + 32D) is suckage compared to the 256 L1 (128I + 128D) on AMD chips.
Except that what you describe will just place your site somewhere on the result page. With the 302, you can REPLACE the site you are targeting because google will think the original site is a duplicate.
I started using gnome when 0.13 was released because I already liked the desktop better than KDE at that time. Unfortunately, when gnome 2.0 came out, I found out that most features/options I found useful in gnome 1 were gone. It got so annoying that I ended up switching to KDE 3 (which I can't say I like that much). All I heard was along the lines of "this feature isn't part of a good user interface". As far as I'm concerned gnome 1 was much better than gnome 2 (except for anti-aliased fonts of course).
Some of the annoyances I'm talking about: - No more virtual desktops (switching from sawfish to metacity) - Panel auto-hide sucks (takes forever to come back again and time preference is gone) - New filemanager in 2.8 takes forever to open correctly just because I have lots of files in my home directory (that's on a 1.6 Pentium-M)...and many other
Sometimes I'm really wondering if the gnome developers are really using their desktops (without many custom fixes).
You may actually be right. Maybe what they wanted to say is that, given enought solar power, each square inch could produce up to 120 Watts (before saturating, melting, whatever). That means you can use mirrors and the like to reduce the size of the panel....or the number could just be completely off.
Of course, it depends on the CPU, the exact size and the operations performed, but I really doubt you can find a case where where 1 is 50 times faster. If you use prefetch carefully, bath can actually be completely equivalent.
Unfortunately nothing that I know of. The main thing that's missing is a good networking layer that goes through NATs. That's something I've wanted to do for a while, but my main expertise is in codecs (see sig). If anyone is interested (and serious) about making a VoIP that goes through NATs and the like, let me know.
I'm surprised nobody has really mentioned error correction. In the same way that correction codes can work around RAM unreliability, you could have checksums built into each instruction to detect and correct errors. You would basically trade speed for reliability, something that has existed in communications for decades (refering to Shannon's work). I don't see why it wouldn't be the same for CPUs. I also remember clearly Richard Feynman proposing the idea (sorry, don't remember which book), so the idea isn't exactly new.
Actually, you forgot the possibility that an event at time t-T caused both the event at time t and the prediction at time t*. Not that I really believe in that anyway...
Doctors, managers and many others are one type of job where you need to work "in real-time". For many other jobs, including science, engineering (not all cases) and art, the result matters a lot more than the time aspect. Great discoveries, inventions or work of art usually aren't done under pressure.
Slashdot Mirror
You would be right if this weren't a prototype (it's not like they're selling it). What they demonstrated is that it's possible to do fusion outside of a tokamak (or similar device). From there, you can always try making the thing work at a higher scale, with less energy.
What??? The pgcc project basically died, though maybe some features were merged back to gcc. What you're probably refering to is the egcs project that forked gcc and proved so much better than it eventually replaced gcc. pgcc is (AFAIK) a version of gcc that was modified by Intel engineers to produce better code on the original Pentium.
OK, let's see who else is steeling out there...
So OpenOffice is steeling from Microsoft because they're trying to read Office files. Of course, Microsoft also stole Windows from Apple by having a windowing system. Perhaps Apple also stole OS X by using BSD, which in turn stole from AT&T UNIX. Damn, the would be so much better if everyone stopped stealing!
Not to mention that he's likely bound by the free BK license saying he's not allowed to work on another SCM for a year.
There is no such clause in patent law. There are many companies that are only licensing patents to larger companies that actually build the product.
Actually, I would say that there are types of organizations that can benefit from patents:
1) Big guy, as you mentioned, using patents to crush the innovative small guy; or
2) Small guy, as long as he doesn't produce anything useful. The example of this would companies like Eolas and many others that sue over stupid patents. The fact that they don't produce anything is important here, since it's the only way not to get countersued over patents by the big company you're suing.
I see. Could you poing me to the actual place in the license that says so, because I wasn't able to find it just by looking quickly at the terms.
I just read the definition of the definition of the CDDL. Outside of the fact that it's not compatible with the GPL (which can always be solved by dual-licensing), I find it reasonable. Of course, I may have missed some details, so I'd be happy if someone could point out what the problem is (outside of "it's from Sun" or "it's another license"). Also, I see that they distinguish between "Initial Developer" and "Contributor", but I don't see any difference between the rights given to one and the other. Can anyone enlighten me on this too?
Optical computers would transport information ten times faster than traditional electronic devices
That's simply not true. In "traditional electronic devices" signals travel faster than just 1/10 the speed of light (I think it's around 60%), so there's no way anything can transport information 10 times faster (unless they claim to go faster than light).
I think if the data is really sensitive, then there no way any deletion program would work. The reason is that sometimes the drive can detect bad sectors before they fail completely and copy it to a previously "hidden" sector. At this point, there no way you can access data on the "bad sector" to delete it.
The answer is that there would be no way to get a refund if it weren't for the EULA. When you buy a car, does your spare tire tell you "If you want to use me, you have to agree on this and that. Do you accept?". That's what Windows does, it lets you accept or refuse. If you chose to refuse (they can't force you to accept), then the product is completely useless, so you should be given a refund. Originally, I think the EULA even said something like "If you refuse, bring this back to your vendor" or something like that, but they removed it.
Your second story has to be the funniest example of "wierd things your do when asleep" that I've every heard! Thanks for sharing.
First, the DMCA is not (yet) law in all countries. Second, it's not clear whether this is illegal under the DMCA. Third, the tool was not made in the US. Fourth, use of the tool for fair use *is* legal under the DMCA.
When you're distributing GPL software without the source, you're violating copyright law. When you're listening some song you purchased on your Linux box (despite Apples attempts to remove your fair use rights) you are *not* violating copyright law. You're just working around (De-DRM) a workaround (DRM) on the copyright law.
512K of L2 cache is suckage
Depends on the load. For most of my applications, I would say that the 64 KB L1 (32I + 32D) is suckage compared to the 256 L1 (128I + 128D) on AMD chips.
Except that what you describe will just place your site somewhere on the result page. With the 302, you can REPLACE the site you are targeting because google will think the original site is a duplicate.
I started using gnome when 0.13 was released because I already liked the desktop better than KDE at that time. Unfortunately, when gnome 2.0 came out, I found out that most features/options I found useful in gnome 1 were gone. It got so annoying that I ended up switching to KDE 3 (which I can't say I like that much). All I heard was along the lines of "this feature isn't part of a good user interface". As far as I'm concerned gnome 1 was much better than gnome 2 (except for anti-aliased fonts of course).
...and many other
Some of the annoyances I'm talking about:
- No more virtual desktops (switching from sawfish to metacity)
- Panel auto-hide sucks (takes forever to come back again and time preference is gone)
- New filemanager in 2.8 takes forever to open correctly just because I have lots of files in my home directory (that's on a 1.6 Pentium-M)
Sometimes I'm really wondering if the gnome developers are really using their desktops (without many custom fixes).
You may actually be right. Maybe what they wanted to say is that, given enought solar power, each square inch could produce up to 120 Watts (before saturating, melting, whatever). That means you can use mirrors and the like to reduce the size of the panel. ...or the number could just be completely off.
Of course, it depends on the CPU, the exact size and the operations performed, but I really doubt you can find a case where where 1 is 50 times faster. If you use prefetch carefully, bath can actually be completely equivalent.
I know that... and I don't see why we couldn't have an open-source app doing this too.
Unfortunately nothing that I know of. The main thing that's missing is a good networking layer that goes through NATs. That's something I've wanted to do for a while, but my main expertise is in codecs (see sig). If anyone is interested (and serious) about making a VoIP that goes through NATs and the like, let me know.
I'm surprised nobody has really mentioned error correction. In the same way that correction codes can work around RAM unreliability, you could have checksums built into each instruction to detect and correct errors. You would basically trade speed for reliability, something that has existed in communications for decades (refering to Shannon's work). I don't see why it wouldn't be the same for CPUs. I also remember clearly Richard Feynman proposing the idea (sorry, don't remember which book), so the idea isn't exactly new.
Actually, you forgot the possibility that an event at time t-T caused both the event at time t and the prediction at time t*. Not that I really believe in that anyway...
Doctors, managers and many others are one type of job where you need to work "in real-time". For many other jobs, including science, engineering (not all cases) and art, the result matters a lot more than the time aspect. Great discoveries, inventions or work of art usually aren't done under pressure.
Well, history shows that far fewer clone makers existed in the 486 era.
I don't quite remember how it was before, but for the 486, AMD had a pretty good clone that was (as far as I remember) both faster and cheaper.