Consumers can equate it just as readily to music, movies, broadcast media, free web sites, etc. Those are rife with just as many commercial interests as physical goods.
People who reject that quality, responsible, trustworthy code can be acquired without money being exchanged can be more readily be persuaded if you show where the money can/does flow in an understandable way than trying to out of the gate convince them of philosophical justifications that in a real world are insufficient to explain the degree of success of the free software movement.
It is unrealistic to expect that everyone that could be significantly important will do the appropriate legwork before making incorrect assertions. In this case, the response was the first opportunity for a non-student to provide active feedback and educational data. Teachers sadly are inherently distrustful of their students justifications, so this may have been the most opportune moment to prove that mature, intelligent people are behind the movement. Instead, she got a response that in her mind essentially proved her expectation of selfishness/immaturity. Even if her mind is unreachable, there is nothing to be gained from that response.
A lot of people want to focus on high ideals as motivation for Free software, and that's just not easy for most people to believe. Most people who do contribute either would not be able (no time, contracts forbidding) to or wouldn't want to without other conditions being met.
Is listening to the radio free? Watching broadcast television? Reading an article excerpt on the front page of a newspaper in a vending machine? Free software represents to people and corporations a good advertising mechanism. There often are services or other products that cost money and augment them.
Was going to high-school free? Not in the strictest sense, as tax money funds it, but the same applies to many Free software. Institutions often contribute software open-source in order to best serve the public trust. Given the nebulous nature of the funding (all taxpayers), open source is most often a best-fit model to reciprocate that investment in that specific scope.
If a repairman had a hard time with a particular bolt, and lent you a wrench and asked you to hold the nut as he tried to turn the bolt, would he charge you excess for access to the wrench? Of course not, he isn't running a tool rental business, it just happens in the course of his actual job. This sort of incidental work is common in the technology world. A company needs an email server. They aren't going to hire an army of developers to write from scratch, and they might not buy a commercial solution. They'll have their administrator download an Open Source email server and that administrator has no motivation to keep required code changes private. On the other hand, getting local modifications accepted upstream absolves them of maintenance efforts on a local patchset.
Generally, education has gaps, particularly in technology. Not all is bad, it went in phases. In Junior high, they explicitly called me into the Library whenever the DOS computers acted weird to get me to fix it. However, in my first high school days, I was disciplined for 'harming' the school's computers. Some examples of what I did that got me banned from using their computers: -Windows 3 displayed a blue screen, instructing to hit control-alt-delete. I did so. Evidently, their policy was to put an out-of-order sign and call the local computer company on a per-incident fee because that company told them those screens required such action. -On their new Win95 computer, I opened a full-screen DOS window. They claimed I had deleted the OS and I barely had time to exit and show them it was still there before they called that company again to fix it. -They had brand new deskjet printers that printed at minutes per page for simple text. I figured out their misconfiguration, and was called down for 'making the printers go too fast'. They said they were lucky they hadn't broken from going too fast and they called that company to 'fix' them back too slow (which they did all too readily, they knew how to exploit the ignorance).
For trying to develop and exercise my professional skillset of choice, I was actively precluded in instructing myself. My second high school refreshingly reverted to my junior high days of being explicitly called to assist the faculty.
As to Linux, I'm actually married to a teacher. Students were generally surprised to see Linux on the Desktop (didn't look like Mac or Windows) and the IT guy was happy to see a teacher using Linux. None of her peers would make this mistake.
All that said, the response was pretty dumb. don't be belligerent. You don't fix the problem by being an asshole. You provide education, links to the legal content of popular licenses and a layman's explanation. Provide reasonable motivations that lead to no-cost software development. Saying 'oh, MS bought you off' doesn't provide the requisite context to counter. Educational and other public institution contribution would be a good starting point, as it hits close to home. Corporate contribution in the name of marketing leverage, development costs (particularly for companies for whom the software is not their revenue source) and in order to obtain some government contracts would be another source perceived as both logical and quality. Finally, personal contributions for personal marketing (resume building) and hobbyist rounds out the major motivations. Mention companies like Dell, HP, and IBM doing open source to move hardware and services. Mention that even Microsoft invests in Novell and others due to their recognition of Linux as a legitimate market participant (assigning no value judgment to that, the statement is true regardless of whether you dislike or like the agreement). Mention that most supercomputers run the platform, many without paying explicitly for it.
You can craft a well-thought out, educational response that may actually spread in a positive way. Telling a teacher she is a bribed shill for MS is going to make her warn her peers in the teacher lounge more about this 'free' software rather than get her perhaps to discuss some interesting stuff she learned. You only have the get one teacher in a school interested enough to talk to get an entire school to at least basically understand Linux.
It's pervasive in software, the developers decide changing behavior without preserving the old should be fine, as their opinion is that the new behavior must be better.
Take, for an additional example, the 'keyhole'. They decided the same context menu should open up regardless of the forward or back button being clicked on. In fact, it is just one control instead of separate. It would be great if they had added this variant and let the user choose between the unified keyhole or the classic distinct buttons, but they forced the choice upon upgraders.
For one, you insinuate the parent poster must be willing to infringe on copyright just because they express disdain for SecuROM. There is no indication he chose to play a pirated game, just expressed the simple fact the DRM technology hurts legitimate users and dose nothing to faze illegitimate usage.
Though I don't agree with the method, the message is not much more clear for a game to have no piracy and no purchases. If a game releases, infested with invasive DRM mechanisms, and sees negligible sales, the company can't be sure whether the game flopped because the baggage that came with it was too burdensome, or if it flopped because the game just wasn't desirable. If it is accompanied by rampant copyright infringement, they can at least know their DRM didn't help their cause, and depending on the circumstances realize the product itself may have otherwise been successful. The Spore case is different as I believe the infringement incidents are artificially high due to a protest justification rather than genuine interest.
The company signed up for the most user inconvenience they could in the name of protected their intellectual property, and as a consequence, experienced the counter-intuitive result of a higher-than-normal percentage of piracy to legitimate sales.
I haven't bothered pirating anything as I'm just too lazy to care, however, I could see it being worth the effort to drive a point home.
If you care about any games, and know DRM afflicts the games you care about, and hear about the Spore debacle, you may wish to prove the DRM only hurt the industry in this case. You may not care about Spore (quite possible given the critical reception), but the opportunity is ripe to show that legitimate customers suffered for naught (given the ostensible reason would be to fight piracy and this proves it doesn't help).
For a given node count, we've seen increases in performance. The claimed problem is that for the workloads that concern these researchers, they don't see people mentioning significant enhancements to the fundamental memory architecture projected to follow the scale at which multi-core systems go. So you buy a 16 core chip system to upgrade your quad-core based system and hypothetically gain little despite the expense. Power efficencies drop and getting more performance requires more nodes. Additionally, who is to say that clock speeds won't lower if programming models in the mass market change such that distributed workloads are common and single-core performance isn't all that impressive.
All that said, talk beyond 6-core/8-core is mostly grandstanding at this time. As memory architecture for the mass market is not considered as intrinsically exciting, I would wager there will be advancements that no one speaks to. For example, Nehalem leapfrogs AMD memory bandwidth by a large margin (like by a factor of 2). It means if Shanghai parts are considered satisfactory today to get respectable yield memory wise to support four cores, Nehalem, by a particular metric, supports 8 equally satisfactorily. The whole picture is a tad more complicated (i.e. latency, numbers I don't know off hand), but the one metric is a highly important one in the supercomputer field.
For all the worry over memory bandwidth though, it hasn't stopped supercomputer purchasers from buying into Core2 all this time. Despite improvements in their chipset, Intel Core2 still doesn't reach AMD performance. Despite that, people spending money to get into the Top500 still chose to put their money on Core2 in general. Sure, Cray and IBM supercomputers in the Top2 used AMD, but from the time of its release, Core2 has decimated AMD supercomputer market share despite an inferior memory architecture.
I've never noticed poor text quality on the text stream subs. Must be your player of choice and configuration, not the fundamental platform on which it runs.
But I think two criteria are important: -Tendency for lettering to alternate between the hands -Most frequent words being typed using the home keys, home rows, and use of fingers tending towards the stronger ones.
I think dvorak does well enough, though I'm sure someone with letter/word frequency could come up with something at least marginally better than dvorak. Though since qwerty is ubiquitous, and I don't have to learn to retype.
More interesting statistics: The longest word I can type without leaving the home keys (not even the whole home row) is sensuousness. 231 words longer than six chars are possible with the dvorak home keys without moving 6 words longer than six chars on the qwerty home keys 1,091 words longer than 6 charsare possible with the dvorak home row 9 words longer with the qwerty home row.
1,139 words longer than six letters can be typed with left hand in qwerty (bad sign) 9 words longer than six letters can be typed with left hand in dvorak.
46 words longer than six letters can be typed in qwerty right hand. 0 words longer than six letters can by typed by dvorak right hand.
Never played around with this, but now that I have, it does kinda prove that dvorak does a good job of keeping the activity balanced between hands and on the home row.
You don't remove GUI overhead, it's still there, just unusable. Even 'Core' runs a gui. The amount of overhead may be low, but it's not fair to say the GUI is no overhead just because no monitor is connected.
For this, for MS my concerns would be: -Manageablitiy. MS is a pain in the ass to administer at scales of systems. -Flexibility. You are confined to MS vision as to how to use the platform. Looking at the rest of the top10, many are incapable of running windows due to the following characteristics: -Non-x86 hardware -No local storage or bootable SAN, using stateless images in RAM. That's just the basics, some projects go so far as a customized stack down to the kernel for certain tasks, to milk the absolute maximum out of the platform without any extraneous processes in the way.
Cost is an issue for the common configuration as well that isn't so fortune to be a bragging point. For this deployment, you can bet cost was an issue too, but in the opposite sense. I imagine MS paid the involved parties a significant sum to help subsidize this effort. Hardware vendors are notorious for this, and MS would be no different.
Microsoft is currently in the throw money at the market phase of trying to get adoption. They will help subsidize a cluster purchase even if it will effectively be all-the-time linux, so long as you will submit a Top500 score running their OS.
They moved to consumer space and Intel processors. The one impressive showing in the Top500 they had was probably at least in part to prove 'see, PPC is powerful, buy us instead of Intel systems!'. Now they don't need that market leverage to vindicate their desktop processor choice, and it's simply not worth it to them as it has no relevance to the high-margin market they are successfully pursuing.
It terms as 'barely even qualifies as 64-bit', that seems a strange characterization. I would say it either is or it isn't, not that it is 'barely'.
I could have sworn they claimed an Rpeak of 1.6 petaflops for Jaguar, yet Top500 lists it as 1.3. Or were they unable to get the whole configuration to run a single job in time?
Also, I guess Los Alamos must have added some rather than just tuned, because the Rpeak rose on Roadrunner (given the nature of Rpeak, only new hardware can explain increases.
The average time to crack it wouldn't require exhausting key space, the chances are very low that you would have to.
Also, the type of operations make a huge difference. Top500 is a specific set of 64-bit precision operations being measured. These systems may be much faster at the AES calculation.
That said, no matter how you fiddle the math, it comes out to a uselessly long time even assuming they had *one* protected piece of data they needed to function (given my laptop has about a dozen AES encrypted streams for utterly boring linux shells, the signal-to-noise of important encrypted data v. uninmportant is pretty against cracking too.
Technically impressive, absolute given, as the current top few offer unique technologies (Cray's Interconnect, IBM's processors).
From a profitability standpoint, undoubtedly the more successful a vendor is closer to the top of the list, the more they undoubtedly had to give up margin-wise for the bragging rights.
From a marketability standpoint, things get a tad awkward I think for the vendors at the top. IBM has Cell and BlueGene showcased, which we all understand can be used to great ends at the expense of a more complicated programming model (though HPC is already so parallel, they shouldn't be too bothered by this aspect of Cell), but in terms of more day-to-day operations, the sole bragging point relevant to the common market they get is that they know how to deliver and help architect 9,000 systems to work efficiently together (which is not a small feat, but the hardware itself has too high a barrier to get into the common market). For Cray, I honestly am surprised they have stuck around with only this niche market and still have held on. Their IO architecture is interesting, but of lower value at more common scales. These systems are certainly not cheap, but they are up against stiff competition willing to slice prices to grandstand about success.
The other configurations are made out of servers and processors and networking equipment that can more readily be used by a wider audience. So as cool as the exotic tech is, I think the commodity aspects are more compelling to the profitable customers for these vendors. Given the type of HP's share, HP may be quite ecstatic with the current list.
While the number itself is highly synthetic, it isn't necessarily out of date any more than any other single benchmark. HPCC attempts to score more stuff through a more varied test suite, but any one score of the suite is equally capable of being called irrelevant or uninteresting.
That said, there are a number of interesting items in the list. Cray's #2 showcased AMD's current generation's IO capabilities to the extent they've never been showcased before (right before they lose that exclusive benefit with Intel's Nehalem). IBM's non-x86 configurations show fantastic performance and even more drastic performance/watt (though I looked at the IBM website and they certainly price those QS22s up there, for a platform that is tricky to get the benefit out of).
But outside of exotic HT based interconnects and non-x86 servers, it is accurate to say mostly they are play-it-safe configurations that tie commodity equipment together with varying degrees of commodity networking equipment. It's a safe thing to do that people know how to do, and they can do it without making a unique custom solution the big companies wouldn't be able to productize on a small scale. I admit it is a shame, as I'd love to see more things along the lines of the Cray configuration, but none of the likely candidates seem to be willing to make that sort of investment.
I don't think IBM/Los Alamos suddenly plopped down something in response to cray at the last minute (frankly, I don't think you could move that fast).
Any hardware upgrades were almost certainly in plan, and if there were none, they've had 6 months of tuning to extract better numbers oout of what they had.
I wouldn't be surprised if in 6 months, the Cray without any additional hardware managed a better number than RoadRunner without additional hardware. However, such a victory is diminished somewhat by the energy the Cray undoubtedly consumes to acheive what performance they do get.
So first off, current incarnations of MS Office are considered the clear market leader. That's a fair observation. A traditionally installed local application manipulating files in a traditional way is popular.
OpenOffice.org is making inroads as a free alternative. More people are starting to find it a viable alternative for many circumstances, and opt not to explicitly buy MS Office. It behaves fundamentally the same way, and does basically the same stuff. Incidentally, I'm happy as it is a cross-platform application, but I think a greater portion of the userbase doesn't think about the source code or the cross-platform, they just didn't have to give money for it.
Then Google docs comes along. In terms of a strong brand to back the concept, it doesn't get much better than the word 'Google'. They find that despite the strong name and potential ability to fulfill at least the basic needs, people aren't excited about using it. The reason seems self-evident, people are more comfortable with traditional software models for this task. They feel they 'own' the software and have the most control over it. They may or may not back up to online storage, but they want to use a local application to edit it.
MS feels this means issuing their own webapp therefore would cement their lead. I think Google's failure indicates that such an offering is moot. People don't want subscription based software if non-subscription software can do the same thing or better. I've seen people throw out how it comes out cheaper in the long haul than buying the software every time, but it ignores the obvious, that people don't buy every iteration. I know people still using their copies of Office97 because they never had a reason to move. MS and many other companies hate this, but it is a simple fact.
It's absolutely clear IBM is contributing tons to Open Source. It's easy to note 'ibm.com' addresses in changelogs for tons of projects.
Generic Linux contributions and the success of Linux is key to them. You sell Linux to x86 users, you suddenly have a nice bridge to Linux on POWER. That seems to have been a pillar of their strategy.
While this is a good thought, and they also derive some recognition as general Linux experts, I'm suprised they haven't more directly capitalized on their investment. I would expect IBM to take ownership of a customer-facing distribution, and gain the ability to own the 'whole stack' in the x86 world. This has been their proud marketing message behind their power/mainframe systems, and yet they haven't applied much of that business or technical strategy to their x86 servers. By and large their x86 solutions aren't that much different from everybody else's. Linux could be a good puzzle piece to glue their middleware and hardware together.
Whether the result would be a good product or not, I cannot say, just surprised IBM hasn't done anything so intriguing in the x86 space.
Slashdot Mirror
Consumers can equate it just as readily to music, movies, broadcast media, free web sites, etc. Those are rife with just as many commercial interests as physical goods.
People who reject that quality, responsible, trustworthy code can be acquired without money being exchanged can be more readily be persuaded if you show where the money can/does flow in an understandable way than trying to out of the gate convince them of philosophical justifications that in a real world are insufficient to explain the degree of success of the free software movement.
It is unrealistic to expect that everyone that could be significantly important will do the appropriate legwork before making incorrect assertions. In this case, the response was the first opportunity for a non-student to provide active feedback and educational data. Teachers sadly are inherently distrustful of their students justifications, so this may have been the most opportune moment to prove that mature, intelligent people are behind the movement. Instead, she got a response that in her mind essentially proved her expectation of selfishness/immaturity. Even if her mind is unreachable, there is nothing to be gained from that response.
A lot of people want to focus on high ideals as motivation for Free software, and that's just not easy for most people to believe. Most people who do contribute either would not be able (no time, contracts forbidding) to or wouldn't want to without other conditions being met.
Is listening to the radio free? Watching broadcast television? Reading an article excerpt on the front page of a newspaper in a vending machine? Free software represents to people and corporations a good advertising mechanism. There often are services or other products that cost money and augment them.
Was going to high-school free? Not in the strictest sense, as tax money funds it, but the same applies to many Free software. Institutions often contribute software open-source in order to best serve the public trust. Given the nebulous nature of the funding (all taxpayers), open source is most often a best-fit model to reciprocate that investment in that specific scope.
If a repairman had a hard time with a particular bolt, and lent you a wrench and asked you to hold the nut as he tried to turn the bolt, would he charge you excess for access to the wrench? Of course not, he isn't running a tool rental business, it just happens in the course of his actual job. This sort of incidental work is common in the technology world. A company needs an email server. They aren't going to hire an army of developers to write from scratch, and they might not buy a commercial solution. They'll have their administrator download an Open Source email server and that administrator has no motivation to keep required code changes private. On the other hand, getting local modifications accepted upstream absolves them of maintenance efforts on a local patchset.
Generally, education has gaps, particularly in technology. Not all is bad, it went in phases. In Junior high, they explicitly called me into the Library whenever the DOS computers acted weird to get me to fix it. However, in my first high school days, I was disciplined for 'harming' the school's computers. Some examples of what I did that got me banned from using their computers:
-Windows 3 displayed a blue screen, instructing to hit control-alt-delete. I did so. Evidently, their policy was to put an out-of-order sign and call the local computer company on a per-incident fee because that company told them those screens required such action.
-On their new Win95 computer, I opened a full-screen DOS window. They claimed I had deleted the OS and I barely had time to exit and show them it was still there before they called that company again to fix it.
-They had brand new deskjet printers that printed at minutes per page for simple text. I figured out their misconfiguration, and was called down for 'making the printers go too fast'. They said they were lucky they hadn't broken from going too fast and they called that company to 'fix' them back too slow (which they did all too readily, they knew how to exploit the ignorance).
For trying to develop and exercise my professional skillset of choice, I was actively precluded in instructing myself. My second high school refreshingly reverted to my junior high days of being explicitly called to assist the faculty.
As to Linux, I'm actually married to a teacher. Students were generally surprised to see Linux on the Desktop (didn't look like Mac or Windows) and the IT guy was happy to see a teacher using Linux. None of her peers would make this mistake.
All that said, the response was pretty dumb. don't be belligerent. You don't fix the problem by being an asshole. You provide education, links to the legal content of popular licenses and a layman's explanation. Provide reasonable motivations that lead to no-cost software development. Saying 'oh, MS bought you off' doesn't provide the requisite context to counter. Educational and other public institution contribution would be a good starting point, as it hits close to home. Corporate contribution in the name of marketing leverage, development costs (particularly for companies for whom the software is not their revenue source) and in order to obtain some government contracts would be another source perceived as both logical and quality. Finally, personal contributions for personal marketing (resume building) and hobbyist rounds out the major motivations. Mention companies like Dell, HP, and IBM doing open source to move hardware and services. Mention that even Microsoft invests in Novell and others due to their recognition of Linux as a legitimate market participant (assigning no value judgment to that, the statement is true regardless of whether you dislike or like the agreement). Mention that most supercomputers run the platform, many without paying explicitly for it.
You can craft a well-thought out, educational response that may actually spread in a positive way. Telling a teacher she is a bribed shill for MS is going to make her warn her peers in the teacher lounge more about this 'free' software rather than get her perhaps to discuss some interesting stuff she learned. You only have the get one teacher in a school interested enough to talk to get an entire school to at least basically understand Linux.
It's pervasive in software, the developers decide changing behavior without preserving the old should be fine, as their opinion is that the new behavior must be better.
Take, for an additional example, the 'keyhole'. They decided the same context menu should open up regardless of the forward or back button being clicked on. In fact, it is just one control instead of separate. It would be great if they had added this variant and let the user choose between the unified keyhole or the classic distinct buttons, but they forced the choice upon upgraders.
For one, you insinuate the parent poster must be willing to infringe on copyright just because they express disdain for SecuROM. There is no indication he chose to play a pirated game, just expressed the simple fact the DRM technology hurts legitimate users and dose nothing to faze illegitimate usage.
Though I don't agree with the method, the message is not much more clear for a game to have no piracy and no purchases. If a game releases, infested with invasive DRM mechanisms, and sees negligible sales, the company can't be sure whether the game flopped because the baggage that came with it was too burdensome, or if it flopped because the game just wasn't desirable. If it is accompanied by rampant copyright infringement, they can at least know their DRM didn't help their cause, and depending on the circumstances realize the product itself may have otherwise been successful. The Spore case is different as I believe the infringement incidents are artificially high due to a protest justification rather than genuine interest.
The company signed up for the most user inconvenience they could in the name of protected their intellectual property, and as a consequence, experienced the counter-intuitive result of a higher-than-normal percentage of piracy to legitimate sales.
I haven't bothered pirating anything as I'm just too lazy to care, however, I could see it being worth the effort to drive a point home.
If you care about any games, and know DRM afflicts the games you care about, and hear about the Spore debacle, you may wish to prove the DRM only hurt the industry in this case. You may not care about Spore (quite possible given the critical reception), but the opportunity is ripe to show that legitimate customers suffered for naught (given the ostensible reason would be to fight piracy and this proves it doesn't help).
For a given node count, we've seen increases in performance. The claimed problem is that for the workloads that concern these researchers, they don't see people mentioning significant enhancements to the fundamental memory architecture projected to follow the scale at which multi-core systems go. So you buy a 16 core chip system to upgrade your quad-core based system and hypothetically gain little despite the expense. Power efficencies drop and getting more performance requires more nodes. Additionally, who is to say that clock speeds won't lower if programming models in the mass market change such that distributed workloads are common and single-core performance isn't all that impressive.
All that said, talk beyond 6-core/8-core is mostly grandstanding at this time. As memory architecture for the mass market is not considered as intrinsically exciting, I would wager there will be advancements that no one speaks to. For example, Nehalem leapfrogs AMD memory bandwidth by a large margin (like by a factor of 2). It means if Shanghai parts are considered satisfactory today to get respectable yield memory wise to support four cores, Nehalem, by a particular metric, supports 8 equally satisfactorily. The whole picture is a tad more complicated (i.e. latency, numbers I don't know off hand), but the one metric is a highly important one in the supercomputer field.
For all the worry over memory bandwidth though, it hasn't stopped supercomputer purchasers from buying into Core2 all this time. Despite improvements in their chipset, Intel Core2 still doesn't reach AMD performance. Despite that, people spending money to get into the Top500 still chose to put their money on Core2 in general. Sure, Cray and IBM supercomputers in the Top2 used AMD, but from the time of its release, Core2 has decimated AMD supercomputer market share despite an inferior memory architecture.
I've never noticed poor text quality on the text stream subs. Must be your player of choice and configuration, not the fundamental platform on which it runs.
But I think two criteria are important:
-Tendency for lettering to alternate between the hands
-Most frequent words being typed using the home keys, home rows, and use of fingers tending towards the stronger ones.
I think dvorak does well enough, though I'm sure someone with letter/word frequency could come up with something at least marginally better than dvorak. Though since qwerty is ubiquitous, and I don't have to learn to retype.
More interesting statistics:
The longest word I can type without leaving the home keys (not even the whole home row) is sensuousness.
231 words longer than six chars are possible with the dvorak home keys without moving
6 words longer than six chars on the qwerty home keys
1,091 words longer than 6 charsare possible with the dvorak home row
9 words longer with the qwerty home row.
1,139 words longer than six letters can be typed with left hand in qwerty (bad sign)
9 words longer than six letters can be typed with left hand in dvorak.
46 words longer than six letters can be typed in qwerty right hand.
0 words longer than six letters can by typed by dvorak right hand.
Never played around with this, but now that I have, it does kinda prove that dvorak does a good job of keeping the activity balanced between hands and on the home row.
It was in china, they just walked to the corner and picked up a few thousand copies of Windows and probably paid 5 to 10 bucks.
You don't remove GUI overhead, it's still there, just unusable. Even 'Core' runs a gui. The amount of overhead may be low, but it's not fair to say the GUI is no overhead just because no monitor is connected.
For this, for MS my concerns would be:
-Manageablitiy. MS is a pain in the ass to administer at scales of systems.
-Flexibility. You are confined to MS vision as to how to use the platform. Looking at the rest of the top10, many are incapable of running windows due to the following characteristics:
-Non-x86 hardware
-No local storage or bootable SAN, using stateless images in RAM.
That's just the basics, some projects go so far as a customized stack down to the kernel for certain tasks, to milk the absolute maximum out of the platform without any extraneous processes in the way.
Cost is an issue for the common configuration as well that isn't so fortune to be a bragging point. For this deployment, you can bet cost was an issue too, but in the opposite sense. I imagine MS paid the involved parties a significant sum to help subsidize this effort. Hardware vendors are notorious for this, and MS would be no different.
Microsoft is currently in the throw money at the market phase of trying to get adoption. They will help subsidize a cluster purchase even if it will effectively be all-the-time linux, so long as you will submit a Top500 score running their OS.
They moved to consumer space and Intel processors. The one impressive showing in the Top500 they had was probably at least in part to prove 'see, PPC is powerful, buy us instead of Intel systems!'. Now they don't need that market leverage to vindicate their desktop processor choice, and it's simply not worth it to them as it has no relevance to the high-margin market they are successfully pursuing.
It terms as 'barely even qualifies as 64-bit', that seems a strange characterization. I would say it either is or it isn't, not that it is 'barely'.
I could have sworn they claimed an Rpeak of 1.6 petaflops for Jaguar, yet Top500 lists it as 1.3. Or were they unable to get the whole configuration to run a single job in time?
Also, I guess Los Alamos must have added some rather than just tuned, because the Rpeak rose on Roadrunner (given the nature of Rpeak, only new hardware can explain increases.
Not to diminish the scale any, but...
The average time to crack it wouldn't require exhausting key space, the chances are very low that you would have to.
Also, the type of operations make a huge difference. Top500 is a specific set of 64-bit precision operations being measured. These systems may be much faster at the AES calculation.
That said, no matter how you fiddle the math, it comes out to a uselessly long time even assuming they had *one* protected piece of data they needed to function (given my laptop has about a dozen AES encrypted streams for utterly boring linux shells, the signal-to-noise of important encrypted data v. uninmportant is pretty against cracking too.
Technically impressive, absolute given, as the current top few offer unique technologies (Cray's Interconnect, IBM's processors).
From a profitability standpoint, undoubtedly the more successful a vendor is closer to the top of the list, the more they undoubtedly had to give up margin-wise for the bragging rights.
From a marketability standpoint, things get a tad awkward I think for the vendors at the top. IBM has Cell and BlueGene showcased, which we all understand can be used to great ends at the expense of a more complicated programming model (though HPC is already so parallel, they shouldn't be too bothered by this aspect of Cell), but in terms of more day-to-day operations, the sole bragging point relevant to the common market they get is that they know how to deliver and help architect 9,000 systems to work efficiently together (which is not a small feat, but the hardware itself has too high a barrier to get into the common market). For Cray, I honestly am surprised they have stuck around with only this niche market and still have held on. Their IO architecture is interesting, but of lower value at more common scales. These systems are certainly not cheap, but they are up against stiff competition willing to slice prices to grandstand about success.
The other configurations are made out of servers and processors and networking equipment that can more readily be used by a wider audience. So as cool as the exotic tech is, I think the commodity aspects are more compelling to the profitable customers for these vendors. Given the type of HP's share, HP may be quite ecstatic with the current list.
While the number itself is highly synthetic, it isn't necessarily out of date any more than any other single benchmark. HPCC attempts to score more stuff through a more varied test suite, but any one score of the suite is equally capable of being called irrelevant or uninteresting.
That said, there are a number of interesting items in the list. Cray's #2 showcased AMD's current generation's IO capabilities to the extent they've never been showcased before (right before they lose that exclusive benefit with Intel's Nehalem). IBM's non-x86 configurations show fantastic performance and even more drastic performance/watt (though I looked at the IBM website and they certainly price those QS22s up there, for a platform that is tricky to get the benefit out of).
But outside of exotic HT based interconnects and non-x86 servers, it is accurate to say mostly they are play-it-safe configurations that tie commodity equipment together with varying degrees of commodity networking equipment. It's a safe thing to do that people know how to do, and they can do it without making a unique custom solution the big companies wouldn't be able to productize on a small scale. I admit it is a shame, as I'd love to see more things along the lines of the Cray configuration, but none of the likely candidates seem to be willing to make that sort of investment.
I don't think IBM/Los Alamos suddenly plopped down something in response to cray at the last minute (frankly, I don't think you could move that fast).
Any hardware upgrades were almost certainly in plan, and if there were none, they've had 6 months of tuning to extract better numbers oout of what they had.
I wouldn't be surprised if in 6 months, the Cray without any additional hardware managed a better number than RoadRunner without additional hardware. However, such a victory is diminished somewhat by the energy the Cray undoubtedly consumes to acheive what performance they do get.
Viable is less controversial to claim ;)
Taking things at face value to show their official perception and how it doesn't match.
I wouldn't touch webapps with a 10 foot pole if I could help it, personally.
Since cars are kinda complicated, I'll break it down to an anti-car analogy so people can easily understand.
It's like sending DMCA notices for screenshots of your application to a site boasting about the total awesomeness of the application.
So first off, current incarnations of MS Office are considered the clear market leader. That's a fair observation. A traditionally installed local application manipulating files in a traditional way is popular.
OpenOffice.org is making inroads as a free alternative. More people are starting to find it a viable alternative for many circumstances, and opt not to explicitly buy MS Office. It behaves fundamentally the same way, and does basically the same stuff. Incidentally, I'm happy as it is a cross-platform application, but I think a greater portion of the userbase doesn't think about the source code or the cross-platform, they just didn't have to give money for it.
Then Google docs comes along. In terms of a strong brand to back the concept, it doesn't get much better than the word 'Google'. They find that despite the strong name and potential ability to fulfill at least the basic needs, people aren't excited about using it. The reason seems self-evident, people are more comfortable with traditional software models for this task. They feel they 'own' the software and have the most control over it. They may or may not back up to online storage, but they want to use a local application to edit it.
MS feels this means issuing their own webapp therefore would cement their lead. I think Google's failure indicates that such an offering is moot. People don't want subscription based software if non-subscription software can do the same thing or better. I've seen people throw out how it comes out cheaper in the long haul than buying the software every time, but it ignores the obvious, that people don't buy every iteration. I know people still using their copies of Office97 because they never had a reason to move. MS and many other companies hate this, but it is a simple fact.
It's absolutely clear IBM is contributing tons to Open Source. It's easy to note 'ibm.com' addresses in changelogs for tons of projects.
Generic Linux contributions and the success of Linux is key to them. You sell Linux to x86 users, you suddenly have a nice bridge to Linux on POWER. That seems to have been a pillar of their strategy.
While this is a good thought, and they also derive some recognition as general Linux experts, I'm suprised they haven't more directly capitalized on their investment. I would expect IBM to take ownership of a customer-facing distribution, and gain the ability to own the 'whole stack' in the x86 world. This has been their proud marketing message behind their power/mainframe systems, and yet they haven't applied much of that business or technical strategy to their x86 servers. By and large their x86 solutions aren't that much different from everybody else's. Linux could be a good puzzle piece to glue their middleware and hardware together.
Whether the result would be a good product or not, I cannot say, just surprised IBM hasn't done anything so intriguing in the x86 space.