Apparently one of the biggest advantages of the amd64 architecture (aka x86-64) is not the 64-bitness at all (though there are some everyday benefits to that too -- e.g., visiting 1GB files in Emacs:-), but that in 64-bit mode it has more registers (not just wider ones), which allows the compiler to generate better code than it can for the anemic normal x86 architecture.
You'd think that the main developer would welcome i18n patches.
Indeed, but it's not uncommon for such patches to be very intrusive, badly designed, and quite buggy (even when not using the support). For people who really need it, maybe the functionality is worth the problems, but often the tradeoffs are different for the world at large (at least from the maintainer's point of view).
The main FOSS project I'm involved with is GNU Emacs, and the people who post to the development mailing list seem to be rather evenly distributed over the entire globe (it's funny watching replies to a message slowly coming in as people wake up in successive time zones), with the exception of Africa (er, and Antarctica I suppose). What variation there is seems to match what one might expect from variations in population and economic development pretty well; of course as it's an english language list, there's also a bias towards english-speaking countries.
I expect that rather than a correlation with "piracy", you're simply seeing the effect of economic development and other obvious factors (people who don't have computers or reliable net connections, are less likely to participate:-).
Another effect I've noticed is that in countries like Japan there are lots of "local" FOSS projects (often forks of more widespread packages) which never really seem to show their face outside that country. I guess this is due to both language and cultural issues (and some technical ones -- massive changes to support your local language may be very important to you, but the original developer may not accept them). I imagine that in countries with less widespread connectivity, this effect may be even greater.
There's a reason to go to silicon valley. The area is beautiful, and the talent pool for your $COMPANY there is tremendous, if you need 20 engineers to work on some software project, finding 20 skilled individuals in Atlanta, Georgia is going to suck. Trying to find 20 skilled people in silicon valley is a matter of going out to a busy resturant at lunch
I'm sure you're right about lots of talented people living nearby, but beautiful?!? I admit, I've only visited, but as far as I can see, silicon valley is the epitome of horrid suburban sprawl office-park mcmansion wasteland. It's one of the most depressing places I've ever been.
There are a lot of companies there that would seem very cool places to work, and a lot of the industry "focus" seems to be on SV... but I could never live there. I'd rather work for a lesser company in a civilized location.
Well it's not just manga vs. Mishima, there's a huge number of popular modern Japanese (non-manga) writers too. There's (obviously) a vast quantity of stuff which hasn't been translated into English.
AFAIK, manga's taken a big bite of out of non-manga reading, but that seems to have been going on for a long time. It's just an offhand judgement, but in general I think Japan's (non-manga) book scene seems a lot healthier than that in the U.S -- though I guess that says more about the U.S. than Japan...
Mind-share is created by peoples' expectations, and peoples' expectations are decided based on their experience. For every person who knows the ins-and-outs of the game industry, and has some idea of what to expect with the Wii, there are lots of regular gamers who simply remember they bought a PS2 rather than a Gamecube, and will go in preferring the PS3 for that reason.
By that same logic, the xbox -- and the PS1! -- should have failed miserably. There's certainly an element of "just blindly keep buying the same brand", but there are also clearly turning points from time to time where something new comes along and captures the public's imagination.
Now, it's very hard to predict these turning points, but the Wii's extremely strong showing at E3 (both among players and developers) makes people feel that maybe such shift is coming. Maybe Nintendo's innovations will move the market, or maybe you're right that the public will just opt for the same-old same-old -- but it's hardly a sure thing either way [and of course the huge price differential hardly hurts Nintendo's cause!].
What's with this blind faith in Nintendo? Their last two consoles have been decidedly mediocre. Decent hardware, patchy library. Why's is everyone convinced this run-through is going to be different?
There's nothing particularly wrong with the n64/gc libraries -- nowhere near the quantity of ps2 games, but there are some absolutely incredible games, and certainly enough to satisfy. Nintendo's hardware and 1st-party games are of course top-notch.
I think the feeling is that with every new generation, the counter gets reset, and the only thing required for good 3rd-party support is mind-share.
In the last gen, Sony played their cards perfectly and managed to hold onto the the mind-share they had with the ps1 -- but many people seem to feel they've stumbled badly this time, and that Nintendo stands a good chance to get 3rd party support they didn't in the last gen. Given that Nintendo's pretty much guaranteed to offer high quality hardware and 1st-party games, has a vastly lower price point, and is simply doing more interesting things than Sony is, why would you not want to invest in a wii???
M$ only invested about $20 in advertising the system in Japan. Many gamers over there aren't even aware the system exists. Gamers are going to the store to buy an Xbox1 not even knowing there is a 360 out.
No, this isn't true. I live in Japan and I've seen tons of xbox360 advertising. There was an initial advertising blitz during the release, and even now, I see lots of prime-time TV ads for the 360 (they seem to sponsor lots of TV shows). in fact, I've seen far more TV advertising (here in Japan) for the 360 than for any other gaming system.
They also have a fairly strong presence in many Japanese video game stores -- prominent demo kiosks, game section not relegated to back corner despite the low sales, etc. Given the absurdly low income that must come from 360 sales, I assume this means MS is forking over lots of promotional cash to stores.
Yup. In fact with my browser and fonts, it's completely unreadable (half the text overflows the white box and ends up displayed as black on dark blue...).
I suppose it probably displays properly in Internet Explorer if you happen to have exactly the same settings as the page designer ("but, but, it worked for me!!").
OTOH, the page does have a convenient "text" button to bypass the page designer's idiocies. [A lot of sites don't seem to have even that.]
Many people across the country seem to read the New York Times, but I'm a little iffy on them. It seems that their reporters have been caught lying, and doing other unscrupulous things a number of times.
I suspect there are very few major news sources which haven't had problems like that (individual reporters lying in their stories). The main difference is that the NYT is constantly under scrutiny (being essentially the "paper of record" in the USA), especially now that many neo-conservatives seem to feel some kind of personal animosity towards it.
I think the most important issue is whether such activities result from impropriety on the part of the paper itself; as far as I know, this wasn't the case with any of the recent NYT gaffes.
My initial thought was "Oh great, another Japanese-style bloat-o-phone" -- but after looking at the specs, I see the N93 is actually even more bloated and heavy than that (a typical Japanese bloat-phone is around 115g, and already seems kind of clunky and awkard; the N93 is 180g!).
I understand some people like all-in-phone phones because they only have to carry around one device... but it seems rather silly when it reaches the point where you don't want to carry it around at all because your pockets aren't big or strong enough. I rather like the ability to split up my gadgets as is convenient for the situation -- carry one thing in my pocket, another in my bag, leave another at home because I know I won't need it.
[Of course some day all-in-one devices will probably become so small and functional that there will be little reason not to use them... but that N93 seems very far from that ideal.]
Efficient mass transit, unfortunately, requires that we all work in a dense downtown area where a critical mass of people shows up.
No, it only requires that there be some concentration of people. It doesn't have to all be in one place; a network of connected nodes (a bunch of "little downtowns") works just as well, in fact, better -- if you can get people moving in both directions equally, you get double the capacity for free!
Mass transit is also unpleasant to use and generally very slow.
No, badly done mass transit is like that. It's very likely that Houston has bad mass transit of course.
More and bigger roads, logically enough, would be the better solution.
It has been shown time and time again, that this is not the right solution -- building bigger roads just gives people the impression that "it's ok" to drive wherever they want, as much as they want, so people do, and things reach exactly the same state of congestion they were at before -- only worse, because now you've got a bigger load on the rest of the infrastructure, and the landscape is starting to seem like an endless vista of pavement and parking lots. [I don't know about you, but I think giant 12-lane local roads and massive parking lots every five feet are ugly and depressing.]
The real problem is U.S. car worship. No matter what you do, tradeoffs have to be made, and as long as people remain so emotionally invested in their cars, it's very unlikely they'll be able to make rational choices about the matter.
It's interesting, because I've been in this field for 16 years, and I've never met a pure "software architect" or a pure "programmer"
Indeed; as far as I can figure these are weird-ass terms left over from the 1960s.
Certainly there are variations in the amount of coding or design responsibility that different people on a team have, but it's usually very much a matter of degree, and tends to shift around as a project develops.
I think terms like "software architect" are as much vanity titles as anything else these days -- if somebody's very good, he gets a big paycheck and his business card says "software architect," but the less clueful have to make do with "code monkey."
How about at your workplace? (Maybe you are too young now for a job, but in 5 years time?)
Certainly not a given... my workplace is almost entirely windows based, but the first thing I do whenever they give me a new PC is wipe the disk and install my OS of choice (usually debian). [Just for the bragging rights, I make a point of not even booting windows once.]
Many coworkers are a bit bemused by this and occasionally annoyed if I can't access some random IE-only internal website, but well, I can deal.
[It's a very big company, so I think they do actually have IT nazis of some sort, but the rules tend to get bent an awful lot by the time they reach the lower levels. We've got our own sysadmin who is a lot more practical -- if you're getting your work done and not bugging him, he's on your side.]
Not to mention color depth: 24 bit color is clearly inadequate for many uses, and formats like OpenEXR (one of the current best HDR image formats; it uses 48 bits per pixel, in the form of 3 16-bit "half floats") are gaining popularity. For some uses RGB isn't sufficient either, and you want samples at many more wavelengths.
There is always some overhead related to MP support, and there will always be benchmarks which can show its worst effect, but OpenBSD is quite well known for being far worse in this respect than other mainstream FOSS kernels. Linux in particular has spent vast amounts of effort optimizing the MP case, and there are many core linux hackers who have very strong MP experience from IBM, SGI, etc.
However, the environmental problem of "clear cutting forests" isn't as big as is typically let on by envirowackos. We have more forest in the US now than we've had since the 1930s, and that amount is still increasing.
Note that "forest" is far too vague a word to be very meaningful -- untouched old-growth forest is a very different thing than, for instance a mono-culture of non-native trees planted at some absurdly high density. Which you prefer of course, depends on your values and goals (the most horrid examples I've seen were tree plantations in Scotland whose apparent purpose was to abuse the tax laws!).
I disagree and believe that Tivo did innovate and does deserve patent protection.
Tivo may well have thought innovatively, and be a great company -- but remember, you can't patent great ideas, and that's what exactly most of what you describe sounds like.
If they had a sufficiently non-obvious mechanism to implement some of their great ideas, they could patent the mechanism, but anybody else is quite free to come along and use a different method with the same result. Even if Tivo thought of it first.
Beta: Everything works, but need to request public feedback.
Um, you know the reason you want public feedback, right? The reason is because without wider testing, it's pretty damn sure there are bugs remaining that you haven't found. These are the bugs a beta release is intended to find; until you've done a beta release, there's simply no way you can be sure they don't exist.
This appears to perfectly match what happened to Apple.
Re:Google Calendar Reviewed in PC World...
on
Google Calendar
·
· Score: 1
When the domain is pinned down, natural language processing works really well. If you're guaranteed that any sentence will follow a certain formula, parsing the sentences is pretty trivial.
I wish they'd do it for google maps too then -- its parsing is annoyingly rigid. If you screw up the simple patterns it looks for, even a little bit, it completely barfs and interprets what you typed in bizarre ways. The Japanese version of google maps is even worse; it can't even deal with many common postal addresses (even though Japanese has very obvious tag-words marking the different parts of addresses).
I don't think you should expect to be modded down for asking a reasonable and well thought-out question. Isn't that the whole point of having a discussion?
And all this time I thought it was a race to see who could talk about Nazis first...
I currently work near McCarthy Ranch. It's more like a giant strip mall with a BUNCH of standalone restaurants - and except for it and another mall on the opposite side of the freeway it's nearly a services-dead area for miles in all directions. (MIlipitas proper has a few restaurants in the old downtown. But the bulk of them are in those two malls and a third one a few miles away.)
My god... you make it sound like pretty much the most depressing place on earth. Is this really what silicon valley is like?!?
I've run linux (2.6.x!) on a system with 4 MB of RAM (yes, that's four), and a 40 MHz CPU. Everything in flash (no harddisk). Simple GUI (picogui). It worked great and was quite snappy, though the size of RAM and lack of a practical swap device rather limits the amount of multitasking you can do.
The linux kernel is very configurable[*], and of course with FOSS you've got an almost infinite selection of user-land apps to choose from. How many FOSS GUI toolkits are there? I've lost count...
Unless he's planning for his $100 laptop to use an 8-bit microcontroller with 12 KB of on-chip RAM, it sounds like Negroponte is talking out his ass... ("well that nice MS salesman called me up and pointed out how many CDs Redhat 'linux' takes up... OMG!! We'd better use windows!!")
[*] As a hacker of embedded linux, I can of course point out places where I wish it were even more configurable -- e.g., it's annoying to have that disk scheduling infrastructure around when I have no disks, and lots of VM-related code when I have no VM...:-/
Slashdot Mirror
Apparently one of the biggest advantages of the amd64 architecture (aka x86-64) is not the 64-bitness at all (though there are some everyday benefits to that too -- e.g., visiting 1GB files in Emacs :-), but that in 64-bit mode it has more registers (not just wider ones), which allows the compiler to generate better code than it can for the anemic normal x86 architecture.
You'd think that the main developer would welcome i18n patches.
Indeed, but it's not uncommon for such patches to be very intrusive, badly designed, and quite buggy (even when not using the support). For people who really need it, maybe the functionality is worth the problems, but often the tradeoffs are different for the world at large (at least from the maintainer's point of view).
The main FOSS project I'm involved with is GNU Emacs, and the people who post to the development mailing list seem to be rather evenly distributed over the entire globe (it's funny watching replies to a message slowly coming in as people wake up in successive time zones), with the exception of Africa (er, and Antarctica I suppose). What variation there is seems to match what one might expect from variations in population and economic development pretty well; of course as it's an english language list, there's also a bias towards english-speaking countries.
:-).
I expect that rather than a correlation with "piracy", you're simply seeing the effect of economic development and other obvious factors (people who don't have computers or reliable net connections, are less likely to participate
Another effect I've noticed is that in countries like Japan there are lots of "local" FOSS projects (often forks of more widespread packages) which never really seem to show their face outside that country. I guess this is due to both language and cultural issues (and some technical ones -- massive changes to support your local language may be very important to you, but the original developer may not accept them). I imagine that in countries with less widespread connectivity, this effect may be even greater.
There's a reason to go to silicon valley. The area is beautiful, and the talent pool for your $COMPANY there is tremendous, if you need 20 engineers to work on some software project, finding 20 skilled individuals in Atlanta, Georgia is going to suck. Trying to find 20 skilled people in silicon valley is a matter of going out to a busy resturant at lunch
... but I could never live there. I'd rather work for a lesser company in a civilized location.
I'm sure you're right about lots of talented people living nearby, but beautiful?!? I admit, I've only visited, but as far as I can see, silicon valley is the epitome of horrid suburban sprawl office-park mcmansion wasteland. It's one of the most depressing places I've ever been.
There are a lot of companies there that would seem very cool places to work, and a lot of the industry "focus" seems to be on SV
Well it's not just manga vs. Mishima, there's a huge number of popular modern Japanese (non-manga) writers too. There's (obviously) a vast quantity of stuff which hasn't been translated into English.
AFAIK, manga's taken a big bite of out of non-manga reading, but that seems to have been going on for a long time. It's just an offhand judgement, but in general I think Japan's (non-manga) book scene seems a lot healthier than that in the U.S -- though I guess that says more about the U.S. than Japan...
Mind-share is created by peoples' expectations, and peoples' expectations are decided based on their experience. For every person who knows the ins-and-outs of the game industry, and has some idea of what to expect with the Wii, there are lots of regular gamers who simply remember they bought a PS2 rather than a Gamecube, and will go in preferring the PS3 for that reason.
By that same logic, the xbox -- and the PS1! -- should have failed miserably. There's certainly an element of "just blindly keep buying the same brand", but there are also clearly turning points from time to time where something new comes along and captures the public's imagination.
Now, it's very hard to predict these turning points, but the Wii's extremely strong showing at E3 (both among players and developers) makes people feel that maybe such shift is coming. Maybe Nintendo's innovations will move the market, or maybe you're right that the public will just opt for the same-old same-old -- but it's hardly a sure thing either way [and of course the huge price differential hardly hurts Nintendo's cause!].
What's with this blind faith in Nintendo? Their last two consoles have been decidedly mediocre. Decent hardware, patchy library. Why's is everyone convinced this run-through is going to be different?
There's nothing particularly wrong with the n64/gc libraries -- nowhere near the quantity of ps2 games, but there are some absolutely incredible games, and certainly enough to satisfy. Nintendo's hardware and 1st-party games are of course top-notch.
I think the feeling is that with every new generation, the counter gets reset, and the only thing required for good 3rd-party support is mind-share.
In the last gen, Sony played their cards perfectly and managed to hold onto the the mind-share they had with the ps1 -- but many people seem to feel they've stumbled badly this time, and that Nintendo stands a good chance to get 3rd party support they didn't in the last gen. Given that Nintendo's pretty much guaranteed to offer high quality hardware and 1st-party games, has a vastly lower price point, and is simply doing more interesting things than Sony is, why would you not want to invest in a wii???
M$ only invested about $20 in advertising the system in Japan. Many gamers over there aren't even aware the system exists. Gamers are going to the store to buy an Xbox1 not even knowing there is a 360 out.
No, this isn't true. I live in Japan and I've seen tons of xbox360 advertising. There was an initial advertising blitz during the release, and even now, I see lots of prime-time TV ads for the 360 (they seem to sponsor lots of TV shows). in fact, I've seen far more TV advertising (here in Japan) for the 360 than for any other gaming system.
They also have a fairly strong presence in many Japanese video game stores -- prominent demo kiosks, game section not relegated to back corner despite the low sales, etc. Given the absurdly low income that must come from 360 sales, I assume this means MS is forking over lots of promotional cash to stores.
Yup. In fact with my browser and fonts, it's completely unreadable (half the text overflows the white box and ends up displayed as black on dark blue...).
I suppose it probably displays properly in Internet Explorer if you happen to have exactly the same settings as the page designer ("but, but, it worked for me!!").
OTOH, the page does have a convenient "text" button to bypass the page designer's idiocies. [A lot of sites don't seem to have even that.]
Many people across the country seem to read the New York Times, but I'm a little iffy on them. It seems that their reporters have been caught lying, and doing other unscrupulous things a number of times.
I suspect there are very few major news sources which haven't had problems like that (individual reporters lying in their stories). The main difference is that the NYT is constantly under scrutiny (being essentially the "paper of record" in the USA), especially now that many neo-conservatives seem to feel some kind of personal animosity towards it.
I think the most important issue is whether such activities result from impropriety on the part of the paper itself; as far as I know, this wasn't the case with any of the recent NYT gaffes.
More realistically: more taxes on (1) cars, or (2) cheeseburgers.
Think of the insane revenues you could raise with those...
And who are those unpatriotic commies complaining?!? Can't they see it's for the kids???
My initial thought was "Oh great, another Japanese-style bloat-o-phone" -- but after looking at the specs, I see the N93 is actually even more bloated and heavy than that (a typical Japanese bloat-phone is around 115g, and already seems kind of clunky and awkard; the N93 is 180g!).
... but it seems rather silly when it reaches the point where you don't want to carry it around at all because your pockets aren't big or strong enough. I rather like the ability to split up my gadgets as is convenient for the situation -- carry one thing in my pocket, another in my bag, leave another at home because I know I won't need it.
... but that N93 seems very far from that ideal.]
I understand some people like all-in-phone phones because they only have to carry around one device
[Of course some day all-in-one devices will probably become so small and functional that there will be little reason not to use them
Efficient mass transit, unfortunately, requires that we all work in a dense downtown area where a critical mass of people shows up.
No, it only requires that there be some concentration of people. It doesn't have to all be in one place; a network of connected nodes (a bunch of "little downtowns") works just as well, in fact, better -- if you can get people moving in both directions equally, you get double the capacity for free!
Mass transit is also unpleasant to use and generally very slow.
No, badly done mass transit is like that. It's very likely that Houston has bad mass transit of course.
More and bigger roads, logically enough, would be the better solution.
It has been shown time and time again, that this is not the right solution -- building bigger roads just gives people the impression that "it's ok" to drive wherever they want, as much as they want, so people do, and things reach exactly the same state of congestion they were at before -- only worse, because now you've got a bigger load on the rest of the infrastructure, and the landscape is starting to seem like an endless vista of pavement and parking lots. [I don't know about you, but I think giant 12-lane local roads and massive parking lots every five feet are ugly and depressing.]
The real problem is U.S. car worship. No matter what you do, tradeoffs have to be made, and as long as people remain so emotionally invested in their cars, it's very unlikely they'll be able to make rational choices about the matter.
It's interesting, because I've been in this field for 16 years, and I've never met a pure "software architect" or a pure "programmer"
Indeed; as far as I can figure these are weird-ass terms left over from the 1960s.
Certainly there are variations in the amount of coding or design responsibility that different people on a team have, but it's usually very much a matter of degree, and tends to shift around as a project develops.
I think terms like "software architect" are as much vanity titles as anything else these days -- if somebody's very good, he gets a big paycheck and his business card says "software architect," but the less clueful have to make do with "code monkey."
How about at your workplace? (Maybe you are too young now for a job, but in 5 years time?)
... my workplace is almost entirely windows based, but the first thing I do whenever they give me a new PC is wipe the disk and install my OS of choice (usually debian). [Just for the bragging rights, I make a point of not even booting windows once.]
Certainly not a given
Many coworkers are a bit bemused by this and occasionally annoyed if I can't access some random IE-only internal website, but well, I can deal.
[It's a very big company, so I think they do actually have IT nazis of some sort, but the rules tend to get bent an awful lot by the time they reach the lower levels. We've got our own sysadmin who is a lot more practical -- if you're getting your work done and not bugging him, he's on your side.]
Not to mention color depth: 24 bit color is clearly inadequate for many uses, and formats like OpenEXR (one of the current best HDR image formats; it uses 48 bits per pixel, in the form of 3 16-bit "half floats") are gaining popularity. For some uses RGB isn't sufficient either, and you want samples at many more wavelengths.
This is not specific to OpenBSD.
There is always some overhead related to MP support, and there will always be benchmarks which can show its worst effect, but OpenBSD is quite well known for being far worse in this respect than other mainstream FOSS kernels. Linux in particular has spent vast amounts of effort optimizing the MP case, and there are many core linux hackers who have very strong MP experience from IBM, SGI, etc.
However, the environmental problem of "clear cutting forests" isn't as big as is typically let on by envirowackos. We have more forest in the US now than we've had since the 1930s, and that amount is still increasing.
Note that "forest" is far too vague a word to be very meaningful -- untouched old-growth forest is a very different thing than, for instance a mono-culture of non-native trees planted at some absurdly high density. Which you prefer of course, depends on your values and goals (the most horrid examples I've seen were tree plantations in Scotland whose apparent purpose was to abuse the tax laws!).
I disagree and believe that Tivo did innovate and does deserve patent protection.
Tivo may well have thought innovatively, and be a great company -- but remember, you can't patent great ideas, and that's what exactly most of what you describe sounds like.
If they had a sufficiently non-obvious mechanism to implement some of their great ideas, they could patent the mechanism, but anybody else is quite free to come along and use a different method with the same result. Even if Tivo thought of it first.
Beta: Everything works, but need to request public feedback.
Um, you know the reason you want public feedback, right? The reason is because without wider testing, it's pretty damn sure there are bugs remaining that you haven't found. These are the bugs a beta release is intended to find; until you've done a beta release, there's simply no way you can be sure they don't exist.
This appears to perfectly match what happened to Apple.
When the domain is pinned down, natural language processing works really well. If you're guaranteed that any sentence will follow a certain formula, parsing the sentences is pretty trivial.
I wish they'd do it for google maps too then -- its parsing is annoyingly rigid. If you screw up the simple patterns it looks for, even a little bit, it completely barfs and interprets what you typed in bizarre ways. The Japanese version of google maps is even worse; it can't even deal with many common postal addresses (even though Japanese has very obvious tag-words marking the different parts of addresses).
I don't think you should expect to be modded down for asking a reasonable and well thought-out question. Isn't that the whole point of having a discussion?
And all this time I thought it was a race to see who could talk about Nazis first...
[I win!]
I currently work near McCarthy Ranch. It's more like a giant strip mall with a BUNCH of standalone restaurants - and except for it and another mall on the opposite side of the freeway it's nearly a services-dead area for miles in all directions. (MIlipitas proper has a few restaurants in the old downtown. But the bulk of them are in those two malls and a third one a few miles away.)
My god... you make it sound like pretty much the most depressing place on earth. Is this really what silicon valley is like?!?
Indeed.
:-/
I've run linux (2.6.x!) on a system with 4 MB of RAM (yes, that's four), and a 40 MHz CPU. Everything in flash (no harddisk). Simple GUI (picogui). It worked great and was quite snappy, though the size of RAM and lack of a practical swap device rather limits the amount of multitasking you can do.
The linux kernel is very configurable[*], and of course with FOSS you've got an almost infinite selection of user-land apps to choose from. How many FOSS GUI toolkits are there? I've lost count...
Unless he's planning for his $100 laptop to use an 8-bit microcontroller with 12 KB of on-chip RAM, it sounds like Negroponte is talking out his ass... ("well that nice MS salesman called me up and pointed out how many CDs Redhat 'linux' takes up... OMG!! We'd better use windows!!")
[*] As a hacker of embedded linux, I can of course point out places where I wish it were even more configurable -- e.g., it's annoying to have that disk scheduling infrastructure around when I have no disks, and lots of VM-related code when I have no VM...