The post is pretty much standard trolling, but have the moderators gone on crack too? That's like say "Let's learn to drive a car. Let's start by assembling the engine..." and this is less of a problem in the workforce than on a collection of random computers. Every serious IT department runs recommended configurations of hardware and software, you don't just throw parts together, slap an OS on it and hope it works. Some hardware works flawlessly under Linux, others is a paperweight with every variation in between. If you want to run Linux you get hardware that runs Linux and it's not that hard to find, it's more that some brands support open source and others don't. Running it on random hardware is only done by people who want Linux to fail so they can mock it or those that really want the pain of dark magic command-fu or a nasty assignment in C. I really would like to see it as one of the assigned tasks though - run Linux natively from a LiveCD and run through basic checks on what works and what doesn't. That could be rather helpful information to someone trying to find a Linux friendly computer and Linux friendly accessories.
Pretty much everything down at the grocery store has competition without much regulation to speak of, it's just that some markets doesn't naturally lead to competition. Run multiple sets of sewage pipes is a great example that most see is extremely impractical. Forced leasing of lines and rackspace in centrals are vital to a working market, and yes the owner do make a good profit off it anyway. They can't just practically shut off people so that it's their service or no service, speed might not be better but you might get a different modem, different ToS, different support, in short dealing with another company if you want to switch. From those beginnings they will sometimes lay down their own lines in addition to leasing, making it a mixed market with what looks remarkably like competition. Funny to say it, but we socialists seems to have a better functioning market than the US...
"Excessive" tends to mean to the point where only a few are actually doing the work and lots are copying it for favors, status, friendship or plain generosity. It lets people with no clue sail by with no learning and will cause trouble either in later classes or real life. There's every reason to strike down on that.
In more technologically advanced countries the latest generation of broadband is plenty good at home. Even my parents and my uncle are moving to fiber connections now with 10/10 Mbit as the lowest, which is plenty and on upload even faster than my cable line. The whole "limited bandwidth" is going to be some oddity of the past in a few decades because even a fairly notorious HD hog such as myself doesn't download 100 GB/day which is what a saturated 10 Mbit line will upload. For comparison, a complete binary usenet feed is "only" about 500 Mbit/s and includes everything but the kitchen sink. Maybe live BluRay streaming is a little ways off still but then you're trying very hard to find the most marginal case possible.
Spare time? Most open source programmers get paid for their work, and quite well.
[citation needed]. I know it's true for the Linux kernel and probably some other server software, but there's huge amounts of open source that I would be very surprised if someone was getting paid for. Some of the things Sun has been funding has been things where I'm not sure anyone would pick up on, at least not as a whole product. Sure, there might be the odd company that wants a little feature in OpenOffice here and an addition to the standard library in java there but who'll be the ones making OpenOffice 4.0 or the next major version of java?
Just to pick one example, look at the implementation of >8 bit color/channel support in GIMP. It's been a work in progress for many, many years and is still not done in 2.8. Obviously it's a lot of work and touches very many areas of code, including many people that don't care about it though everyone agree it'd be a good feature to have. If there was a cathedral, it'd be done long ago because the order simply came from high up to make all parts of the application support it and people would do it or be fired. Instead some people are trying to run around the bazaar trying to make people switch over even though it's as annoying as driving on the right when everyone else is driving on the left.
Though I'm sure it's an oversimplification, but if Steve Jobs decides then Apple jumps. If whoever in Sun decides, Sun jumps. But you can't make the bazaar jump, and that is something you lose even if you somehow find enough people to fill the void.
On the other hand sticking to one's guns while the company goes down in flames might not be the best career choice either;-)
If the company would go bankrupt by paying for a Cisco certification, you'd better be polishing your resume anyway. If companies were rational, I would say that this is a bluff. The cost of going through a termination process, hiring process and lost productivity getting that person up to speed is huge and you never know what you'll get so firing an otherwise highly productive employee is insane. Still, under these circumstances companies often end up doing irrational things. And if there is a round of layoffs yoy can run into a downsizing policy that says non-certified people are let go first, so it's not entirely without rational risk either.
Personally I would politely decline, saying that with my other personal commitments I don't have the opportunity to do this on my time and dime. Don't discuss it, your personal life is none of their business just state it. Basically tell them that you'd love to continue working there, but that it's either on the job training, waive the requirement or they're indirectly asking you to start sending out resumes because your days with that company are numbered. Make it clear that the last is absolutely not what you want, but you're not seeing many other options.
The last one will make them show their true colors. If they really need you to get the certification, they'll pay for it. If they just tried to make you pay for it, they'll waive it and agree times are tough on everyone, please stay. If you are going to get laid off, you might get an early hint to find another job. Or you can simply be fired either right there or later, but in either case I doubt your career with that company would have been a long one. At the very least, I'd want it to be on their dime paying for course material and exams and that I could read it at work during spare moments.
Some self study would be an acceptable investment in my own career though, even our company will expect that for larger certifications - they're not mandatory in that sense though, but you'll never go from junior to senior developer or developer to architect without passing some. That I think is fairly reasonable, of course there's many job skills don't have clear certifications or where certifications don't make you qualified but overall it works out nice, as far as I know though all exams are paid for by the company.
P.S. I know that for certain huge investments, like funding/cofunding an MBA or something like that you have to commit yourself to working for the company for X months or pay them back. That I think is quite reasonable, otherwise people could get their degree and immediately apply somewhere else. I know for example that airline pilots are often this way, the airlines pay the pilot education and then you're bound to that company a while. I doubt a network certification falls into that category, maybe in the US though since the resignation period is just two weeks.
Flip that around and it sounds almost like you shouldn't get a wife and children because it gets in the way of your zealotry. If so, I say everyone is free to pursue their own happiness but if it was me that'd be going in the wrong direction. Throughout the world and as far as we've historically recorded people have found mates and managed to raise children on next to nothing, including men both ambitious and true to their beliefs. What's killing people is an imposed social standard so high they have to mortgage themselves into bankruptcy despite that a century ago you would raise the same number of children in a shack. Not that this is news, poverty has always been a big taboo but it's still weird to see how many refuse to live by their means and instead kill themselves on debt.
It's not nearly as complicated as it seems, "fjalla" means mountain and "jokull" glacier so a native would read it more like "the glacier of the mountain Eyja" or "Eyja Mountain Glacier". But like the Scandinavian and German language they build one long word out of it.
Oh please, latin was pretty much the universal language of Europe all through the Middle Ages. It was far more important for doctors to be able to talk precisely to all other doctors about the human body than to use common words - which doctors would when talking to patients anyway, just like today. Every bone, every muscle was found and well established in use long before English was any replacement. This is more like that many more IT people will understand "CPU" and "RAM" than who speak English. Common, world wide terms makes things so much easier. Doctors would be fools to try abolishing the latin.
Meh, if GameStation wants to fight the devil over it when I get sent down to the cellar, I'm not seeing too many downsides. Unless GameStation is run by Cthulhu, in which case it's the greater evil.
The real issue I think is, who wants an IP6-only Internet connection? NOBODY. Because despite everything, there's millions of applications and shit that won't work because they assume there's nothing but IPv4. You can pry my IPv4 address from my cold dead hands, being on IPv6 would be very close to being permanently behind NAT - you get out, nothing gets in. And if you're handing out a IPv4 address as well, you've gained nothing. I'm guessing someone at the bottom of some barrel somewhere end up taking it anyway because that's all there is, but it won't be in the first world countries. That is the only way it'll really happen beyond nice bullet points on how we should all go IPv6.
From what I gather, so is significantly changing an orbit and you'd have to do that twice, one to get it to the recycling point and once more to get it to whereever you want to go. Also in lower orbits you need thrusters to stay up or your orbit will decay, which means you can't just dump them somewhere because there's a constant fuel cost. From what I gather this is for when they're out of thrusters and the orbit is decaying, to speed up the deorbit. Without doing more it'd come down anyway, just not so quickly so there'd be much more space junk up there.
Because the context is a problem every time you mix computers and what you're doing on a computer. Let's say you record a CD, 16 bits/sample @ 44.1kHz. That's a bitrate of 16 * 44.1 = 705.6 kbit/s second right? If I want to send it over the LAN too? What if I need to allocate a memory buffer, is it still 705.6 kbit/s? And what if I want to store it to disk, do I need to allocate 705.6 kbit per second of music? Computers aren't not remotely consistent with themselves, a 100 Mbit LAN is 100,000,000 bits/second. Hard drives too but they're hardly the only ones, floppies weren't even consistent with themselves most being 1.44*1000*1024 bytes.
Things get confusing all the time because a 1 MB, 1 KHz (1024*1024*1000) bus is not equal to a 1 kB, 1MHz bus (1024*1000*1000) which is why everyone dealing with networks never used kilo = 1024. The 56k modem is 56,000 bits, ISDN is 64,000 bits and so on right up to SATA 6Gbit/s which is 6,000,000,000 Gbit/s (and even more confusing because it's in 8/10 bit encoding, but that's another story). So both inside and outside the machine we're switching between base 2 and base 10 all the time.
A particularly confusing item was codecs. Should they follow the "size" standard so a 128 kbit/s MP3 would take up 128 kbit/s, or the network standard so that a 128 kbit/s would take 128 kbit/s of network bandwidth? I think now most settled on k = 1000, that is to say if you encode a one second clip at 128 kbit/s it'll only take up 125 kbit on your disk. Confusing as fuck? Hell yeah. Let's just settle this and be done with it, with the i = base 2, without it base 10. Just forget the lame names, and let the prefixes do the talking. MB = megabyte, MiB = megabyte. That's what I'm doing at least.
Well, honestly I don't quite get this card. It's either a much cheaper Quatro or it's an overpriced, hot gaming card. It's six months after the Radeon 5xxx series launched, I wouldn't be very surprised if between this paper launch and actual availability AMD has binned up and announces the HD5890 to go head-to-head with nVidia for the title of fastest single gaming card again. Also AMD has managed to roll out a full series top-to-bottom of 40nm cards already, while I think nVidia will take a lot longer to trickle down.
Yes, there's those that need CUDA but I don't think the intersection between gamers and CUDA users is that big - and if gamers don't buy this card and the CUDA users buy this instead of the Quatros then it's a lose-lose proposition for nVidia. But I guess I see the outlines of the "new nVidia", they're just preparing to leave the "regular" GPU market sooner than I expected. With Intel including graphics on Atoms/Core i3/Core i5 and AMD heading for AMD Fusion then nVidia is being shut out of the market for integrated graphics. But there'll still be a decent market for discrete chips between that and the CUDA-focused cards.
This card looks like a misfit, like if you've put a truck engine in a sports car because they're both big, powerful engines. If AMD just focuses on the "average" discrete market and does that much better they got plenty to live off even if nVidia takes the CUDA market. I'm not so sure nVidia can sustain themselves just on being a niche high-end company. Maybe they can but most companies that have tried have been steamrolled by the huge volume and investment happening mainstream. This card to me at least looks like it's serving customers with one hand and packing its bags with the other.
Well, you could simply declare that multiple monitors is for losers and get a QuadHD (3840x2160) LCD instead, like say this one. It's only supposed to set you back 50,000$ or so. A 2160 cinema projector can easily set you back a few hundred thousands if that's not enough. There's always options if you have enough money...
The whole point of the 1$/day price is to make people that try to "game" the paper and only buy it on the good news days and not the slow ones pay more, while the weekly subscription makes you buy the whole week whether something interesting happens in the folllowing 6 days or not. The risk/value proposition is completely different and so your extrapolation is nonsense too. 104$/year is probably what they expect to make on a regular reader, maybe you can push it down to 80-90$ by paying for all year but there's no way you're getting another 2/3rds rebate just paying yearly instead of weekly.
They don't have to support an open source driver. If they would just publish specs the community could take care of implementing them.
1) You assume that there's a ready set of PDFs that could be uploaded somewhere. There's not, there's actually a mess of various documentation mixed in with tons of internal notes, foreign IP, trade secrets and stuff that was fixed in the driver and commented there but isn't really in a separate document at all. AMD has put a helluva lot of effort in creating a process to produce the documentation and get formal signoff from lawyers, technical experts and executives that this information is safe to release. Often they've given up on documentation and found it's easier to produce a clean code snippet and get that through the review - it's far from a trivial process.
2) Since it's normally a one-to-one hardware-driver combination, things get redone. A lot. Many things are simply removed and replaced by software, kind of like winmodems. It's not like you build a OpenGL 2.0 driver and next generation you have a working 2.0 driver on 3.0 hardware, you have to keep up with the changes to get any hardware support at all. It wouldn't be entirely impossible to do it from specs alone, but it would be difficult. In practice you need people in the project or available to the project to answer questions, correct documentation and work with the internal driver/hardware team. And/or possibly have some sort of NDA program in addition to the public specs, but all of this takes time and effort which equals money.
3) The community is quite frankly not that big. At last headcount there was about a dozen working on the AMD source drivers, of which three are AMD employees. I've heard Bridgman say they use 2-3% of the effort on Linux despite accounting for 1% of the sales, so a back-of-the-napkin calculation says the internal driver team is something like 100 people. With complete access to all the documentation including on unreleased products, the hardware designers, hardware simulators, early engineering samples and so on. So on top of all their other disadvantages, the community is vastly outnumbered. Not to dispute that they could do a lot more specs than without, just that there's a lot more missing than specs.
I have just one question: Why? If it's a choice between Catalyst (AMD's binary driver) and nVidia's binary driver, I'd take nVidia any day of the week on any platform. The two reasons I have a HD5850 in this machine is that a) It's a helluva fast card at a decent price for what it is and b) AMD has been opening their specifications and is building an open driver. Right now though my experience is that the binary driver under Windows is still buggy - far buggier than nVidia, and so far there's no 2D/3D acceleration with the open source drivers under Linux (for this series, r100-r700 has varying degrees of support) but I knew that before I purchased it. The driver is really the last reason I'd want to ditch nVidia, sure they are 100% closed source but being what it is it's very good. Not perfect or anything, but the competition certainly isn't either.
What bugs me about them is that they look just like any other quest. If they had a dollar sign above them, it wouldn't be nearly as annoying. But they absolutely want to give you the sales pitch.
That highly depends on how you look at it, the game as such is complete exactly as purchased including said code so all the 543 pieces are there. It's the publisher that isn't honoring it, and it doesn't say anywhere clearly say "good for one time only". What if the person who originally bought it tries installing it later on a completely different machine, will he get the same message? I would think so. In short, Gamespot shouldn't be held responsible for the publisher's misleading advertising. The closest thing you'd get to this in real life is a coupon for 2 free pieces, but in that case you'd be able to sell all 543+2 pieces. This game has a secret self-destruct button invoked at sale, and if that's allowed second hand sales will go to hell shortly. Hell, they might as well stop giving you a game, you're just get a 60$ code to unlock the game. Want to sell it second hand? "Sorry, this code has already been used."
The idea of this would be that if you can tell that somebody is a 40 year old man, and they are on a forum saying they are a 12 year old girl, you can flag that as being suspicious.
Somehow this sounds more like a tool for pedos to find out who's really a 12 year old girl and not a 40 year old man... Police sting? Anti-pedo vigilante? People just yanking your chain pretending? Aha... a real 12yo girl, jackpot. Not to mention I can imagine many really weird situations from false positives. There's so many holes it this logic there's more holes than logic.
Heh. In my experience algorithms like these are those that have been really, really optimized for a computer and would royally suck to understand without comments. Personally I must admit I don't know any sorting algorithms, I could probably reinvent the bubble sort or the B-tree - poorly. What I do know is that if I use a O(n^2) algorithm with n=1000 and time it to 1 second, then doing it with 100,000 records would take (100000/1000)^2*1 second = 10000 seconds. If I have problems, maybe I should use another algorithm but it's still not one I'd write. I just know there are different tradeoffs for append/update/find performance and should have a clue what I want. You absolutely need not need to understand hand-edited assembly to pick the right one.
There's always some people who need surgery NOW and whether they get cancer in 2-5-10-20 years or not it's still a win. I'm all for medical testing and not rushing out unsafe procedures, but the reactions I see are mostly knee-jerk "it's STEM cells, omg you can't" not based on real research. In fact, they don't want the research done in the first place. Of course it's highly experimental medically, so was heart transplants. The first guy survived two weeks, but today we average 15 years. Research can prove or disprove (ok, don't get all philosophical on me) whether it helps medically, but it won't matter because most of the resistance is due to the fanatic anti-abortion crowd which equate embryonic stem cells with unborn babies. And if it's not embryonic, they'll pretend there's no difference because only blind rage will do.
Slashdot Mirror
The post is pretty much standard trolling, but have the moderators gone on crack too? That's like say "Let's learn to drive a car. Let's start by assembling the engine..." and this is less of a problem in the workforce than on a collection of random computers. Every serious IT department runs recommended configurations of hardware and software, you don't just throw parts together, slap an OS on it and hope it works. Some hardware works flawlessly under Linux, others is a paperweight with every variation in between. If you want to run Linux you get hardware that runs Linux and it's not that hard to find, it's more that some brands support open source and others don't. Running it on random hardware is only done by people who want Linux to fail so they can mock it or those that really want the pain of dark magic command-fu or a nasty assignment in C. I really would like to see it as one of the assigned tasks though - run Linux natively from a LiveCD and run through basic checks on what works and what doesn't. That could be rather helpful information to someone trying to find a Linux friendly computer and Linux friendly accessories.
Pretty much everything down at the grocery store has competition without much regulation to speak of, it's just that some markets doesn't naturally lead to competition. Run multiple sets of sewage pipes is a great example that most see is extremely impractical. Forced leasing of lines and rackspace in centrals are vital to a working market, and yes the owner do make a good profit off it anyway. They can't just practically shut off people so that it's their service or no service, speed might not be better but you might get a different modem, different ToS, different support, in short dealing with another company if you want to switch. From those beginnings they will sometimes lay down their own lines in addition to leasing, making it a mixed market with what looks remarkably like competition. Funny to say it, but we socialists seems to have a better functioning market than the US...
When you finish your MBA- it'll all become clear.
After some cost/benefit analysis on the ideas above, I think yes. It's not going anywhere.
"Excessive" tends to mean to the point where only a few are actually doing the work and lots are copying it for favors, status, friendship or plain generosity. It lets people with no clue sail by with no learning and will cause trouble either in later classes or real life. There's every reason to strike down on that.
In more technologically advanced countries the latest generation of broadband is plenty good at home. Even my parents and my uncle are moving to fiber connections now with 10/10 Mbit as the lowest, which is plenty and on upload even faster than my cable line. The whole "limited bandwidth" is going to be some oddity of the past in a few decades because even a fairly notorious HD hog such as myself doesn't download 100 GB/day which is what a saturated 10 Mbit line will upload. For comparison, a complete binary usenet feed is "only" about 500 Mbit/s and includes everything but the kitchen sink. Maybe live BluRay streaming is a little ways off still but then you're trying very hard to find the most marginal case possible.
Spare time? Most open source programmers get paid for their work, and quite well.
[citation needed]. I know it's true for the Linux kernel and probably some other server software, but there's huge amounts of open source that I would be very surprised if someone was getting paid for. Some of the things Sun has been funding has been things where I'm not sure anyone would pick up on, at least not as a whole product. Sure, there might be the odd company that wants a little feature in OpenOffice here and an addition to the standard library in java there but who'll be the ones making OpenOffice 4.0 or the next major version of java?
Just to pick one example, look at the implementation of >8 bit color/channel support in GIMP. It's been a work in progress for many, many years and is still not done in 2.8. Obviously it's a lot of work and touches very many areas of code, including many people that don't care about it though everyone agree it'd be a good feature to have. If there was a cathedral, it'd be done long ago because the order simply came from high up to make all parts of the application support it and people would do it or be fired. Instead some people are trying to run around the bazaar trying to make people switch over even though it's as annoying as driving on the right when everyone else is driving on the left.
Though I'm sure it's an oversimplification, but if Steve Jobs decides then Apple jumps. If whoever in Sun decides, Sun jumps. But you can't make the bazaar jump, and that is something you lose even if you somehow find enough people to fill the void.
Whose turn it is to not fly for a week or so at a time will depend on the wind direction
At high altitudes it's pretty much always strong eastern winds due to Earth's rotation. So it'll be Europe being screwed at least 9 times out of 10.
On the other hand sticking to one's guns while the company goes down in flames might not be the best career choice either ;-)
If the company would go bankrupt by paying for a Cisco certification, you'd better be polishing your resume anyway. If companies were rational, I would say that this is a bluff. The cost of going through a termination process, hiring process and lost productivity getting that person up to speed is huge and you never know what you'll get so firing an otherwise highly productive employee is insane. Still, under these circumstances companies often end up doing irrational things. And if there is a round of layoffs yoy can run into a downsizing policy that says non-certified people are let go first, so it's not entirely without rational risk either.
Personally I would politely decline, saying that with my other personal commitments I don't have the opportunity to do this on my time and dime. Don't discuss it, your personal life is none of their business just state it. Basically tell them that you'd love to continue working there, but that it's either on the job training, waive the requirement or they're indirectly asking you to start sending out resumes because your days with that company are numbered. Make it clear that the last is absolutely not what you want, but you're not seeing many other options.
The last one will make them show their true colors. If they really need you to get the certification, they'll pay for it. If they just tried to make you pay for it, they'll waive it and agree times are tough on everyone, please stay. If you are going to get laid off, you might get an early hint to find another job. Or you can simply be fired either right there or later, but in either case I doubt your career with that company would have been a long one. At the very least, I'd want it to be on their dime paying for course material and exams and that I could read it at work during spare moments.
Some self study would be an acceptable investment in my own career though, even our company will expect that for larger certifications - they're not mandatory in that sense though, but you'll never go from junior to senior developer or developer to architect without passing some. That I think is fairly reasonable, of course there's many job skills don't have clear certifications or where certifications don't make you qualified but overall it works out nice, as far as I know though all exams are paid for by the company.
P.S. I know that for certain huge investments, like funding/cofunding an MBA or something like that you have to commit yourself to working for the company for X months or pay them back. That I think is quite reasonable, otherwise people could get their degree and immediately apply somewhere else. I know for example that airline pilots are often this way, the airlines pay the pilot education and then you're bound to that company a while. I doubt a network certification falls into that category, maybe in the US though since the resignation period is just two weeks.
Flip that around and it sounds almost like you shouldn't get a wife and children because it gets in the way of your zealotry. If so, I say everyone is free to pursue their own happiness but if it was me that'd be going in the wrong direction. Throughout the world and as far as we've historically recorded people have found mates and managed to raise children on next to nothing, including men both ambitious and true to their beliefs. What's killing people is an imposed social standard so high they have to mortgage themselves into bankruptcy despite that a century ago you would raise the same number of children in a shack. Not that this is news, poverty has always been a big taboo but it's still weird to see how many refuse to live by their means and instead kill themselves on debt.
It's not nearly as complicated as it seems, "fjalla" means mountain and "jokull" glacier so a native would read it more like "the glacier of the mountain Eyja" or "Eyja Mountain Glacier". But like the Scandinavian and German language they build one long word out of it.
Oh please, latin was pretty much the universal language of Europe all through the Middle Ages. It was far more important for doctors to be able to talk precisely to all other doctors about the human body than to use common words - which doctors would when talking to patients anyway, just like today. Every bone, every muscle was found and well established in use long before English was any replacement. This is more like that many more IT people will understand "CPU" and "RAM" than who speak English. Common, world wide terms makes things so much easier. Doctors would be fools to try abolishing the latin.
Meh, if GameStation wants to fight the devil over it when I get sent down to the cellar, I'm not seeing too many downsides. Unless GameStation is run by Cthulhu, in which case it's the greater evil.
The real issue I think is, who wants an IP6-only Internet connection? NOBODY. Because despite everything, there's millions of applications and shit that won't work because they assume there's nothing but IPv4. You can pry my IPv4 address from my cold dead hands, being on IPv6 would be very close to being permanently behind NAT - you get out, nothing gets in. And if you're handing out a IPv4 address as well, you've gained nothing. I'm guessing someone at the bottom of some barrel somewhere end up taking it anyway because that's all there is, but it won't be in the first world countries. That is the only way it'll really happen beyond nice bullet points on how we should all go IPv6.
From what I gather, so is significantly changing an orbit and you'd have to do that twice, one to get it to the recycling point and once more to get it to whereever you want to go. Also in lower orbits you need thrusters to stay up or your orbit will decay, which means you can't just dump them somewhere because there's a constant fuel cost. From what I gather this is for when they're out of thrusters and the orbit is decaying, to speed up the deorbit. Without doing more it'd come down anyway, just not so quickly so there'd be much more space junk up there.
Because the context is a problem every time you mix computers and what you're doing on a computer. Let's say you record a CD, 16 bits/sample @ 44.1kHz. That's a bitrate of 16 * 44.1 = 705.6 kbit/s second right? If I want to send it over the LAN too? What if I need to allocate a memory buffer, is it still 705.6 kbit/s? And what if I want to store it to disk, do I need to allocate 705.6 kbit per second of music? Computers aren't not remotely consistent with themselves, a 100 Mbit LAN is 100,000,000 bits/second. Hard drives too but they're hardly the only ones, floppies weren't even consistent with themselves most being 1.44*1000*1024 bytes.
Things get confusing all the time because a 1 MB, 1 KHz (1024*1024*1000) bus is not equal to a 1 kB, 1MHz bus (1024*1000*1000) which is why everyone dealing with networks never used kilo = 1024. The 56k modem is 56,000 bits, ISDN is 64,000 bits and so on right up to SATA 6Gbit/s which is 6,000,000,000 Gbit/s (and even more confusing because it's in 8/10 bit encoding, but that's another story). So both inside and outside the machine we're switching between base 2 and base 10 all the time.
A particularly confusing item was codecs. Should they follow the "size" standard so a 128 kbit/s MP3 would take up 128 kbit/s, or the network standard so that a 128 kbit/s would take 128 kbit/s of network bandwidth? I think now most settled on k = 1000, that is to say if you encode a one second clip at 128 kbit/s it'll only take up 125 kbit on your disk. Confusing as fuck? Hell yeah. Let's just settle this and be done with it, with the i = base 2, without it base 10. Just forget the lame names, and let the prefixes do the talking. MB = megabyte, MiB = megabyte. That's what I'm doing at least.
Well, honestly I don't quite get this card. It's either a much cheaper Quatro or it's an overpriced, hot gaming card. It's six months after the Radeon 5xxx series launched, I wouldn't be very surprised if between this paper launch and actual availability AMD has binned up and announces the HD5890 to go head-to-head with nVidia for the title of fastest single gaming card again. Also AMD has managed to roll out a full series top-to-bottom of 40nm cards already, while I think nVidia will take a lot longer to trickle down.
Yes, there's those that need CUDA but I don't think the intersection between gamers and CUDA users is that big - and if gamers don't buy this card and the CUDA users buy this instead of the Quatros then it's a lose-lose proposition for nVidia. But I guess I see the outlines of the "new nVidia", they're just preparing to leave the "regular" GPU market sooner than I expected. With Intel including graphics on Atoms/Core i3/Core i5 and AMD heading for AMD Fusion then nVidia is being shut out of the market for integrated graphics. But there'll still be a decent market for discrete chips between that and the CUDA-focused cards.
This card looks like a misfit, like if you've put a truck engine in a sports car because they're both big, powerful engines. If AMD just focuses on the "average" discrete market and does that much better they got plenty to live off even if nVidia takes the CUDA market. I'm not so sure nVidia can sustain themselves just on being a niche high-end company. Maybe they can but most companies that have tried have been steamrolled by the huge volume and investment happening mainstream. This card to me at least looks like it's serving customers with one hand and packing its bags with the other.
Well, you could simply declare that multiple monitors is for losers and get a QuadHD (3840x2160) LCD instead, like say this one. It's only supposed to set you back 50,000$ or so. A 2160 cinema projector can easily set you back a few hundred thousands if that's not enough. There's always options if you have enough money...
The whole point of the 1$/day price is to make people that try to "game" the paper and only buy it on the good news days and not the slow ones pay more, while the weekly subscription makes you buy the whole week whether something interesting happens in the folllowing 6 days or not. The risk/value proposition is completely different and so your extrapolation is nonsense too. 104$/year is probably what they expect to make on a regular reader, maybe you can push it down to 80-90$ by paying for all year but there's no way you're getting another 2/3rds rebate just paying yearly instead of weekly.
They don't have to support an open source driver. If they would just publish specs the community could take care of implementing them.
1) You assume that there's a ready set of PDFs that could be uploaded somewhere. There's not, there's actually a mess of various documentation mixed in with tons of internal notes, foreign IP, trade secrets and stuff that was fixed in the driver and commented there but isn't really in a separate document at all. AMD has put a helluva lot of effort in creating a process to produce the documentation and get formal signoff from lawyers, technical experts and executives that this information is safe to release. Often they've given up on documentation and found it's easier to produce a clean code snippet and get that through the review - it's far from a trivial process.
2) Since it's normally a one-to-one hardware-driver combination, things get redone. A lot. Many things are simply removed and replaced by software, kind of like winmodems. It's not like you build a OpenGL 2.0 driver and next generation you have a working 2.0 driver on 3.0 hardware, you have to keep up with the changes to get any hardware support at all. It wouldn't be entirely impossible to do it from specs alone, but it would be difficult. In practice you need people in the project or available to the project to answer questions, correct documentation and work with the internal driver/hardware team. And/or possibly have some sort of NDA program in addition to the public specs, but all of this takes time and effort which equals money.
3) The community is quite frankly not that big. At last headcount there was about a dozen working on the AMD source drivers, of which three are AMD employees. I've heard Bridgman say they use 2-3% of the effort on Linux despite accounting for 1% of the sales, so a back-of-the-napkin calculation says the internal driver team is something like 100 people. With complete access to all the documentation including on unreleased products, the hardware designers, hardware simulators, early engineering samples and so on. So on top of all their other disadvantages, the community is vastly outnumbered. Not to dispute that they could do a lot more specs than without, just that there's a lot more missing than specs.
I have just one question: Why? If it's a choice between Catalyst (AMD's binary driver) and nVidia's binary driver, I'd take nVidia any day of the week on any platform. The two reasons I have a HD5850 in this machine is that a) It's a helluva fast card at a decent price for what it is and b) AMD has been opening their specifications and is building an open driver. Right now though my experience is that the binary driver under Windows is still buggy - far buggier than nVidia, and so far there's no 2D/3D acceleration with the open source drivers under Linux (for this series, r100-r700 has varying degrees of support) but I knew that before I purchased it. The driver is really the last reason I'd want to ditch nVidia, sure they are 100% closed source but being what it is it's very good. Not perfect or anything, but the competition certainly isn't either.
What bugs me about them is that they look just like any other quest. If they had a dollar sign above them, it wouldn't be nearly as annoying. But they absolutely want to give you the sales pitch.
That highly depends on how you look at it, the game as such is complete exactly as purchased including said code so all the 543 pieces are there. It's the publisher that isn't honoring it, and it doesn't say anywhere clearly say "good for one time only". What if the person who originally bought it tries installing it later on a completely different machine, will he get the same message? I would think so. In short, Gamespot shouldn't be held responsible for the publisher's misleading advertising. The closest thing you'd get to this in real life is a coupon for 2 free pieces, but in that case you'd be able to sell all 543+2 pieces. This game has a secret self-destruct button invoked at sale, and if that's allowed second hand sales will go to hell shortly. Hell, they might as well stop giving you a game, you're just get a 60$ code to unlock the game. Want to sell it second hand? "Sorry, this code has already been used."
The idea of this would be that if you can tell that somebody is a 40 year old man, and they are on a forum saying they are a 12 year old girl, you can flag that as being suspicious.
Somehow this sounds more like a tool for pedos to find out who's really a 12 year old girl and not a 40 year old man... Police sting? Anti-pedo vigilante? People just yanking your chain pretending? Aha... a real 12yo girl, jackpot. Not to mention I can imagine many really weird situations from false positives. There's so many holes it this logic there's more holes than logic.
Heh. In my experience algorithms like these are those that have been really, really optimized for a computer and would royally suck to understand without comments. Personally I must admit I don't know any sorting algorithms, I could probably reinvent the bubble sort or the B-tree - poorly. What I do know is that if I use a O(n^2) algorithm with n=1000 and time it to 1 second, then doing it with 100,000 records would take (100000/1000)^2*1 second = 10000 seconds. If I have problems, maybe I should use another algorithm but it's still not one I'd write. I just know there are different tradeoffs for append/update/find performance and should have a clue what I want. You absolutely need not need to understand hand-edited assembly to pick the right one.
There's always some people who need surgery NOW and whether they get cancer in 2-5-10-20 years or not it's still a win. I'm all for medical testing and not rushing out unsafe procedures, but the reactions I see are mostly knee-jerk "it's STEM cells, omg you can't" not based on real research. In fact, they don't want the research done in the first place. Of course it's highly experimental medically, so was heart transplants. The first guy survived two weeks, but today we average 15 years. Research can prove or disprove (ok, don't get all philosophical on me) whether it helps medically, but it won't matter because most of the resistance is due to the fanatic anti-abortion crowd which equate embryonic stem cells with unborn babies. And if it's not embryonic, they'll pretend there's no difference because only blind rage will do.