What are you smoking? Qt is sold on a per-developer basis, on a sliding cost scale. There are no royalties involved. It's very cheap as it is -- only about a grand or two per developer. That's about 1-2% of a programmer's annual salary. Thus, if he/she is 2% more productive, it pays for itself.
You are misunderstanding how interference works. Signal doesn't have to be on the exact frequency to interfere. There are always various harmonics, sidebands, and so on that get in the way. Furthermore, there's quite a bit of spill-over. You generally cannot filter this very well with filters because filters are good only up to a point. Any signal that leaks through the filter will cause enough interference to be a problem, and that's not even taking into effect all the harmonics and crap.
What are you smoking? There were other music stores YEARS before Apple's that did roughly the same thing. A pretty interface does not an invention make, and that's really the only new thing Apple brought to the table. Ever hear of PressPlay, MusicMatch, or RealOne?
If a handcrafted watch costs more than $100K and the factory-made one costs $5, then I would prefer the cheaper one. After all, if you buy a new $5 watch each year for the next 100 years you will still not spend $100k. And I can put up with a little error. Not many people buy handcrafted watches these days.
No shit, Sherlock, that's exactly what I said. "More difficult to contain" means the same as "harder to keep isolated". Learn some English before you reply.
Why would we just randomly release sterile organisms continuously? I agree, that is a bad idea, no matter how we look at it. Ecosystems are too complex to screw around with. However, breeding aquarium or food fish that live in a controlled environment is very different. Obviously, this precludes large numbers of them being released into the wild.
The grandparent post was talking about an accidental release that would be a single, extremely undesirable event. Clearly, unless such incidents occur very often and involve large number of fish the population won't be significantly affected.
As for the Purdue article: they did not use sterile fish. Their whole point was that the transgenic animals produced sexually favorable but weak offspring. If the animals are sterile, none of their conclusions would be applicable.
The gene *will* escape into the wild. In every single case studied, it escaped despite all precautions in production settings. Lab settings were more successful in keeping the lid on.
These are PLANTS we are talking about. Obviously it's much more difficult to contain plant pollen than it is to contain a zebrafish.
Furthermore, plants can reproduce much faster than zebrafish or other animals. Bt corn is an extreme case: it has been SPECIFICALLY engineered to survive better than normal corn in the wild.
Zebrafish are very fragile and an aquarium variety will not last very long in the wild. It will become sick, die, get eaten, or all of the above. I can not believe how many people forget this simple fact.
That's not a problem either, you dumbass. If the GE fish are sterile, they would eventually die and the problem will go away. Unless you release millions of the damn things, they aren't going to compete much with wild ones. Besides, following your logic, only the single largest fish will reproduce and the rest will not be able to. If it actually worked like that, salmon would not exist.
Re:I use good passwords, and here's how
on
Real Security?
·
· Score: 1
Ummm... You are assuming that the person cracking the password knows exactly how you generate it. If he/she doesn't, then it's as good as any random string.
The GPL says nothing about shipping proprietary stuff with the kernel. The incompatible kernel is created at runtime when the module is inserted, and is not distributed. I don't think that's a GPL violation, especially since many distributions include the commercial nvidia drivers.
Modules do not have to be GPL'd. Nvidia provides kernel modules that are proprietary, as do many others. In any case, you can always add hooks to the kernel and do stuff from userland. Either way should be pretty safe.
Let me guess: you've never actually tried to hook up your Mac to any company networks? If you actually had done that, you would see that many companies need much more than Samba in order to connect to the network.
How exactly would an improvement to SMTP stop spam? Why the fuck is everyone pushing SMTP as the problem? Really, SMTP is just fine as-is.
Spam is as fundamental a problem as junk mail and saying that SMTP causes spam is like saying that the shape of the envelope causes junk mail. SMTP has nothing to do with it. Really, in ANY e-mail system, if you can send someone an email, you can send them a junk e-mail. No improvements to any e-mail protocol are going to change that fundamental factor.
I don't think any programmer who actually creates cutting-edge software that pushes boundaries is in any risk of being outsourced. Outsourcing only works well when you are solving some common problem that's already been solved a few hundred times before. Indian programmers are not exactly known for their creativity or for innovative solutions.
Anyway, while we are on the topic of economics, the simple fact is that if your job can be done by somebody else for $1.00 an hour, that's how much you're worth. For programmers, the geographic location is not that important.
If Indian programmers are willing to do the job for cheaper, with equivalent quality, why should a company hire someone else?
The truth is, programming has evolved into a mindless occupation, like manufacturing, even though it is more akin to engineering (which has mostly stayed in the US). Perhaps that also explains the extremely poor quality of today's software.
Network transparency should not be the job of the window system. The toolkits should be the ones doing it, just like Windows does it. X relies on sending bitmaps over the network, which is slow and unnecessary. Hell, to have acceptable x performance you need at least a 10mbit connection with very low latency. Microsoft Terminal Services require orders of magnitude less bandwidth because it works at the toolkit level (as in, "draw a button and a drop-down menu" and not "draw this pixel, then this pixel, then that pixel...").
why dont you call your congressman and ask him to get up the FCC's ass and ask them why they are limiting wireless speeds ?
I can probably answer that question for you. The reason we cannot have fast wireless networking is because there isn't enough usable spectrum to go around. Just think about it: 100mbps of bandwidth requires more than 100MHz of bandwidth. Just to give you an idea, it would include everything from shortwave to the FM broadcast bands.
The only way you can get huge swaths of spectrum like that is at extremely high frequencies -- tens of GHz. And frequencies above a few GHz are strictly line-of-sight, and are usable mainly for satellites (as long as there aren't any trees in the way). You might as well use lasers or something.
Wireless should be left to what it does well -- mobile communications. Fixed wireless is pointless and inefficient.
on our cluster a tenth that size we blew 60 hard drives in the first 6 months and had to replace 10% of the motherboards.
Who the hell did you buy the cluster from? Emachines? Cluster nodes typically use power supplies, hard drives, memory, and motherboards that are of much higher quality than even the G5 (SCSI, quality ECC, reliable manufacturers, special climate-control systems, and so on). That's why they cost 10x more than the Big Mac. Just watch when the Mac cluster starts going through hard drives and mobos.
Slashdot Mirror
What are you smoking? Qt is sold on a per-developer basis, on a sliding cost scale. There are no royalties involved. It's very cheap as it is -- only about a grand or two per developer. That's about 1-2% of a programmer's annual salary. Thus, if he/she is 2% more productive, it pays for itself.
Look a Mac OS X. You get Aqua. no questions asked. And everyone seems to love it.
Of course. The 98.5% that don't like it use Windows.
Well, let's see.... How does that compare with Apple's offering? Oh wait, it's exactly the same.
You are misunderstanding how interference works. Signal doesn't have to be on the exact frequency to interfere. There are always various harmonics, sidebands, and so on that get in the way. Furthermore, there's quite a bit of spill-over. You generally cannot filter this very well with filters because filters are good only up to a point. Any signal that leaks through the filter will cause enough interference to be a problem, and that's not even taking into effect all the harmonics and crap.
What are you smoking? There were other music stores YEARS before Apple's that did roughly the same thing. A pretty interface does not an invention make, and that's really the only new thing Apple brought to the table. Ever hear of PressPlay, MusicMatch, or RealOne?
You are deluding yourself. IIT in India is widely considered to be the same or better than the top US Ivy League schools. And that's just one example.
If a handcrafted watch costs more than $100K and the factory-made one costs $5, then I would prefer the cheaper one. After all, if you buy a new $5 watch each year for the next 100 years you will still not spend $100k. And I can put up with a little error. Not many people buy handcrafted watches these days.
What are you smoking?
Go to Preferences -> Privacy and Security -> Images -> uncheck "Do not load remote images in Mail and Newsgroup messages". Voila.
I think that the pollen has an advantage here.
No shit, Sherlock, that's exactly what I said. "More difficult to contain" means the same as "harder to keep isolated". Learn some English before you reply.
Why would we just randomly release sterile organisms continuously? I agree, that is a bad idea, no matter how we look at it. Ecosystems are too complex to screw around with. However, breeding aquarium or food fish that live in a controlled environment is very different. Obviously, this precludes large numbers of them being released into the wild.
The grandparent post was talking about an accidental release that would be a single, extremely undesirable event. Clearly, unless such incidents occur very often and involve large number of fish the population won't be significantly affected.
As for the Purdue article: they did not use sterile fish. Their whole point was that the transgenic animals produced sexually favorable but weak offspring. If the animals are sterile, none of their conclusions would be applicable.
The gene *will* escape into the wild. In every single case studied, it escaped despite all precautions in production settings. Lab settings were more successful in keeping the lid on.
These are PLANTS we are talking about. Obviously it's much more difficult to contain plant pollen than it is to contain a zebrafish.
Furthermore, plants can reproduce much faster than zebrafish or other animals. Bt corn is an extreme case: it has been SPECIFICALLY engineered to survive better than normal corn in the wild.
Zebrafish are very fragile and an aquarium variety will not last very long in the wild. It will become sick, die, get eaten, or all of the above. I can not believe how many people forget this simple fact.
That's not a problem either, you dumbass. If the GE fish are sterile, they would eventually die and the problem will go away. Unless you release millions of the damn things, they aren't going to compete much with wild ones. Besides, following your logic, only the single largest fish will reproduce and the rest will not be able to. If it actually worked like that, salmon would not exist.
Ummm... You are assuming that the person cracking the password knows exactly how you generate it. If he/she doesn't, then it's as good as any random string.
SuSE, for one.
The GPL says nothing about shipping proprietary stuff with the kernel. The incompatible kernel is created at runtime when the module is inserted, and is not distributed. I don't think that's a GPL violation, especially since many distributions include the commercial nvidia drivers.
Modules do not have to be GPL'd. Nvidia provides kernel modules that are proprietary, as do many others. In any case, you can always add hooks to the kernel and do stuff from userland. Either way should be pretty safe.
Let me guess: you've never actually tried to hook up your Mac to any company networks? If you actually had done that, you would see that many companies need much more than Samba in order to connect to the network.
In that case, it's just a temporary problem that will simply go away once a few companies get seriously burned on it.
How exactly would an improvement to SMTP stop spam? Why the fuck is everyone pushing SMTP as the problem? Really, SMTP is just fine as-is.
Spam is as fundamental a problem as junk mail and saying that SMTP causes spam is like saying that the shape of the envelope causes junk mail. SMTP has nothing to do with it. Really, in ANY e-mail system, if you can send someone an email, you can send them a junk e-mail. No improvements to any e-mail protocol are going to change that fundamental factor.
I don't think any programmer who actually creates cutting-edge software that pushes boundaries is in any risk of being outsourced. Outsourcing only works well when you are solving some common problem that's already been solved a few hundred times before. Indian programmers are not exactly known for their creativity or for innovative solutions.
Anyway, while we are on the topic of economics, the simple fact is that if your job can be done by somebody else for $1.00 an hour, that's how much you're worth. For programmers, the geographic location is not that important.
If Indian programmers are willing to do the job for cheaper, with equivalent quality, why should a company hire someone else?
The truth is, programming has evolved into a mindless occupation, like manufacturing, even though it is more akin to engineering (which has mostly stayed in the US). Perhaps that also explains the extremely poor quality of today's software.
WTF is NX? Link please.
Network transparency should not be the job of the window system. The toolkits should be the ones doing it, just like Windows does it. X relies on sending bitmaps over the network, which is slow and unnecessary. Hell, to have acceptable x performance you need at least a 10mbit connection with very low latency. Microsoft Terminal Services require orders of magnitude less bandwidth because it works at the toolkit level (as in, "draw a button and a drop-down menu" and not "draw this pixel, then this pixel, then that pixel...").
why dont you call your congressman and ask him to get up the FCC's ass and ask them why they are limiting wireless speeds ?
I can probably answer that question for you. The reason we cannot have fast wireless networking is because there isn't enough usable spectrum to go around. Just think about it: 100mbps of bandwidth requires more than 100MHz of bandwidth. Just to give you an idea, it would include everything from shortwave to the FM broadcast bands.
The only way you can get huge swaths of spectrum like that is at extremely high frequencies -- tens of GHz. And frequencies above a few GHz are strictly line-of-sight, and are usable mainly for satellites (as long as there aren't any trees in the way). You might as well use lasers or something.
Wireless should be left to what it does well -- mobile communications. Fixed wireless is pointless and inefficient.
on our cluster a tenth that size we blew 60 hard drives in the first 6 months and had to replace 10% of the motherboards.
Who the hell did you buy the cluster from? Emachines? Cluster nodes typically use power supplies, hard drives, memory, and motherboards that are of much higher quality than even the G5 (SCSI, quality ECC, reliable manufacturers, special climate-control systems, and so on). That's why they cost 10x more than the Big Mac. Just watch when the Mac cluster starts going through hard drives and mobos.