I never read anything definitive on the matter either.
Just once that was indirectly confirmed by news media when journalists dug up something on a politician. Government officials commented that intelligence, due to thickness of his biography, simply couldn't finish due background check on him in allotted time before appointment.
The impression I got was that they do background check, but the check isn't something what can actually bar a person from an office.
After election/appointment it is all up to fourth branch of government.
To me foreigner (not US resident) the whole birth certificate show down looked quite silly from the start.
I was surprised that same US citizens complaining on overreaching powers of their intelligence services, fail to understand that CIA starts background check on all involved people right after they declare their intent to run for the office. So that politicians who have something cloudy in their past can bail out from race long before they are actually nominated by parties. AFAIK this is standard for pretty much all branches of office: not only president candidates get screened, all of the officials who hold any kind of responsibility before taking post are checked by CIA. That's actually one of the reasons why such nominations take that long time to proceed.
There are lots of instructions and other craft inside 80x86 processors that occupy silicon that is never used.
Rarely used instructions are not need to be optimized - then they would take very little of transistors to implement. Only heavily used instructions needs to be optimized.
The programming languages that will benefit from Larrabee though will not be C/C++. It will be Fortran and the purely functional programming languages.
I hope you do understand that Fortran was fast because programs written in it were also simple. Modern programs combine lots and lots of math, memory and I/O operations. You can't easily parallelize that. Even now it can be already perfectly parallelized in C/C++, yet resulting software is quite complicated to manage and maintain.
It sounds stupid, but Sun actually already optimized their Java's JIT for SPART T1/T2 which are highly multithreaded CPUs.
Plus Itanium killed of most of the Risc architectures and x64 looks likely to kill off or nicheify Itanium.
This is misinformed B.S. Itanium didn't kill anything.
That was (and is) triumphant march of Linux/x64 all the time.
It is true that Intel and HP made out of PA-RISC and Alpha sacrificial lambs on Itanic's altar. Yet, Itanic's never caught up (and never will) to the levels where both PA-RISC and Alpha in the times were.
I bet a Larrabee like CPU would be great in a server too, and it's trivially highly scalable by changing the number of cores.
Servers are I/O heavy - CPU parallelism is very secondary. I doubt Larrabee would make any dent in server market. Unless of course OnLive/similar would catch up or Intel add something interesting for e.g. XML processing.
Tools would mean dick and it would take long time for developers to actually adopt such tools to exploit the architecture.
As an example, take recent Sun's highly-multi-threaded CPUs - T1 and T2. Benchmark team ran our software (essentially message processing) on the CPUs and it is dog slow - unless you disable the CPU multi-threading. On other hand Java, have seen good boost to performance: because Sun's JIT already can on-the-fly optimize code for such architecture.
It is a long road for Larrabee before general acceptance. Let's just that Intel wouldn't make same mistake twice.
If they do Windows programming, they should be used to working with random collections of spare parts that might have whatever random collection of system files that were put there by who knows what software written under the assumption they can spew whatever they want
wherever they want.
Probably that's why new PC games start at $60.
From what I have seen, the EVO folks target at $20.
P.S. And actually after M$ forced everybody into DirectX slavery, that's not the case anymore. It's only M$ developed/published games which also reinstall half of Windows.
I guess then EVO is for people who do not want to install kernels - but want to play games instead.
... And for developers who do not want to waste time writing scripts detecting and properly setting up software for all possible combinations of kernel+glibc+X+sdl+mesa.
Roguelike - most of them can be easily scripted around (and most people do that)
Not if you're playing over SSH.
SSH isn't a problem.
This brings me to another widely popular game genre without quickload: MMORPGs.
MMORPGs are by definition for kids: adults never have that much spare time to waste on leveling-up. And this are weeks of playing a game - all that to be able play the actual game in the end. Under adults I obviously mean people who have real life and under kids - pretty much everybody else.
Do not get me wrong: I really envy those who are older than 30 and can play WoW. It's just life dragged me into/pinned down to the real world where I can't afford such luxury (anymore).
So I take it you don't play roguelikes. Nor do you play Diablo 2 hardcore, Final Fantasy Tactics Advance, or any other game that has consequences for the player's actions. Nor do you play Animal Crossing games, or at least you see the mole a lot.
In Diablo 2 dying is rare occasion and generally reloading is not needed most of the time.
Roguelike - most of them can be easily scripted around (and most people do that) to preserve number of save file to be able to restart game from some earlier point. (BTW, Diablo 2 save system can be hacked in a similar way.)
FFTA - sucks for many other reasons, like e.g. it is heavily watered down, lots of unskippable crap, redundant unskippable dialogs, lacking story, etc. The whole bouquet of console retardism.
and 5s (at any moment of game!) needed to save progress and quit game.
Would you accept a game that has 5 second save (or backgrounded auto-save) but doesn't allow multiple loads of a given save file?
Actually in Doom3 the "5 seconds" are for saving and quitting: press F6 (quick save) and type "~" followed by "quit[Enter]". (Starting is even easier: double click Doom3 icon, wait few seconds for game to enter full screen mode and press immediately F9 (quick load): Doom3 would skip into screens and load game immediately.)
And no, I obviously wouldn't accept game which doesn't allow me to start game from the save two/more times. That's just stupid and senseless limitation.
Actually such games remind me of a reason my friend in times became early adopter of DVD: there was a bunch of movies the guy could watch (depending on mood) constantly and sometime was simply leaving the movie playing in background while doing something else. (With such usage, VHSs were dying withing less than year - DVDs were lasting much much longer).
My point here is that some Wii games are reaching now that level now too. It's not something to sit before TV NNN hours non stop til your eyes start bleeding, but actually something that you can pick up any time you are in mood for it and enjoy it for half hour or more.
For quite some time I was alone who had only very little time to play games and consequently my game library (mainly DS now, many PC games, some Wii ones too) was generally made of games which I can pick play for 15-30 minutes, save progress and move on to home chores/whatever. (e.g. try to find console game which can match 15s load time of Doom3 (that's from double click on desktop's icon to actual game play) and 5s (at any moment of game!) needed to save progress and quit game.)
It's still beyond me why game industry is so stupid and tries to appeal exclusively to the 12-14yo gamer audience (and those "hardcore gamers" who never grew out of their 12-14 basement times), completely disregard all other ages and other audiences.
I guess then You have never been IBM's customer...
Because only problem customers have with IBM is their prices. IBM can actually do anything you would ask them. Literally. But it will cost you a bunch... and your kidney too.
In short term I do not think there are any dangers in IBM's acquisition of Sun. Their either buy it (then we all have two/something years to make a decision where to migrate) or not (then we all have two/something years before Sun goes bankrupt, so better start planning migration now).
Sun is one of the last big product oriented companies. And it sucks to be a product company now: sales went down a lot, while services slightly improved.
Yep, right handed swastika - right in the Sun's logo. Which in some mythologies is actually associated with... right - the Sun. (The star under which like where are living - not the corporation.)
As somebody born long after WWII I hardly have anything negative against swastika.
Still. Sun's logo was so many year popping up randomly before my eyes and I never spotted the pun: it is quite literally 100% sun wheel or good or right-handed swastika.
But it's really hard to argue now that their F/LOSS portfolio is a burden and was a waste of money.
Whoever planned those acquisitions in past was visionary.
But whoever would have to capitalize on that now... good luck. He'd better be at least a double visionary of the previous guy.
I'd say Sun, a not-so-much service company compared to HP or IBM, is in quite bad position. Their corporate culture is also not that nice to customers: they are used too much to selling boxes. All that "service" thing is way to new for them. And cloud is primarily a service, not product. They have been changing in past few years (my employer is actually "strategical partner" of Sun) but I'm not sure they have much time left to convince everybody to try to deal with them again. (That's probably why the aforementioned focus on start-ups - what is also bad timing, since in the financial state, there would be much less of those.)
Quality is almost always CHEAPER, not more expensive. Fix your process and produce quality, or you'll be bleeding $$$ and STILL have a broken process that can't produce quality.
Forgot to reply to that.
Quality is most expensive for one simple reason: there is no practical limit to QA.
That's because in modern software and hardware, internal state is described with hundreds and thousands items of configuration. Total number of possible states a PC system can be in long time ago exceeded number of stars in our galaxy.
That means that though there is a theoretical limit to number of of configurations and states, the limit is so frigging high that you have no chance to test a tiny speck of it in any time frame for your product to remain relevant to consumers.
Only choice is to limit number of configurations and then define focus areas where tests would be conducted. E.g. another industry average I know is 95%. Or in other words: 5% of returns are tolerable. I do not know how it is with game consoles, but some similar metric I'm sure is in Xbox360 business plan of MS.
P.S. Airplains cost what they cost mainly due to QA and certifications they have to have before they can be sold or take off - not because they just expensive.
I hope you never design an airplane... or traffic signal system... or a roof for a house... or an elevator... or anything more complicated than a sand castle.
To answer your unasked questions: yes, most of what you enumerate has dedicated safety standards. Except for sand castle.
Yes, even roofs. Because in past some first industrial roofs where made from materials which were actually dangerous to health in long term. That's why there are standards for that now.
And hey, I have worked in safety market - it's between traffic signal system and avionics. Close to elevator's level of mechanical safety - but in electronics/software. Well, in safety market the "fast, cheap, secure" works the same: with security (realiability) commanding features and lion share of costs going to testing. But that's niche market too;)
But game consoles they are very different from e.g. planes or medical equipment or fire alarm (fault tolerant system, life support system, safety system correspondently) in way that if you get RRoD or E74 nobody's life is endangered. That's why btw you can acquire Xbox360 for as low as $200 (compared to e.g. $800 for cheapest safety box (based on Motorola 68K CPU with 64K RAM) plus $3k safety software to actually be able to use the box plus $Nk for certification from state that whatever you coded in the box actually conforms to corresponding ISO safety standard.)
So what you're saying is that Microsoft sacrificed quality for price and features.
Yep.
Sony appears to have sacrificed price for quality and features. I'll take Sony.
What identifies you as niche buyer. And mass market companies do not care about niche buyers. (Even Sony.)
People see price first - but experience quality only later. That's why it is important to balance the both. If entry price is too high, provided there is competition, many wouldn't bother to even try. But once people bought a console, natural instinct of buyer to protect investments would actually smooth the negative perception of most quality problems.
In other words, in mass market, low quality is forgivable, high price isn't.
P.S. Just recall how Dell improved quality of their PCs. It's not that it improved anything, but they have people standing by with spare replacement parts. Quality is the same usual crap. But if it breaks, they simply replace it - real fast. And perception of Dell's quality really soared in past years.
Its just shitty hardware Q&A from the same company that excels in bad quality software.
You are - as most people on/. - are totally wrong about it.
Quality is a metric. Product has to balance quality vs. features vs. price.
N.B. Network people I'm sure can already recognize the famous "fast, cheap, secure: pick any two" pattern.
M$ as business has to balance all the three to make the product not too expensive (== cut features and quality); stable enough (== fewer better tested featured) and feature-full.
While feature set and price often are predefined (e.g. in case of Xbox360) quality remains only metric business can use to regulate development/etc costs.
In the end: Yes, M$ has shitty hardware Q&A. But it's "as designed". Otherwise, Xbox360 might have cost only few chips less than PS3. It's cheaper (RRoD fiasco was really exceptional) to deal with quality issues later, than introducing delays into initial phases of development/promotion of new product.
Slashdot Mirror
I never read anything definitive on the matter either.
Just once that was indirectly confirmed by news media when journalists dug up something on a politician. Government officials commented that intelligence, due to thickness of his biography, simply couldn't finish due background check on him in allotted time before appointment.
The impression I got was that they do background check, but the check isn't something what can actually bar a person from an office.
After election/appointment it is all up to fourth branch of government.
To me foreigner (not US resident) the whole birth certificate show down looked quite silly from the start.
I was surprised that same US citizens complaining on overreaching powers of their intelligence services, fail to understand that CIA starts background check on all involved people right after they declare their intent to run for the office. So that politicians who have something cloudy in their past can bail out from race long before they are actually nominated by parties. AFAIK this is standard for pretty much all branches of office: not only president candidates get screened, all of the officials who hold any kind of responsibility before taking post are checked by CIA. That's actually one of the reasons why such nominations take that long time to proceed.
No.
As Colbert (as in ISS module name) pointed out, the hawaiian birth certificate is very likely fake as it is not etched on a coconut.
And knowing current state of US intelligence services, one never can be sure that somebody actually run a background check before he got elected.
With same success, I can run "for(;;;);" in several threads and run all CPUs/cores to ground.
No matter what Java folks try to make out of it, Java on servers is pretty niche - precisely because of inefficient use of resources.
Server Java is of course not so niche in whole Java market. But not other way around.
There are lots of instructions and other craft inside 80x86 processors that occupy silicon that is never used.
Rarely used instructions are not need to be optimized - then they would take very little of transistors to implement. Only heavily used instructions needs to be optimized.
The programming languages that will benefit from Larrabee though will not be C/C++. It will be Fortran and the purely functional programming languages.
I hope you do understand that Fortran was fast because programs written in it were also simple. Modern programs combine lots and lots of math, memory and I/O operations. You can't easily parallelize that. Even now it can be already perfectly parallelized in C/C++, yet resulting software is quite complicated to manage and maintain.
It sounds stupid, but Sun actually already optimized their Java's JIT for SPART T1/T2 which are highly multithreaded CPUs.
Plus Itanium killed of most of the Risc architectures and x64 looks likely to kill off or nicheify Itanium.
This is misinformed B.S. Itanium didn't kill anything.
That was (and is) triumphant march of Linux/x64 all the time.
It is true that Intel and HP made out of PA-RISC and Alpha sacrificial lambs on Itanic's altar. Yet, Itanic's never caught up (and never will) to the levels where both PA-RISC and Alpha in the times were.
I bet a Larrabee like CPU would be great in a server too, and it's trivially highly scalable by changing the number of cores.
Servers are I/O heavy - CPU parallelism is very secondary. I doubt Larrabee would make any dent in server market. Unless of course OnLive/similar would catch up or Intel add something interesting for e.g. XML processing.
Tools would mean dick and it would take long time for developers to actually adopt such tools to exploit the architecture.
As an example, take recent Sun's highly-multi-threaded CPUs - T1 and T2. Benchmark team ran our software (essentially message processing) on the CPUs and it is dog slow - unless you disable the CPU multi-threading. On other hand Java, have seen good boost to performance: because Sun's JIT already can on-the-fly optimize code for such architecture.
It is a long road for Larrabee before general acceptance. Let's just that Intel wouldn't make same mistake twice.
If they do Windows programming, they should be used to working with random collections of spare parts that might have whatever random collection of system files that were put there by who knows what software written under the assumption they can spew whatever they want wherever they want.
Probably that's why new PC games start at $60.
From what I have seen, the EVO folks target at $20.
P.S. And actually after M$ forced everybody into DirectX slavery, that's not the case anymore. It's only M$ developed/published games which also reinstall half of Windows.
I guess then EVO is for people who do not want to install kernels - but want to play games instead.
Roguelike - most of them can be easily scripted around (and most people do that)
Not if you're playing over SSH.
SSH isn't a problem.
This brings me to another widely popular game genre without quickload: MMORPGs.
MMORPGs are by definition for kids: adults never have that much spare time to waste on leveling-up. And this are weeks of playing a game - all that to be able play the actual game in the end. Under adults I obviously mean people who have real life and under kids - pretty much everybody else.
Do not get me wrong: I really envy those who are older than 30 and can play WoW. It's just life dragged me into/pinned down to the real world where I can't afford such luxury (anymore).
So I take it you don't play roguelikes. Nor do you play Diablo 2 hardcore, Final Fantasy Tactics Advance, or any other game that has consequences for the player's actions. Nor do you play Animal Crossing games, or at least you see the mole a lot.
In Diablo 2 dying is rare occasion and generally reloading is not needed most of the time.
Roguelike - most of them can be easily scripted around (and most people do that) to preserve number of save file to be able to restart game from some earlier point. (BTW, Diablo 2 save system can be hacked in a similar way.)
FFTA - sucks for many other reasons, like e.g. it is heavily watered down, lots of unskippable crap, redundant unskippable dialogs, lacking story, etc. The whole bouquet of console retardism.
Animal Crossing - not my cup of tea.
and 5s (at any moment of game!) needed to save progress and quit game.
Would you accept a game that has 5 second save (or backgrounded auto-save) but doesn't allow multiple loads of a given save file?
Actually in Doom3 the "5 seconds" are for saving and quitting: press F6 (quick save) and type "~" followed by "quit[Enter]". (Starting is even easier: double click Doom3 icon, wait few seconds for game to enter full screen mode and press immediately F9 (quick load): Doom3 would skip into screens and load game immediately.)
And no, I obviously wouldn't accept game which doesn't allow me to start game from the save two/more times. That's just stupid and senseless limitation.
The link was posted above and is quite insightful read.
Actually such games remind me of a reason my friend in times became early adopter of DVD: there was a bunch of movies the guy could watch (depending on mood) constantly and sometime was simply leaving the movie playing in background while doing something else. (With such usage, VHSs were dying withing less than year - DVDs were lasting much much longer).
My point here is that some Wii games are reaching now that level now too. It's not something to sit before TV NNN hours non stop til your eyes start bleeding, but actually something that you can pick up any time you are in mood for it and enjoy it for half hour or more.
One of Midway's teams actually figured out how to sell games on the Wii.
Thanks for the link.
For quite some time I was alone who had only very little time to play games and consequently my game library (mainly DS now, many PC games, some Wii ones too) was generally made of games which I can pick play for 15-30 minutes, save progress and move on to home chores/whatever. (e.g. try to find console game which can match 15s load time of Doom3 (that's from double click on desktop's icon to actual game play) and 5s (at any moment of game!) needed to save progress and quit game.)
It's still beyond me why game industry is so stupid and tries to appeal exclusively to the 12-14yo gamer audience (and those "hardcore gamers" who never grew out of their 12-14 basement times), completely disregard all other ages and other audiences.
Do not litter into deprecated interfaces like select().
There is whole Async I/O API especially for that purpose.... apropos for aio.
As for NTFS: At least from the application side you know which problems will hit you and which ones not.
Because in Linux, problems - given enough community exposure - are getting eventually fixed.
Hard-coding broken behavior and workarounds into applications on Linux - unlike on Windows and commercial *NIXs - doesn't work.
Power to the people.
Jedem das Seine?
I guess then You have never been IBM's customer...
Because only problem customers have with IBM is their prices. IBM can actually do anything you would ask them. Literally. But it will cost you a bunch ... and your kidney too.
In short term I do not think there are any dangers in IBM's acquisition of Sun. Their either buy it (then we all have two/something years to make a decision where to migrate) or not (then we all have two/something years before Sun goes bankrupt, so better start planning migration now).
Sun is one of the last big product oriented companies. And it sucks to be a product company now: sales went down a lot, while services slightly improved.
Yep, right handed swastika - right in the Sun's logo. Which in some mythologies is actually associated with ... right - the Sun. (The star under which like where are living - not the corporation.)
For comparison: Swastika vs. Sun's logo
As somebody born long after WWII I hardly have anything negative against swastika.
Still. Sun's logo was so many year popping up randomly before my eyes and I never spotted the pun: it is quite literally 100% sun wheel or good or right-handed swastika.
But it's really hard to argue now that their F/LOSS portfolio is a burden and was a waste of money.
Whoever planned those acquisitions in past was visionary.
But whoever would have to capitalize on that now ... good luck. He'd better be at least a double visionary of the previous guy.
I'd say Sun, a not-so-much service company compared to HP or IBM, is in quite bad position. Their corporate culture is also not that nice to customers: they are used too much to selling boxes. All that "service" thing is way to new for them. And cloud is primarily a service, not product. They have been changing in past few years (my employer is actually "strategical partner" of Sun) but I'm not sure they have much time left to convince everybody to try to deal with them again. (That's probably why the aforementioned focus on start-ups - what is also bad timing, since in the financial state, there would be much less of those.)
Quality is almost always CHEAPER, not more expensive. Fix your process and produce quality, or you'll be bleeding $$$ and STILL have a broken process that can't produce quality.
Forgot to reply to that.
Quality is most expensive for one simple reason: there is no practical limit to QA.
That's because in modern software and hardware, internal state is described with hundreds and thousands items of configuration. Total number of possible states a PC system can be in long time ago exceeded number of stars in our galaxy.
That means that though there is a theoretical limit to number of of configurations and states, the limit is so frigging high that you have no chance to test a tiny speck of it in any time frame for your product to remain relevant to consumers.
Only choice is to limit number of configurations and then define focus areas where tests would be conducted. E.g. another industry average I know is 95%. Or in other words: 5% of returns are tolerable. I do not know how it is with game consoles, but some similar metric I'm sure is in Xbox360 business plan of MS.
P.S. Airplains cost what they cost mainly due to QA and certifications they have to have before they can be sold or take off - not because they just expensive.
I hope you never design an airplane ... or traffic signal system ... or a roof for a house ... or an elevator ... or anything more complicated than a sand castle.
To answer your unasked questions: yes, most of what you enumerate has dedicated safety standards. Except for sand castle.
Yes, even roofs. Because in past some first industrial roofs where made from materials which were actually dangerous to health in long term. That's why there are standards for that now.
And hey, I have worked in safety market - it's between traffic signal system and avionics. Close to elevator's level of mechanical safety - but in electronics/software. Well, in safety market the "fast, cheap, secure" works the same: with security (realiability) commanding features and lion share of costs going to testing. But that's niche market too ;)
But game consoles they are very different from e.g. planes or medical equipment or fire alarm (fault tolerant system, life support system, safety system correspondently) in way that if you get RRoD or E74 nobody's life is endangered. That's why btw you can acquire Xbox360 for as low as $200 (compared to e.g. $800 for cheapest safety box (based on Motorola 68K CPU with 64K RAM) plus $3k safety software to actually be able to use the box plus $Nk for certification from state that whatever you coded in the box actually conforms to corresponding ISO safety standard.)
So what you're saying is that Microsoft sacrificed quality for price and features.
Yep.
Sony appears to have sacrificed price for quality and features. I'll take Sony.
What identifies you as niche buyer. And mass market companies do not care about niche buyers. (Even Sony.)
People see price first - but experience quality only later. That's why it is important to balance the both. If entry price is too high, provided there is competition, many wouldn't bother to even try. But once people bought a console, natural instinct of buyer to protect investments would actually smooth the negative perception of most quality problems.
In other words, in mass market, low quality is forgivable, high price isn't.
P.S. Just recall how Dell improved quality of their PCs. It's not that it improved anything, but they have people standing by with spare replacement parts. Quality is the same usual crap. But if it breaks, they simply replace it - real fast. And perception of Dell's quality really soared in past years.
Finally, several months after the loss in elections, Sarah Palin let the steam out.......
Its just shitty hardware Q&A from the same company that excels in bad quality software.
You are - as most people on /. - are totally wrong about it.
Quality is a metric. Product has to balance quality vs. features vs. price.
N.B. Network people I'm sure can already recognize the famous "fast, cheap, secure: pick any two" pattern.
M$ as business has to balance all the three to make the product not too expensive (== cut features and quality); stable enough (== fewer better tested featured) and feature-full.
While feature set and price often are predefined (e.g. in case of Xbox360) quality remains only metric business can use to regulate development/etc costs.
In the end: Yes, M$ has shitty hardware Q&A. But it's "as designed". Otherwise, Xbox360 might have cost only few chips less than PS3. It's cheaper (RRoD fiasco was really exceptional) to deal with quality issues later, than introducing delays into initial phases of development/promotion of new product.