Apparently this is somewhere between 2 months to half a year old. Someone on digg was trying to get ahold of these for mass purchase and failed to get any replies. I haven't seen any pictures floating around of this thing in the wild online. You can't buy them anywhere. Multiplied by the fact that BLX hasn't acquired a MIPS liscence (their CPU likely borrows certain patents originally designed for MIPS, given how similar it is), I doubt you'll see many of these floating around on eBay. Some theorize it's a grandstand by the Chinese government to demonstrate how much better they are than other local countries such as Korea, Taiwan or Japan. Plausible, but I suspect this is just the first visible vaporware company out of China.
In summary, it's not a 150 dollar device until you can acquire one somewhere at that price.
Interestingly, "Open Source" is one of the flows labelled in their final stage, "Market Enablement." I'm not sure what sort of "market opportunities" this company has in mind, but I hope it doesn't involve purchasing SCO's tactics.
Which is what this startup appears to have been designed to challenge. 400 million dollars worth of patents is a lot. The other people with this size of a portfolio claim it's mostly for defensive purposes, so if another company sues them, they can sue right back. Well no such luck here. They aren't actively selling or developing technology, and the fact that your patent looks very similar to another patent doesn't mean you're liable for damages.
Effectively, this is like those guys who abuse the rules of games to the point where it becomes obvious that the system, when taken to profitable extremes, is totally broken. Everyone has to either begin playing the game his way, or agree to fix the rules. All of the "too powerful" have products in addition to patents, so they'd much rather fix the rules than change their entire business to play by the methods the intelligent but otherwise intellectually unproductive have discovered. Obviously with a half billion to lose, this patent troll could be a potentially become a an entrenched power itself.
One reason is called "Research." I2 access can greatly improve access time to some incredibly large databases that are infeasible to ship. It also allows for teleconferencing classes, but that's kind of a toss up. That might not really be worth 300k a year though. However, a 2 million dollar grant to research high bandwidth internet technologies pretty much requires I2. And the physicists seem to love the I2 as well for research. I suspect that for most degree granting institutions, the costs far exceed the benefits of nonexistant research. However, I refrain from naming institutions that conduct no research "Universities."
The costs of I2 is rather small if you're actually using it for something more than a fast pipe for mirroring linux.isos and bittorrent. It's even smaller when you recieve a grant
While I agree with the overall attitude of your post, I am just reminding everyone that one of the primary reasons Einstein and the rest of those European scientists came to the U.S. was because they were trying to escape Nazi Germany.
You see, ACM finals require you to have a lot of practice in certain idiomatic programming problems and an ability to code map any new problem to one of the standards and code it up quickly.
This is actually a pretty useful CS skill, if you're faced with writing something more complicated than Accounts Receivable. The technique is called reduction, and requires you A) be familiar with a large set of problems solved by Dijkstra and a ton of dead men from the 50s and B) the efficiency of these. Being able to reduce a problem to one of those well studied ones has a number of benefits, not least of which is that you'll be able to demonstrate, when the boss approaches you, that the problem is intractable. Whether he believes or understands you, well, lets hope the world's best business people aren't German or something.
I'm curious to hear what you think it means to be very smart and good at CS, when reduction isn't part of that qualification.
What's that even get them? Do people read the headlines and think, "'First day sellout?' Shit, I need to get an 360!" ? But if the damn things really sold out, they'll just come back from the store empty handed. The math on your theory doesn't quite add up.
Loss leader economics only works when your production capacity is larger than the equalibrium price. Without the razor blade approach, there's a number of units to produce that maximized returns, and a set of numbers greater and fewer than that that result in less profit. The razor blade model says that it's okay to dip into the set of numbers larger than exceed the optimal production, because you'll benefit in the auxiliery market, video games sales. Ideally there is no shortage of the loss leader product-- selling out is a missed opportunity to expand the installed base. Yet shortages not only happened but persisted for some time (is the shortage even over? I see walmart selling online for 500 dollars =/). So while microsoft may have intended to be a loss leader, what we instead saw was the market raising prices in the face of scarcity. Microsoft was either unwilling or incapable of providing more 360s at the price listed. Effectively, they lost out on both the demand for the product at MSRP by not producing enough, and/or the profits they could have realized by pricing closer to the open market average.
Sure, you can say that they simply did far better than they expected, or that the shortages were intentional. But do you realize how many games you have to buy to justify the difference between 400 dollars and 800? My napkin math suggests somewhere between 8 and 20, depending on how much of each sale goes to MS. And this doesn't even include the fact that people who bought that JC Penny package still presumably wanted to purchase a game or two for it!
Re:I got my anti-windmill dvd in the mail last wee
on
Tilting At Windmills
·
· Score: 1
All the turbines I've seen appear to have the same aestetic as the ever popular "Apple knows how to make things visually appealing" ipod. When the county commission puts a six month moratorium on discussion, it becomes apparent that the cost of a pretty design is cheaper than the cost of spending a year doing nothing.
Avie leaving is only half the equation. Bringing out a new kernel that ditches mach would be a serious transition that you can't accomplish without hiring someone skilled in the design of such, or borrowing one of the open source systems. Neither of these options can be done in clandestine operation, no matter how hard Apple wants to silence the media. Simply put, building a new OSX requires Avie or someone like him.
Sure, one could imagine that Cringely is deliberately leaving out half of his argument to avoid suits and jeopardizing his sources, but claiming that Wine is somehow subject to deliberate breakage that doing the SAME DAMN thing in kernel is somehow immune from suggests he hasn't finished the digging on this one.
Even in "infinite supply" situations like selling music downloads, the pricing doesn't derive from trying to sell for as little as possible. You still need to work out the demand elasticity, the difference is that you're expressly trying to maximize revenues rather than fill two thousand seats. The reason companies have to move prices lower, towards the cost of production, is that they have competition. The question I have is whether there are any serious substitutes for a Madonna download. Of course we know the answer is "Yes, allofMP3 does a fantastic job offering a substitute for iTunes / whatever she uses." But as long as copyright is observed I don't think there's a lot of direct substitutes for a given song, which is why the music industry wants flexible pricing. They want to be rewarded for dumping millions into promotions of a single band, for paying independent promoters to get their music heard on the radio, for the spot on Saturday Night live, for the other TV and radio ads that make their old ways work.
As far as the relationship between concerts and music sales, your theory is as good as the others I've heard. At the very least, I won't have to subjected to the notion that "any concert is far better than any recorded album" quite as often.
You've got to be careful here. When you conduct a geniuine behavioral interview, you're asking a person to come up with a situation that matches your criteria, and then infer from their actions (and perhaps follow up questions on the event) their personality. The fundamental attribution error states that people generally overattribute a person's actions to their personality rather than the situation.
For example, an interviewer might attribute a candidate's decision to implement a genetic algorithm for the Frequency Assignment Problem as "over-engineering" rather than the fact that it's NP-Hard. Of course, a candidate who fails to get this across to the interviewer may not be the kind of person your company needs, but there are PLENTY other situations in which the situation defines what is to be done versus the person, and the behavioral interview rarely digs into the situation and focuses on a) checking that the person isn't lying by asking for details and b) examining the person's behavior and thought process during the scenario.
Arcade games came up with this in the early 80s, and Nintendo perfected it in the SNES. What I'm talking about is Super Mario Bros. This is a game where in five minutes you either finish a level (make progress) or time expires. If you're complaining about situations where don't have five minutes to play and nothing accomodates your needs, perhaps its time to check into a clinic;) This is probably the biggest problem with Zelda. Your average dungeon is epic in scope. It's fun when you have the time to dedicate to finishing (or don't but try anyways and ignore that paper due date), but if you only want to invest 30 minutes you may be SoL.
Note that episoidic gaming doesn't address this really. It's like getting a new dungeon every 30 days, while you still don't have 2 free contiguous hours to sit down and solve it. Some games address this with a save anywhere approach. Pokemon was like that. In five minutes you could at least make progress, if not accomplish anything exactly "meaningful." The original quake was save anywhere. And while it might not fit your definition of "meaningful" there are plenty of First Person Shooters with a time cap on multiplayer. Although, I'm pretty sure that rearranging a sequence of bits on a computer isn't the highest of human endevors anyways.
I think the last job title I want is "Mafioso Cryptologist;" once you've got them set up, the only thing left for them to do is solve the fact that you know about it.
I believe the gentlemen's point was that a man who needs to stop typing things is the wrong person to have type up a program to do speech to text conversions.
Two obvious reasons: 1) Ubuntu focuses its efforts on the desktop. Oracle's software is server oriented. The two could mix but there's many better options out there for Oracle.
2) Ubuntu is Shuttleworth's pet project. It's not his get rich quick scheme. Profitability is important, but I think he's more interested in Ubuntu actively contributing to the communities from which his fortune grew. Selling to anyone, Oracle included, would be an inadvertant admission that someone else is more qualified to direct how Ubuntu contributes than he is. And given Oracle's track record with acquisitions, I doubt someone who's number one priority isn't profit would sell to Oracle.
Well, their application benchmarks aren't much better. They MANUALLY played to a specific point 5 times in F.E.A.R., and in BF2. The only timedemo was in ut2k4. It would have been interesting to see them repeat their efforts in the same manner and compare it against the timedemo results.
The whole fucking point of the article was that 3d drivers are by and large, closed source. Yes, there are plenty of regular 2d xorg drivers, even for ati and nvidia. The reason this comes up now is that Xgl is starting to make its way into distros as an option, and aiglx isn't far behind. If the problem was simply getting a working screen, I'd agree with you. But that isn't the case. The problem is that the open source drivers for ati and nvidia won't perform well enough for the 3d accellerated desktops coming down the line. This is a recurrant discussion that will likely come up once again when nvidia releases their next driver with GLX_EXT_texture_from_pixmap support. It was supposed to be with "the next major driver release," but that was before 8756, so either that wasn't a major release or they changed their minds.
Well, the mount thing is actually pretty nice. The/media directory has the distinction of containing removable media, on behalf of users via pmount and hal. The benefits are twofold: you can't accidently kill off another predefined mount in/mnt, and if you're looking to create a list of available media sort of like My Computer, you've got an directory containing em. I suppose it can be confusing for people most comfortable with the command line, but I always liked puzzles;)
Screenshot 6 is just another member of the "lets ditch the 70s terminal concept where possible" parade. I'm sure you know that reboots are required to get a new kernel running. This is all that is. It's been around since at least breezy, and its as equally as non threatening as the reminders in the console to "run lilo before you reboot, and by the way, reboot soon!"
Personally, I'm much more scared of the Orangification. I better remember to save the current theme in case the new one is too painful!
Well, as long as your prefered method of playing music on a computer is patented, you'll be subjected to the wiles of a patent holder lording over who can distribute programs and how much it will cost them. I can't help it the pirate scene saw an interesting opportunity and jumped in head first, and the tech scene followed them into muddy waters. Nor can Ubuntu, as long as they intend to disitribute Ubuntu at NO COST to its users then they'll be at odds with people who insist their ideas must cost people.
It really isn't nessecary to reformat, I don't think, although it could be instructive for you to walk through the install process again. Just install ubunutu-desktop and you've installed ubuntu. Removing KDE / kubuntu is a bit more tricky, since apt-get doesn't keep track of things in quite the right manner to detect orphaned packages automatically. A fairly reliable way is to remove kdelibs -- if you want some of the neater kde apps like k3b, you can get them back later, but this should catch a lot of the things that kubuntu installs that regular ubuntu doesn't.
I'm running gcc4 on the current unbuntu stable. If you really need the older versions for compatibility reasons, they're also available. For example, nachOS by default doesn't compile on gcc4 (or gcc3.4), but 3.3 is available for my shitty needs.
Slashdot Mirror
Yeah, Ninjas are cool, Ninjas are mysterious, but face it: They're outdated. As outdated as the record industry, but they have a worse lobby.
Indeed, ninjas are out. Pirates are the future!
Apparently this is somewhere between 2 months to half a year old. Someone on digg was trying to get ahold of these for mass purchase and failed to get any replies. I haven't seen any pictures floating around of this thing in the wild online. You can't buy them anywhere. Multiplied by the fact that BLX hasn't acquired a MIPS liscence (their CPU likely borrows certain patents originally designed for MIPS, given how similar it is), I doubt you'll see many of these floating around on eBay. Some theorize it's a grandstand by the Chinese government to demonstrate how much better they are than other local countries such as Korea, Taiwan or Japan. Plausible, but I suspect this is just the first visible vaporware company out of China.
In summary, it's not a 150 dollar device until you can acquire one somewhere at that price.
Interestingly, "Open Source" is one of the flows labelled in their final stage, "Market Enablement." I'm not sure what sort of "market opportunities" this company has in mind, but I hope it doesn't involve purchasing SCO's tactics.
Which is what this startup appears to have been designed to challenge. 400 million dollars worth of patents is a lot. The other people with this size of a portfolio claim it's mostly for defensive purposes, so if another company sues them, they can sue right back. Well no such luck here. They aren't actively selling or developing technology, and the fact that your patent looks very similar to another patent doesn't mean you're liable for damages.
Effectively, this is like those guys who abuse the rules of games to the point where it becomes obvious that the system, when taken to profitable extremes, is totally broken. Everyone has to either begin playing the game his way, or agree to fix the rules. All of the "too powerful" have products in addition to patents, so they'd much rather fix the rules than change their entire business to play by the methods the intelligent but otherwise intellectually unproductive have discovered. Obviously with a half billion to lose, this patent troll could be a potentially become a an entrenched power itself.
One reason is called "Research." I2 access can greatly improve access time to some incredibly large databases that are infeasible to ship. It also allows for teleconferencing classes, but that's kind of a toss up. That might not really be worth 300k a year though. However, a 2 million dollar grant to research high bandwidth internet technologies pretty much requires I2. And the physicists seem to love the I2 as well for research. I suspect that for most degree granting institutions, the costs far exceed the benefits of nonexistant research. However, I refrain from naming institutions that conduct no research "Universities."
.isos and bittorrent. It's even smaller when you recieve a grant
The costs of I2 is rather small if you're actually using it for something more than a fast pipe for mirroring linux
While I agree with the overall attitude of your post, I am just reminding everyone that one of the primary reasons Einstein and the rest of those European scientists came to the U.S. was because they were trying to escape Nazi Germany.
"America: at least we're not Nazi Germany!"
You see, ACM finals require you to have a lot of practice in certain idiomatic programming problems and an ability to code map any new problem to one of the standards and code it up quickly.
This is actually a pretty useful CS skill, if you're faced with writing something more complicated than Accounts Receivable. The technique is called reduction, and requires you A) be familiar with a large set of problems solved by Dijkstra and a ton of dead men from the 50s and B) the efficiency of these. Being able to reduce a problem to one of those well studied ones has a number of benefits, not least of which is that you'll be able to demonstrate, when the boss approaches you, that the problem is intractable. Whether he believes or understands you, well, lets hope the world's best business people aren't German or something.
I'm curious to hear what you think it means to be very smart and good at CS, when reduction isn't part of that qualification.
What's that even get them? Do people read the headlines and think, "'First day sellout?' Shit, I need to get an 360!" ? But if the damn things really sold out, they'll just come back from the store empty handed. The math on your theory doesn't quite add up.
Loss leader economics only works when your production capacity is larger than the equalibrium price. Without the razor blade approach, there's a number of units to produce that maximized returns, and a set of numbers greater and fewer than that that result in less profit. The razor blade model says that it's okay to dip into the set of numbers larger than exceed the optimal production, because you'll benefit in the auxiliery market, video games sales. Ideally there is no shortage of the loss leader product-- selling out is a missed opportunity to expand the installed base. Yet shortages not only happened but persisted for some time (is the shortage even over? I see walmart selling online for 500 dollars =/). So while microsoft may have intended to be a loss leader, what we instead saw was the market raising prices in the face of scarcity. Microsoft was either unwilling or incapable of providing more 360s at the price listed. Effectively, they lost out on both the demand for the product at MSRP by not producing enough, and/or the profits they could have realized by pricing closer to the open market average.
Sure, you can say that they simply did far better than they expected, or that the shortages were intentional. But do you realize how many games you have to buy to justify the difference between 400 dollars and 800? My napkin math suggests somewhere between 8 and 20, depending on how much of each sale goes to MS. And this doesn't even include the fact that people who bought that JC Penny package still presumably wanted to purchase a game or two for it!
And this discourages monopolistic behaviors how?
All the turbines I've seen appear to have the same aestetic as the ever popular "Apple knows how to make things visually appealing" ipod. When the county commission puts a six month moratorium on discussion, it becomes apparent that the cost of a pretty design is cheaper than the cost of spending a year doing nothing.
Avie leaving is only half the equation. Bringing out a new kernel that ditches mach would be a serious transition that you can't accomplish without hiring someone skilled in the design of such, or borrowing one of the open source systems. Neither of these options can be done in clandestine operation, no matter how hard Apple wants to silence the media. Simply put, building a new OSX requires Avie or someone like him.
Sure, one could imagine that Cringely is deliberately leaving out half of his argument to avoid suits and jeopardizing his sources, but claiming that Wine is somehow subject to deliberate breakage that doing the SAME DAMN thing in kernel is somehow immune from suggests he hasn't finished the digging on this one.
Even in "infinite supply" situations like selling music downloads, the pricing doesn't derive from trying to sell for as little as possible. You still need to work out the demand elasticity, the difference is that you're expressly trying to maximize revenues rather than fill two thousand seats. The reason companies have to move prices lower, towards the cost of production, is that they have competition. The question I have is whether there are any serious substitutes for a Madonna download. Of course we know the answer is "Yes, allofMP3 does a fantastic job offering a substitute for iTunes / whatever she uses." But as long as copyright is observed I don't think there's a lot of direct substitutes for a given song, which is why the music industry wants flexible pricing. They want to be rewarded for dumping millions into promotions of a single band, for paying independent promoters to get their music heard on the radio, for the spot on Saturday Night live, for the other TV and radio ads that make their old ways work.
As far as the relationship between concerts and music sales, your theory is as good as the others I've heard. At the very least, I won't have to subjected to the notion that "any concert is far better than any recorded album" quite as often.
You've got to be careful here. When you conduct a geniuine behavioral interview, you're asking a person to come up with a situation that matches your criteria, and then infer from their actions (and perhaps follow up questions on the event) their personality. The fundamental attribution error states that people generally overattribute a person's actions to their personality rather than the situation.
For example, an interviewer might attribute a candidate's decision to implement a genetic algorithm for the Frequency Assignment Problem as "over-engineering" rather than the fact that it's NP-Hard. Of course, a candidate who fails to get this across to the interviewer may not be the kind of person your company needs, but there are PLENTY other situations in which the situation defines what is to be done versus the person, and the behavioral interview rarely digs into the situation and focuses on a) checking that the person isn't lying by asking for details and b) examining the person's behavior and thought process during the scenario.
Arcade games came up with this in the early 80s, and Nintendo perfected it in the SNES. What I'm talking about is Super Mario Bros. This is a game where in five minutes you either finish a level (make progress) or time expires. If you're complaining about situations where don't have five minutes to play and nothing accomodates your needs, perhaps its time to check into a clinic ;) This is probably the biggest problem with Zelda. Your average dungeon is epic in scope. It's fun when you have the time to dedicate to finishing (or don't but try anyways and ignore that paper due date), but if you only want to invest 30 minutes you may be SoL.
Note that episoidic gaming doesn't address this really. It's like getting a new dungeon every 30 days, while you still don't have 2 free contiguous hours to sit down and solve it. Some games address this with a save anywhere approach. Pokemon was like that. In five minutes you could at least make progress, if not accomplish anything exactly "meaningful." The original quake was save anywhere. And while it might not fit your definition of "meaningful" there are plenty of First Person Shooters with a time cap on multiplayer. Although, I'm pretty sure that rearranging a sequence of bits on a computer isn't the highest of human endevors anyways.
I think the last job title I want is "Mafioso Cryptologist;" once you've got them set up, the only thing left for them to do is solve the fact that you know about it.
I believe the gentlemen's point was that a man who needs to stop typing things is the wrong person to have type up a program to do speech to text conversions.
If Oracle wants their stuff to run on every possible system, then why's it so damn hard to get it up and running?
Two obvious reasons: 1) Ubuntu focuses its efforts on the desktop. Oracle's software is server oriented. The two could mix but there's many better options out there for Oracle.
2) Ubuntu is Shuttleworth's pet project. It's not his get rich quick scheme. Profitability is important, but I think he's more interested in Ubuntu actively contributing to the communities from which his fortune grew. Selling to anyone, Oracle included, would be an inadvertant admission that someone else is more qualified to direct how Ubuntu contributes than he is. And given Oracle's track record with acquisitions, I doubt someone who's number one priority isn't profit would sell to Oracle.
Well, their application benchmarks aren't much better. They MANUALLY played to a specific point 5 times in F.E.A.R., and in BF2. The only timedemo was in ut2k4. It would have been interesting to see them repeat their efforts in the same manner and compare it against the timedemo results.
The whole fucking point of the article was that 3d drivers are by and large, closed source. Yes, there are plenty of regular 2d xorg drivers, even for ati and nvidia. The reason this comes up now is that Xgl is starting to make its way into distros as an option, and aiglx isn't far behind. If the problem was simply getting a working screen, I'd agree with you. But that isn't the case. The problem is that the open source drivers for ati and nvidia won't perform well enough for the 3d accellerated desktops coming down the line. This is a recurrant discussion that will likely come up once again when nvidia releases their next driver with GLX_EXT_texture_from_pixmap support. It was supposed to be with "the next major driver release," but that was before 8756, so either that wasn't a major release or they changed their minds.
Well, the mount thing is actually pretty nice. The /media directory has the distinction of containing removable media, on behalf of users via pmount and hal. The benefits are twofold: you can't accidently kill off another predefined mount in /mnt, and if you're looking to create a list of available media sort of like My Computer, you've got an directory containing em. I suppose it can be confusing for people most comfortable with the command line, but I always liked puzzles ;)
Screenshot 6 is just another member of the "lets ditch the 70s terminal concept where possible" parade. I'm sure you know that reboots are required to get a new kernel running. This is all that is. It's been around since at least breezy, and its as equally as non threatening as the reminders in the console to "run lilo before you reboot, and by the way, reboot soon!"
Personally, I'm much more scared of the Orangification. I better remember to save the current theme in case the new one is too painful!
Well, as long as your prefered method of playing music on a computer is patented, you'll be subjected to the wiles of a patent holder lording over who can distribute programs and how much it will cost them. I can't help it the pirate scene saw an interesting opportunity and jumped in head first, and the tech scene followed them into muddy waters. Nor can Ubuntu, as long as they intend to disitribute Ubuntu at NO COST to its users then they'll be at odds with people who insist their ideas must cost people.
It really isn't nessecary to reformat, I don't think, although it could be instructive for you to walk through the install process again. Just install ubunutu-desktop and you've installed ubuntu. Removing KDE / kubuntu is a bit more tricky, since apt-get doesn't keep track of things in quite the right manner to detect orphaned packages automatically. A fairly reliable way is to remove kdelibs -- if you want some of the neater kde apps like k3b, you can get them back later, but this should catch a lot of the things that kubuntu installs that regular ubuntu doesn't.
I'm running gcc4 on the current unbuntu stable. If you really need the older versions for compatibility reasons, they're also available. For example, nachOS by default doesn't compile on gcc4 (or gcc3.4), but 3.3 is available for my shitty needs.