What if the parent process forks a gazillion child processes which never write, and the the parent process writes to a memory location? Do you then have to allocate memory for all gazillion child processes at once? That could suck.
How many child processes never write to memory? What exactly is it doing if not writing to memory? Does Copy on Write only save you memory if you fork a bunch of child processes that do nothing? Or is it a matter of saving the allocation of memory until the parent process is done, thus speeding up the parent process at the time of the fork? Why not then copy on parent-write or child-execution, whichever comes first?
Dell has appeared to be far more expensive than eMachines/Gateway, Acer, and Compaq/HP for low-end desktops over the past year, even without shipping. These are no doubt low margin machines, but probably a high percentage of consumer retail sales. Dell's just not trying very hard at the low end.
It's too bad that they were forced to give up on getting over to McCool hill. If you look at the map referenced in this update, you realize that they just gave up on the farther safe slopes in favor of the slope immediately at hand. But if it survives to survey through another martian summer, I suppose it's worth it.
All slashdot stories must end with a dumb rhetorical question that triggers useless comments pointing out the stupidity of the rhetorical question. Q.E.D.
Uh, no. This is the Wall Street Journal, and they're talking business. "Guidance" refers to a company providing a range of revenue and profit for the coming quarters and years. Usually no more than two years. Stock analysts also come up with their own estimates, which are averaged to decide each quarter whether a company met or exceeded expectations. If Google provides guidance, it has an affect on controlling those expectations. Part of the wild rise in Google's stock price was that the analyst estimates were on the low side, until last quarter, when Google finally missed. Even though it is their job to read the tea leaves, the analysts are essentially asking Google to do so as well.
How does providing specific profits and revenue guidance have anything to do with being more than a company that gets its revenue and profit stream from adwords? If Google provides guidance, it's not revealing the secret playbook to become the next Microsoft, it's estimating how many ads it will sell next quarter and year. Investors will learn nothing really new from this, other than a better idea of whether the Forward P/E is 40 or 20 rather than 32. Wall Street analysts, on the other hand, will get a nice benchmark for their own estimates, and get a pat on the head if their estimates are more accurate because of the guidance from google.
Google's trying to become the next Microsoft by opening up their APIs to web developers for free, and more importantly, ASAP. They want to be the web platform the same way that MS became the desktop platform by providing a number of bundled applications and cheap SDKs. The larger the base of established sites using Google services, the better for them. Google's differentiator seems to be providing the content along with the services. Its competitors offer the content and services ala carte.
In five years will google APIs be as ingrained in your average website as windows is ingrained in your average desktop application? Google will continue to provide the framework for that to happen, ASAP. Think of it as Windows 3.0, with 3.1 coming soon. Unless wikipedia suddenly morphs into free maps and free storefronts and free classifieds and free file hosting and free email and free search, I don't see an open alternative to Google's free--as in beer-- content. You can mock the betas the same way people mocked early versions of Windows, but have no doubt that now as then, developers will use what is cheap and easy and available. It doesn't matter if it's coming out of Mountain View or Redmond as long as it pays the rent. Moral qualms are for Stallman.
The controller wasn't bad, actually. Indestructable and not oversensitive. Or maybe it was undersensitive and that's why I always sucked at arcade games. Too much TI-99 time. Alpiner, TI Invaders, great stuff.
27-Jan-06 - Close of All Fact Discovery Except As to Defenses to Claims Relating to Allegedly Misused Material
17-Mar-06 - Close of All Remaining Discovery (i.e., Fact Discovery As to Defenses to Any Claim Relating to Allegedly Misused Material)
This would be fact discovery relating to the material which IBM allegedly misused. They're looking for evidence that their use is no different from the companies that threw SCO a bone.
In the summer of 2001 I was a Microsoft intern and got to attend the annual company meeting at SafeCo field and listen to each product group enumerate which competitor they were going to crush. It was quite a list, you're right.
Your link doesn't actually explain what EWL is, but it's probably a reasonable assumption to assume that it won't be very compatible with 193nm (light wavelength) litho equipment.
IBM's annoucement has a lot to do with stretching the usefulness of existing litho equipment and materials down to nodes that it was never expected to reach. This has been done again (65nm) and again (45nm) from what was once expected. IBM is saying, add water, and we'll do it again (30nm).
You know, three years ago I semi-chided a friend for buying an iPod when I was sure that the price would drop way, way below the $300 (or was it $400) price range. I was sure that there would soon be iPod clones that did everything an iPod did, for less, and as soon as that happened, the price of the top-end models would come way down.
That hasn't happened. First of all, because the iPod experience has not been duplicated. I'm not an iPod owner, but I now take this for granted. You're sort of right about brand cachet, but regardless of cachet it's nicer riding in a luxury sedan than it is in an econowreck. Second of all, I don't know that the top-end iPods really are that overpriced. If the hardware specs can be substantially undercut price-wise, I don't see it happening. I'm talking like more than 25% less. Is a 50% premium too much to ask for the iPod experience? Probably not.
With its monthly fees, the Live service must also be a key aspect of justifying losing hundreds of dollars on every XBox sold. Given the fact that the PS3 will also have a few C-notes wrapped around every console, Sony absolutely needs this to make up that money.
Well, logically, when you figure out that stars don't orbit galactic centers as you expect you can theorize that gravity is not acting as you expect, or you can theorize that gravity is acting as expected, and that there is mass that you cannot detect through other means.
If you a nineteenth century astronomer and you noticed that Newtonian physics didn't accurately predict the orbit of Mercury, would you come up with the theory of relativity, or would you look for Vulcan? I agree with you that Dark Matter seems to be the 21st-century equivalent of searching for Vulcan, but trying to explain the observation without changing the theory of gravity was not necessarily a stupid thing to do, and it's a heck of a lot more straightforward than developing a new theory.
If they couldn't do it, they shouldn't have promised Apple the chips. Any "charade" was IBM's fault, not Apple's.
As I said, the overcommitment to clock speed and losing money was certainly IBM's problem, but the whole dog-and-pony show with the media is most certainly Apple's speciality. Especially in regards to clouding the issues regarding G5 workstation chips and G5 laptop chips. Apple characterizes their own unwillingness to fund development of a laptop G5 as IBM's inability to deliver. IBM was done supplying the cheap stuff, thanks to game processor volumes filling their fab.
If anything, this article could be characterized as IBM stating, "We have no problems creating low power processors." (Well hell, look at Blue Gene) And in that regard, yes, this may have something to do with Apple.
CNET didn't even mention this in regards to Apple. This processer is a big deal for 3 of IBM's four primary server brands, and it's a big deal because they're promising a clock frequency that's really fast. And they have IBM's chief egghead talking about two relatively unexotic technologies that are being implemented for the first time with 65nm chips.
Sigh. This is not a PowerPC chip that would've been sold to Apple. They'd have to take a relatively successful server chip, staple on AltiVec, figure out how to overcommit and lose money, and then get raked over the coals by Steve Jobs. I doubt they're looking to repeat that charade.
Not vaporware, just bad reporting. IBM's indicating that the chip will use SOI and strained silicon... if the reporter is too daft to realize that's unrevolutionary, whatever.
McNealy said Oracle CEO Larry Ellison recently gave Sun a big shot in the arm by lowering the license fee that the database company charges for Sun servers.
"I sat down with Larry and said, 'Larry, you're killing us,'" McNealy said. "Part of the problem is we didn't have the fastest microprocessors, so you had to throw a lot of microprocessors at it. When you charge $30,000 per core, we ended up looking very expensive."
I have a friend with a name similar to "Bob Smith." Turns out another Bob Smith was born on the very same day in a different state. That other Bob Smith eventually got his license suspended in PA. Well, that makes things tough on my friend Bob Smith, because the national database of shitty drivers has only birthdate and name as identifying information. So in order to get a license, my friend has had to negotiate with the DMV in two separate states swearing that he was not the Bob Smith of PA, and that he in fact living in completely separate state at the time. If ever one feels glad that the Social Security administration has blessed us all with truly unique identifying numbers, I guess this is it. And as the John Shuttleworths in England might learn one you all start paying for your national ID cards, credit agencies just love using your SSN.
Slashdot Mirror
How many child processes never write to memory? What exactly is it doing if not writing to memory? Does Copy on Write only save you memory if you fork a bunch of child processes that do nothing? Or is it a matter of saving the allocation of memory until the parent process is done, thus speeding up the parent process at the time of the fork? Why not then copy on parent-write or child-execution, whichever comes first?
Dell has appeared to be far more expensive than eMachines/Gateway, Acer, and Compaq/HP for low-end desktops over the past year, even without shipping. These are no doubt low margin machines, but probably a high percentage of consumer retail sales. Dell's just not trying very hard at the low end.
It's too bad that they were forced to give up on getting over to McCool hill. If you look at the map referenced in this update, you realize that they just gave up on the farther safe slopes in favor of the slope immediately at hand. But if it survives to survey through another martian summer, I suppose it's worth it.
All slashdot stories must end with a dumb rhetorical question that triggers useless comments pointing out the stupidity of the rhetorical question. Q.E.D.
Looks a little small for a tenor. Maybe a C-Melody.
Indeed, she appears to have a C-Melody rather than a Baritone. That, or she's a frickin huge girl and it's actually a Tenor as it appears.
Uh, no. This is the Wall Street Journal, and they're talking business. "Guidance" refers to a company providing a range of revenue and profit for the coming quarters and years. Usually no more than two years. Stock analysts also come up with their own estimates, which are averaged to decide each quarter whether a company met or exceeded expectations. If Google provides guidance, it has an affect on controlling those expectations. Part of the wild rise in Google's stock price was that the analyst estimates were on the low side, until last quarter, when Google finally missed. Even though it is their job to read the tea leaves, the analysts are essentially asking Google to do so as well.
How does providing specific profits and revenue guidance have anything to do with being more than a company that gets its revenue and profit stream from adwords? If Google provides guidance, it's not revealing the secret playbook to become the next Microsoft, it's estimating how many ads it will sell next quarter and year. Investors will learn nothing really new from this, other than a better idea of whether the Forward P/E is 40 or 20 rather than 32. Wall Street analysts, on the other hand, will get a nice benchmark for their own estimates, and get a pat on the head if their estimates are more accurate because of the guidance from google.
In five years will google APIs be as ingrained in your average website as windows is ingrained in your average desktop application? Google will continue to provide the framework for that to happen, ASAP. Think of it as Windows 3.0, with 3.1 coming soon. Unless wikipedia suddenly morphs into free maps and free storefronts and free classifieds and free file hosting and free email and free search, I don't see an open alternative to Google's free--as in beer-- content. You can mock the betas the same way people mocked early versions of Windows, but have no doubt that now as then, developers will use what is cheap and easy and available. It doesn't matter if it's coming out of Mountain View or Redmond as long as it pays the rent. Moral qualms are for Stallman.
If you're designing PCBs, $400 should be chump change-- right? I'm used to EDA packages that cost well over six figures per seat.
Or SHIT.
The controller wasn't bad, actually. Indestructable and not oversensitive. Or maybe it was undersensitive and that's why I always sucked at arcade games. Too much TI-99 time. Alpiner, TI Invaders, great stuff.
In the summer of 2001 I was a Microsoft intern and got to attend the annual company meeting at SafeCo field and listen to each product group enumerate which competitor they were going to crush. It was quite a list, you're right.
IBM's annoucement has a lot to do with stretching the usefulness of existing litho equipment and materials down to nodes that it was never expected to reach. This has been done again (65nm) and again (45nm) from what was once expected. IBM is saying, add water, and we'll do it again (30nm).
What exactly are you worried about? Are you distributing child porn and exhorting terrorism when playing XBox games?
That hasn't happened. First of all, because the iPod experience has not been duplicated. I'm not an iPod owner, but I now take this for granted. You're sort of right about brand cachet, but regardless of cachet it's nicer riding in a luxury sedan than it is in an econowreck. Second of all, I don't know that the top-end iPods really are that overpriced. If the hardware specs can be substantially undercut price-wise, I don't see it happening. I'm talking like more than 25% less. Is a 50% premium too much to ask for the iPod experience? Probably not.
With its monthly fees, the Live service must also be a key aspect of justifying losing hundreds of dollars on every XBox sold. Given the fact that the PS3 will also have a few C-notes wrapped around every console, Sony absolutely needs this to make up that money.
If you a nineteenth century astronomer and you noticed that Newtonian physics didn't accurately predict the orbit of Mercury, would you come up with the theory of relativity, or would you look for Vulcan? I agree with you that Dark Matter seems to be the 21st-century equivalent of searching for Vulcan, but trying to explain the observation without changing the theory of gravity was not necessarily a stupid thing to do, and it's a heck of a lot more straightforward than developing a new theory.
As I said, the overcommitment to clock speed and losing money was certainly IBM's problem, but the whole dog-and-pony show with the media is most certainly Apple's speciality. Especially in regards to clouding the issues regarding G5 workstation chips and G5 laptop chips. Apple characterizes their own unwillingness to fund development of a laptop G5 as IBM's inability to deliver. IBM was done supplying the cheap stuff, thanks to game processor volumes filling their fab.
If anything, this article could be characterized as IBM stating, "We have no problems creating low power processors." (Well hell, look at Blue Gene) And in that regard, yes, this may have something to do with Apple.
Sigh. This is not a PowerPC chip that would've been sold to Apple. They'd have to take a relatively successful server chip, staple on AltiVec, figure out how to overcommit and lose money, and then get raked over the coals by Steve Jobs. I doubt they're looking to repeat that charade.
Not vaporware, just bad reporting. IBM's indicating that the chip will use SOI and strained silicon... if the reporter is too daft to realize that's unrevolutionary, whatever.
I have a friend with a name similar to "Bob Smith." Turns out another Bob Smith was born on the very same day in a different state. That other Bob Smith eventually got his license suspended in PA. Well, that makes things tough on my friend Bob Smith, because the national database of shitty drivers has only birthdate and name as identifying information. So in order to get a license, my friend has had to negotiate with the DMV in two separate states swearing that he was not the Bob Smith of PA, and that he in fact living in completely separate state at the time. If ever one feels glad that the Social Security administration has blessed us all with truly unique identifying numbers, I guess this is it. And as the John Shuttleworths in England might learn one you all start paying for your national ID cards, credit agencies just love using your SSN.
Sounds like a non-virtual private network, or perhaps an intranet.