You can buy an XP Home license/sticker (sans media) for about $60 mail-order. It's presumed that the big OEMs buying direct from Microsoft have enough negotiating power to drive their cost down to $45.
We have a lower violent crime rate than England-Wales and a lower violent crime rate than Australia, according to the International Crime Victims Survey, conducted by Leiden University in the Netherlands.
I mean, if you're going to compare the U.S. to other countries, your best choices would not just be other First World nations, but large ones that share similar legal systems, cultural traditions, and the like. Australia and Britain fit that well, and are more dangerous than the U.S. We must be doing something right . . .
Yes, those CD encyclopedias (Encarta, Grolier) suck, but it's because they're bad encyclopedias, not because they're on CD.
You can get the full 32-volume Encyclopedia Brittanica on CD, along with a whole bunch of additional material (an atlas, history timelines, Merriam-Webster dicionary and thesaurus, the obligatory multimedia bits, aa selection of articles by famous authors for previous editions of the encyclopedia, some other bits).
Where else could a high school student obtain up to date economic information?
The World Almanac and Book of Facts (or another almanac of that style). Printed on cheap, pulp paper; printed every year; inexpensive (glancing at the copies I have, the '04 edition was $11.95, the '91 was $6.95), available in any decent library and most wothwhile bookstores.
"Gas" is merely the contracted form of "gasoline", which is from
gas : Gas as in air and the like -ol : from oleum, "oil" (petroleum being rock-oil) -line : "of or relating to" or "made of, like"
Thus accurately describing the substance. Gasoline is the distilled fraction (a -line) of petroleum (-ol) that readily vaporizes (forms a gas).
Whereas, "petrol" is an Anglicized contracted form of the French essence de petrole, literally meaning "essence of petroleum". Essence, when used in distilling, meaning "a volatile (readily vaporizable) substance or constituent." Which is also an accurate description of the fuel for automobiles, being a readily vaporizable constituent of petroleum.
Thus both make perfect sense, since both are contractions of accurate descriptive terms. "Gas" is merely more prone to confusion with other meanings.
Nope, AMD doesn't have to pay Intel a dime, either.
Back when the first IBM PC was coming out, IBM made second-sourcing of the 8086/8088 family a requirement of the contract for Intel. So Intel sold a number of companies full rights to Intel's x86 IP. AMD then used genuine Intel microcode through to its 486s, a right confirmed in a court case over the 386 after Intel terminated the second-source agreement.
Fast-forward. A small chip company called NexGen has made a splash with its powerful not Pentium-socket compatible Nx586, the first chip to decode x86 instructions to RISC-like ones. It announces in October, 1995, that it's building multimedia extensions into the Nx686, the first attempt by a non-Intel company to extend the x86 instruction set. This is especially bold since NexGen is the smallest maker of x86 chips, trailing Intel, AMD, and Cyrix.
AMD then buys NexGen before the end of the year, getting the Nx686, and renaming it the K6. With the money and power of #2 x86 maker, the new multi-media enabled K6 looks poised to beat Intel's P55C (Pentium MMX) to market. But can even making it first to market ensure that software makers will adopt its instructions?
Intel and AMD, both unsure of their position and under pressure from software makers, cut a deal. They cross-license the technology and agree to a unified set of multimedia instructions, basically based on the P55C. This delays the K6 a bit as its MMX unit has to be re-engineered to use the new instructions, but guarantees AMD that they'll have software support.
The cross-license agreement is very broad, although it doesn't extend to actual silicon, and with supplemental agreements reached later over SSE, allows AMD and Intel to implement each other's extensions to x86. And so AMD can freely add SSE3 instructions like it's planning for next-generation Opterons/Athlon 64s, and Intel can freely add AMD64 instructions, or 3DNow if they wanted.
Sets it apart from what? It's the same basic architecture as used in Opteron systems, and the technology was co-developed by Apple and AMD -- after Apple jumped on AMD's HyperTransport bandwagon, since AMD is the one that started the development (based on the Alpha bus tech they licensed). The architectural commonalities between the G5 and the Opteron are in fact greater than those between an Opteron and a P4.
Apple has put an IBM chip on a joint AMD-Apple bus, which then connects to completely standard PC components. It's a nice computer, yes, but what makes Macs unique comes down to aesthetic design, not techology.
Oh, lemme correct that. NCSA Mosaic 3.0 allowed you to save the sessions and load them, history intact. So the minor hack would be to do automatic saving and allow loading . . .
There once were browsers that actually had this partly implemented -- Mosaic 3.0 and IBM's Web Explorer. They only saved the tree data for that browser session, but it was incredibly useful.
A quick hack to that would be saving the tree history to disk, and only load previous sesssion history trees into memory when you click on the specific previous session . . .
Unobtanium designates a class of science fiction/fantasy materials (e.g. adamant, adamantium, cavorite, dilithium, duranium, flubber, kryptonite, mithril, octiron, orichalcum, scrith, tiberium, tritanium) that fill roles in the story that known materials cannot.
Right now, there is no existing substance that could be used to build a space elevator, so such a material is (at the moment) a form of unobtanium. it will cease to be unobtanium as soon as it can be manufactured, much like the case of the skin of the SR-71 "Blackbird" went from unobtanium to titanium.
(Why not? Inflation, of course. Six million is just a drop in the bucket nowadays.;-) )
Re:This is all cool, but...
on
Mind Over Machine
·
· Score: 2, Informative
Is a neuron that complex?
Yes. There exists no complete, accurate simulation of even an ordinary human cell; neurobiochemistry is even more complicated even before you try to model interconnections.
Now, currently, interesting things are being done with simplified models of the neuron in moderately large arrays, and it may be that we won't need a complete model of neural chemistry to create the first truly intelligent neural-net models. But even then, 10^10 elements is well beyond the capacities of even our best supercomputers to model.
I don't see any particular reason why the approach won't work, but even with a simplified model working, a frog-level intelligence is a 20+ year problem.
Note LADPW had a special advantage not related to anything it did as an organization, however. Under Federal law, power generated by Federal power projects is available at reduced rates and increased priority to public-owned utilities. So, during the price crisis, LADPW was able to buy Federally generated power below market rates, and then sell any more than it needed to cover LA at a tidy profit, merely because of its ownership.
[Note I called the "power crisis" a price crisis. There was never a shortage of power generating capacity; the trouble was that, with the price spike in natural gas but a price cap on retail electricity, it was impossible for a significant portion of California's generator capacity to generate power for less than SCE and PGE could afford to buy it for wholesale. So rather than sell at a loss, the generator owners shut them down. If retail electricity prices had been allowed by law to float to meet changes in gas prices, SCE and PGE could have raised rates, paid more wholesale, and nobody would have shut down their generators. Thus we had another demonstration of the undisputable economic fact that price controls designed to prevent "gouging" on an "essential item" almost inevitably causes a shortage of the "essential item" instead.]
Wealth != money; if it did, hyperinflation would just make people insanely rich, instead of destroying economies.
It's actually pretty easy to destroy wealth. You just destroy things with value. Nuke a city, slam a plane into a skyscraper, pour out good wine into a sewer, apply a hammer to silicon chips, burn crops, launch art in a rocket to the Sun . . . wealth is quite easily destroyed.
So, assuming they're the 10 most-spoken Indian languages (counting Bengali but not Urdu), Microsoft is just now, twenty years after the first version of Excel, getting around to the second, fifth, fifteenth, sixteenth, eighteenth, twenty-third, twenty-ninth, thirtieth, thirty-second and thirty-sixth most-spoken native languages on Earth, spoken by 922 million people as their native tounge.
With that kind of response time, I'm sure the OSS translators are going to be caught up to in their efforts to support small African languages. In about five hundred years.
It's quite possible for a disproportionate number of Mac users to be Palm users; in fact, it's pretty likely that Mac users use Palms at a higher rate than Windows users, because Palms work better with Macs than the competitors to Palm do.
However, "a significant number of Palm users own Macs" doesn't follow. If a town of 150 people in Angola had 50 Ford F150 owners, sure, the residents would have a disproportionate number of F150s. Despite that, it's not true that those 50 Angolans would constitute a significant number of Ford F150 owners, given that Ford sells thousands of them a year.
Frankly, I doubt anyone, except maybe Palm, has stats on what percentage of its users have Macs. (Given the rigor of most marketing surveys, Palm likely thinks it knows, and is wrong.)
Second, what computer current Palm owners have is largely irrelevant, which is why I ignored it. Cobalt devices currently have zero marketshare in any market; Palm's decisions have to be made in the context of who they think they can sell the Cobalt devices to, not whom they've sold different devices to in the past.
It may be giving up its special advantage in the Mac marketplace will hurt them; on the other hand, they may have run estimates and think the development costs are more than they'll lose by having Mac owner marketshare move to the same proportions of Pocket PC/WinCE vs. Palm devices that the Windows world has.
Third, "rant"? That was just numbers thrown to show that "3%" is not confusing marketshare with installed base. The Mac marketshare was around 1.7% in 2003 (Mac unit sales from the Apple 10K filing for the year divided by Gartner's released-to-the-news estimate of PC sales). 3% is one estimate of installed base, as was my 4.3%.
You want a rant, you should see me on the subject of [muffled sounds of a man yelling through a gag] . . .
If we assume every Mac sold, 1999-2003, is part of the current Mac installed base (and given Mac sales over time, that's actually more generous than "average Mac is used five years") that's 17,206,000 machines.
If we assume that the total non-Mac installed base is merely equal to the number of PCs sold in 2003 (using the Gartner preliminary estimate), then there are roughly 170 million non-Mac machines in the world today.
So even if we assume average Mac users buy a computer every five years and the average non-Mac user buys one every year, the Mac percentage of the installed base is merely 10%. So the absolute worst-case scenario for Palm here is losing 10% of its market.
Now, if we change those numbers to something more reasonable, like an average of a Mac every four years and a PC every two, the Mac installed base drops to 13,758,000 computers, and the PC installed base goes up to 317,000,000. So Macs have 4.3% of the installed base.
Which should more-than-adequately account for UA spoofing in the 3% Google number; I doubt more than 30% of Mac users are claiming to be a Windows PC.
A non-profit organization was set up to run Internet name assignments, with international participation, representation of major infrastructure players, and even a nascent direct interested-person representation system.
It was called the Internet Corporation for Assigned Names and Numbers, and it's the organization that went ahead and so solidly entrenched VeriSign in the first place.
Merely passing along control to another NGO is not, in itself, a solution; there is no reason to expect it won't be politicized and turned into another ICANN.
Office is both generic and descriptive when referring to office-oriented software; you will notice Microsoft has yet to sue Sun over Star Office or Open Office.
Word is similarly both generic and descriptive for word-processing software, and Microsoft has not gone after AbiWord or Scientific Word.
SQL Server is another generic and descriptive term, since SQL and Server are both terms used in general computer science. And that's reflected in the use of the term by other groups: there's GNU SQL Server, Sybase SQL Server, Adabas SQL Server . ..
However, although they're generic words, the generic meanings of Access, Excel, and Outlook are unrelated to the software underlying them. Excel makes as good and as arbitrary a name for a database, an operating system, a food processor, or a brand of women's undergarments as it does for a spreadsheet.
That the term is generic and descriptive doesn't mean it has to be renamed; it means that it can't be protected like a trademark from use by other companies. If Microsoft wants to avoid confusion with Star Office, Scientific Word, and Adabas SQL Server, then, yes, they'll have to change the names; if they don't mind, they can keep selling Microsoft Office, Microsoft Word, and Microsoft SQL Server.
Slashdot Mirror
You can buy the XP Home OEM sticker on PriceWatch for $60 if you don't need media, which the big OEMs who just give you a rescue disk do not.
Presumably Dell and other major OEMs can get it for less than $60, given they're not going through a middleman and have clout as large buyers.
You can buy an XP Home license/sticker (sans media) for about $60 mail-order. It's presumed that the big OEMs buying direct from Microsoft have enough negotiating power to drive their cost down to $45.
Well, then:
We have a lower violent crime rate than England-Wales and a lower violent crime rate than Australia, according to the International Crime Victims Survey, conducted by Leiden University in the Netherlands.
I mean, if you're going to compare the U.S. to other countries, your best choices would not just be other First World nations, but large ones that share similar legal systems, cultural traditions, and the like. Australia and Britain fit that well, and are more dangerous than the U.S. We must be doing something right . . .
Yes, those CD encyclopedias (Encarta, Grolier) suck, but it's because they're bad encyclopedias, not because they're on CD.
You can get the full 32-volume Encyclopedia Brittanica on CD, along with a whole bunch of additional material (an atlas, history timelines, Merriam-Webster dicionary and thesaurus, the obligatory multimedia bits, aa selection of articles by famous authors for previous editions of the encyclopedia, some other bits).
For $49 at Amazon.
Before the $20 rebate.
Yep, but they did finally figure it out. Today you can buy the Brittanica CD-ROM for $49 on Amazon, and then send in the $20 rebate.
Where else could a high school student obtain up to date economic information?
The World Almanac and Book of Facts (or another almanac of that style). Printed on cheap, pulp paper; printed every year; inexpensive (glancing at the copies I have, the '04 edition was $11.95, the '91 was $6.95), available in any decent library and most wothwhile bookstores.
"Gas" is merely the contracted form of "gasoline", which is from
gas : Gas as in air and the like
-ol : from oleum, "oil" (petroleum being rock-oil)
-line : "of or relating to" or "made of, like"
Thus accurately describing the substance. Gasoline is the distilled fraction (a -line) of petroleum (-ol) that readily vaporizes (forms a gas).
Whereas, "petrol" is an Anglicized contracted form of the French essence de petrole, literally meaning "essence of petroleum". Essence, when used in distilling, meaning "a volatile (readily vaporizable) substance or constituent." Which is also an accurate description of the fuel for automobiles, being a readily vaporizable constituent of petroleum.
Thus both make perfect sense, since both are contractions of accurate descriptive terms. "Gas" is merely more prone to confusion with other meanings.
Nope, AMD doesn't have to pay Intel a dime, either.
Back when the first IBM PC was coming out, IBM made second-sourcing of the 8086/8088 family a requirement of the contract for Intel. So Intel sold a number of companies full rights to Intel's x86 IP. AMD then used genuine Intel microcode through to its 486s, a right confirmed in a court case over the 386 after Intel terminated the second-source agreement.
Fast-forward. A small chip company called NexGen has made a splash with its powerful not Pentium-socket compatible Nx586, the first chip to decode x86 instructions to RISC-like ones. It announces in October, 1995, that it's building multimedia extensions into the Nx686, the first attempt by a non-Intel company to extend the x86 instruction set. This is especially bold since NexGen is the smallest maker of x86 chips, trailing Intel, AMD, and Cyrix.
AMD then buys NexGen before the end of the year, getting the Nx686, and renaming it the K6. With the money and power of #2 x86 maker, the new multi-media enabled K6 looks poised to beat Intel's P55C (Pentium MMX) to market. But can even making it first to market ensure that software makers will adopt its instructions?
Intel and AMD, both unsure of their position and under pressure from software makers, cut a deal. They cross-license the technology and agree to a unified set of multimedia instructions, basically based on the P55C. This delays the K6 a bit as its MMX unit has to be re-engineered to use the new instructions, but guarantees AMD that they'll have software support.
The cross-license agreement is very broad, although it doesn't extend to actual silicon, and with supplemental agreements reached later over SSE, allows AMD and Intel to implement each other's extensions to x86. And so AMD can freely add SSE3 instructions like it's planning for next-generation Opterons/Athlon 64s, and Intel can freely add AMD64 instructions, or 3DNow if they wanted.
Sets it apart from what? It's the same basic architecture as used in Opteron systems, and the technology was co-developed by Apple and AMD -- after Apple jumped on AMD's HyperTransport bandwagon, since AMD is the one that started the development (based on the Alpha bus tech they licensed). The architectural commonalities between the G5 and the Opteron are in fact greater than those between an Opteron and a P4.
Apple has put an IBM chip on a joint AMD-Apple bus, which then connects to completely standard PC components. It's a nice computer, yes, but what makes Macs unique comes down to aesthetic design, not techology.
Oh, and you can get NCSA Mosaic 3.0 here, to see how it worked.
Oh, lemme correct that. NCSA Mosaic 3.0 allowed you to save the sessions and load them, history intact. So the minor hack would be to do automatic saving and allow loading . . .
There once were browsers that actually had this partly implemented -- Mosaic 3.0 and IBM's Web Explorer. They only saved the tree data for that browser session, but it was incredibly useful.
A quick hack to that would be saving the tree history to disk, and only load previous sesssion history trees into memory when you click on the specific previous session . . .
Well theoretically yes, but, the matches would be so perfect as to eliminate the thrill of romantic conquest.
Unobtanium designates a class of science fiction/fantasy materials (e.g. adamant, adamantium, cavorite, dilithium, duranium, flubber, kryptonite, mithril, octiron, orichalcum, scrith, tiberium, tritanium) that fill roles in the story that known materials cannot.
Right now, there is no existing substance that could be used to build a space elevator, so such a material is (at the moment) a form of unobtanium. it will cease to be unobtanium as soon as it can be manufactured, much like the case of the skin of the SR-71 "Blackbird" went from unobtanium to titanium.
No.
;-) )
(Why not? Inflation, of course. Six million is just a drop in the bucket nowadays.
Is a neuron that complex?
Yes. There exists no complete, accurate simulation of even an ordinary human cell; neurobiochemistry is even more complicated even before you try to model interconnections.
Now, currently, interesting things are being done with simplified models of the neuron in moderately large arrays, and it may be that we won't need a complete model of neural chemistry to create the first truly intelligent neural-net models. But even then, 10^10 elements is well beyond the capacities of even our best supercomputers to model.
I don't see any particular reason why the approach won't work, but even with a simplified model working, a frog-level intelligence is a 20+ year problem.
Please. This is /. The only wives we have are from here.
Note LADPW had a special advantage not related to anything it did as an organization, however. Under Federal law, power generated by Federal power projects is available at reduced rates and increased priority to public-owned utilities. So, during the price crisis, LADPW was able to buy Federally generated power below market rates, and then sell any more than it needed to cover LA at a tidy profit, merely because of its ownership.
[Note I called the "power crisis" a price crisis. There was never a shortage of power generating capacity; the trouble was that, with the price spike in natural gas but a price cap on retail electricity, it was impossible for a significant portion of California's generator capacity to generate power for less than SCE and PGE could afford to buy it for wholesale. So rather than sell at a loss, the generator owners shut them down. If retail electricity prices had been allowed by law to float to meet changes in gas prices, SCE and PGE could have raised rates, paid more wholesale, and nobody would have shut down their generators. Thus we had another demonstration of the undisputable economic fact that price controls designed to prevent "gouging" on an "essential item" almost inevitably causes a shortage of the "essential item" instead.]
Wealth != money; if it did, hyperinflation would just make people insanely rich, instead of destroying economies.
It's actually pretty easy to destroy wealth. You just destroy things with value. Nuke a city, slam a plane into a skyscraper, pour out good wine into a sewer, apply a hammer to silicon chips, burn crops, launch art in a rocket to the Sun . . . wealth is quite easily destroyed.
Hindi and "nine other Indian languages". Hmm.
So, assuming they're the 10 most-spoken Indian languages (counting Bengali but not Urdu), Microsoft is just now, twenty years after the first version of Excel, getting around to the second, fifth, fifteenth, sixteenth, eighteenth, twenty-third, twenty-ninth, thirtieth, thirty-second and thirty-sixth most-spoken native languages on Earth, spoken by 922 million people as their native tounge.
With that kind of response time, I'm sure the OSS translators are going to be caught up to in their efforts to support small African languages. In about five hundred years.
First, those two statements are not equivalent.
It's quite possible for a disproportionate number of Mac users to be Palm users; in fact, it's pretty likely that Mac users use Palms at a higher rate than Windows users, because Palms work better with Macs than the competitors to Palm do.
However, "a significant number of Palm users own Macs" doesn't follow. If a town of 150 people in Angola had 50 Ford F150 owners, sure, the residents would have a disproportionate number of F150s. Despite that, it's not true that those 50 Angolans would constitute a significant number of Ford F150 owners, given that Ford sells thousands of them a year.
Frankly, I doubt anyone, except maybe Palm, has stats on what percentage of its users have Macs. (Given the rigor of most marketing surveys, Palm likely thinks it knows, and is wrong.)
Second, what computer current Palm owners have is largely irrelevant, which is why I ignored it. Cobalt devices currently have zero marketshare in any market; Palm's decisions have to be made in the context of who they think they can sell the Cobalt devices to, not whom they've sold different devices to in the past.
It may be giving up its special advantage in the Mac marketplace will hurt them; on the other hand, they may have run estimates and think the development costs are more than they'll lose by having Mac owner marketshare move to the same proportions of Pocket PC/WinCE vs. Palm devices that the Windows world has.
Third, "rant"? That was just numbers thrown to show that "3%" is not confusing marketshare with installed base. The Mac marketshare was around 1.7% in 2003 (Mac unit sales from the Apple 10K filing for the year divided by Gartner's released-to-the-news estimate of PC sales). 3% is one estimate of installed base, as was my 4.3%.
You want a rant, you should see me on the subject of [muffled sounds of a man yelling through a gag] . . .
If we assume every Mac sold, 1999-2003, is part of the current Mac installed base (and given Mac sales over time, that's actually more generous than "average Mac is used five years") that's 17,206,000 machines.
If we assume that the total non-Mac installed base is merely equal to the number of PCs sold in 2003 (using the Gartner preliminary estimate), then there are roughly 170 million non-Mac machines in the world today.
So even if we assume average Mac users buy a computer every five years and the average non-Mac user buys one every year, the Mac percentage of the installed base is merely 10%. So the absolute worst-case scenario for Palm here is losing 10% of its market.
Now, if we change those numbers to something more reasonable, like an average of a Mac every four years and a PC every two, the Mac installed base drops to 13,758,000 computers, and the PC installed base goes up to 317,000,000. So Macs have 4.3% of the installed base.
Which should more-than-adequately account for UA spoofing in the 3% Google number; I doubt more than 30% of Mac users are claiming to be a Windows PC.
A non-profit organization was set up to run Internet name assignments, with international participation, representation of major infrastructure players, and even a nascent direct interested-person representation system.
It was called the Internet Corporation for Assigned Names and Numbers, and it's the organization that went ahead and so solidly entrenched VeriSign in the first place.
Merely passing along control to another NGO is not, in itself, a solution; there is no reason to expect it won't be politicized and turned into another ICANN.
Office is both generic and descriptive when referring to office-oriented software; you will notice Microsoft has yet to sue Sun over Star Office or Open Office.
.
Word is similarly both generic and descriptive for word-processing software, and Microsoft has not gone after AbiWord or Scientific Word.
SQL Server is another generic and descriptive term, since SQL and Server are both terms used in general computer science. And that's reflected in the use of the term by other groups: there's GNU SQL Server, Sybase SQL Server, Adabas SQL Server . .
However, although they're generic words, the generic meanings of Access, Excel, and Outlook are unrelated to the software underlying them. Excel makes as good and as arbitrary a name for a database, an operating system, a food processor, or a brand of women's undergarments as it does for a spreadsheet.
That the term is generic and descriptive doesn't mean it has to be renamed; it means that it can't be protected like a trademark from use by other companies. If Microsoft wants to avoid confusion with Star Office, Scientific Word, and Adabas SQL Server, then, yes, they'll have to change the names; if they don't mind, they can keep selling Microsoft Office, Microsoft Word, and Microsoft SQL Server.
So multitasking, like talking to your wife and thinking about your mistress
That's dangerous; human brains don't have memory protection.