Isn't this mail fraud? I mean, the FBI adds mail (or wire) fraud to just about every case they prosecute; it's an incredibly broad statute. And I wouldn't be surprised if these letters were sent over state lines, so it's a federal case...
We've all known since the P4 appeared that IF SSE2 is adopted en masse by software developers, it would have a significant advantage over every other chip on the market.
It's just that that's one heck of an "if", especially as long as sales of the P4 are as weak as they have been so far.
Too many variations of Linux
Which one is really better as they all claim one or two niches over the other.
Gnome vs. KDE vs. etc.
Why so many desktop environments, sure alternatives are good, but when work needs to be done, money is burnt by time spent figuring out whats what on Linux vs. point and click MS
Others have already noted the flaws with the "command line is too hard" attitude. As far as these two points go, what people fail to consider is that a company that's rolling out Linux on the desktop is going to choose ONE distribution for everyone, with ONE desktop environment. The end user doesn't need to be bothered with the multitude of choices.
In fact, I'd imagine that the most common situation would be where most desktops were just X terminals to a central server, giving tech support full control over desktop configurations. This would actually provide a greater degree of conformity and control than you can easily get under Windows without add-on products (like MS's SMS or Novell's ZENworks).
I can only assume that this is intentional blindness, as many of the people that advance these arguments are the same people that would not hesitate to establish corporate computing standards in other situations. It's really just a matter of planning.
(Note that I am not saying that Linux is actually ready for the corporate desktop -- that's a separate discussion, and I don't personally think it is outside of very specific situations. But the argument that the diversity of options available actually hurts Linux is at best silly.)
To get back to the contention of the original article, even if you only consider the corporate environment, I don't think we'll see the playing field narrowed to Caldera and Red Hat. I think some Debian-based distribution (possibly Debian itself) will very likely be in the mix, by virtue of the apt system (would make it trivial to install and upgrade new -- even custom -- packages on every Linux system in the company; just put up your own server with.deb packages and add that server to sources.list and there you are). This could change if someone created a similarly easy system for RPM, but the options out there now for RPM just aren't as good yet.
Citrix MetaFrame, while a nice solution to this problem (on the platforms it's available for), is mucho expensive. Not really appropriate for a probably budget-conscious school.
Is someone under the silly impression that web-surfing, etc., takes less concentration than talking on a cell phone?
At least with a cell, there are hands-free kits and some limited voice-dialing functionality available; similar tech for computers, while it's been in the works for (it seems like) forever, just isn't up to the job yet. So you'll have Joe Public bending over to check the score of the hockey game from last night (or, worse, checking the porn he got in his e-mail last night *shudder*) and meeting a tree at 40 mph.
If they weren't so likely to take others along with them, I'd think it was a nice bit of auto-Darwination (pun not intended).
Let me preface this by saying that I don't believe these scores myself.
That said, I'm not sure where the author you're quoting got the idea that it's difficult to overclock a 1.2Ghz Athlon to 1.53Ghz. I've seen a 1.1Ghz Athlon clocked that high, and 1.2Ghz clocked significantly higher (1.7 isn't a stretch at all).
It might be more correct to say that without changing the FSB speed, it's very difficult to get 1.53 out of a 1.2 Ghz Athlon, and that would be a reasonable caveat to make in this case, since the Tyan board doesn't provide for changing the FSB speed easily, but that's not what the passage you quoted said.
True, but unlikely. They don't hold that much stock (i.e., most of the shares are outstanding, in investor's hands), so it would have to be at least part cash. They've stated (in conference calls, etc.) that they are interested in expanding their fab capacity; spending most of their $1.8B or so of cash on hand on an acquisition/merger would make this almost impossible.
Also, acquisitions usually are at a premium to the current stock price, so AMD would probably have to pay well over TMTA's current $3+ billion to acquire them. They just can't afford it.
Granted, but that isn't worth $3+ billion (to a company that's currently worth $7 billion). If AMD had more cash on hand, or TMTA was cheaper, it might be different.
It might well be a great long-term investment, but AMD just couldn't afford it without strangling themselves for the next several years.
This isn't going to happen -- not now, at any rate. Transmeta's market cap is almost half of AMD's; that's a huge amount of money that AMD could better put to use building more fab capacity to continue chipping away at Intel.
Not to mention that AMD already is licensing the most important thing it could get from Transmeta -- the code-morphing tech so that they can simulate their upcoming chips.
If Transmeta had some fab capacity of their own, it might be different, but IBM produces their chips, and I don't think AMD's quite ready to buy IBM.:-)
Maybe in a couple years when AMD has an extra couple billion dollars sitting around and/or Transmeta's stock crashes, but for now, AMD is selling every Athlon they can produce without any help. There's just no good financial or technical reason for AMD to do this right now, not in their current position.
AMD's Hammer chips (the Sledgehammer for servers and the Clawhammer for desktops -- the core is the same; the main difference, IIRC, is in the amount of cache) will (according to AMD, anyway) run 32-bit software just as quickly as a 32-bit chip. From what I have heard, this is actually a credible claim, and not just marketing blather. It is also expected to debut at speeds near 2 Ghz. Unfortunately, not until 1Q2002.:-(
The Itanium, on the other hand, will run 32-bit software like a one-legged garden slug; it will debut no higher than 800 Mhz, and clock-for-clock will be terrible on 32-bit code (as in, much worse than any other Intel chip currently on the market). But if you must have a 64-bit chip now (for values of now equal to early next year), it's the only x86-ish game in town.
(Though given its performance shortfalls, that it will be a brand new chip -- with all the baggage that carries -- and the expense, I'm not sure why anyone who needs 64-bit now wouldn't go buy something from one of the big-box vendors...)
10,000 machines runing Windows and MS Office.
10,000 machines running Linux with StarOffice.
the difference is six figures.
Seven, actually. Even at quantities of 10,000, the price for Office 2000 is well over $100 each. That's a million right there, not counting Windows, vs. less than $100 for one boxed set of the Linux distribution of your choice.
(Yes, MS site licensing will change the numbers a bit, but trust me -- you'll end up paying over $1 million for 10,000 machines w/ Office and Windows.)
An interesting question would be what is the fastest single processor (including ASCI's speed divided by 8192) currently in existance. Is it still a Cray? Are the individual processors that make up the ASCI particularly impressive in themselves?
Fastest single processor depends heavily on what the problem is. Doing vector ops, a Cray's going to be a lot more impressive than if it's doing floating point. For some things, the AMD's 1.2 Ghz Athlon is almost certainly the fastest available (i.e., actually for sale as opposed to just in a chipmaker's labs).
As for ASCI-White, according to IBM, each processor is a 375 MHz POWER3.
Maybe I'm overlooking something...
on
3D Printers
·
· Score: 2
...but it seems like there's a reasonably simple answer for the problem the article speaks of in regard to the huge amounts of data needed to encode all of the "voxels" of a complex 3D object, at least in most cases.
In most objects, you're not going to be using a great variety of different materials, and those you use are going to be arranged in (usually) fairly logical patterns. You've got a lump of aluminum over here, a strip of some polymer over there, etc. You usually wouldn't want a voxel of aluminum here, the next one is iron, the next one is plastic, randomly dispersed. (Unless the voxel resolution was very coarse.)
So, like with some 2-D graphical formats, adjacent voxels of the same color, material, etc., could get encoded together. To represent a 200x200x200 cube of aluminum, you don't need to specify each of the 8 million voxels; rather, a handful of coordinates will suffice. Obviously, this is a degenerate case, but even in more "real" cases, this would provide dramatic improvements over what they seemed to be discussing in that part of the story.
It also seems like some sort of "polyhedronal" (like polygonal, except in 3d) encoding, like is used by many video accelerators, might also provide some benefit.
Of course, this all seems too obvious for the companies working on this not to have thought of, so I'll just go back to my corner now...
But that's not what's at issue here. It doesn't matter (for the purposes of the claim "Linux has absolutely no market share", or even the weaker claim "Linux has very little market share") whether the people using Linux are geeks, Grandma, or my hairdresser's niece.
The concept of market share is blind to all of that, and it's pretty clear that someone is using Linux on the desktop, and it's a growing someone.
(One could even make the argument that the people using Linux now are some of the most important in terms of market share: mostly geeks -> geeks are higher-paid on average -> have more expendable income to spend on software. Also, those Linux users working in IT/IS at some company will sometimes have some degree of influence over software choices...or at least are the people who will make those decisions in a few years. At the very least, they have more influence than "Joe Public".)
Ordered a Prius myself a month ago (delivery sometime in November -- they're massively backordered), and the warranty is incredible. Three years bumper-to-bumper everything, including normal tune-ups, EIGHT years on all of the electric parts (including the battery). And they estimate the battery should last quite a bit longer than that under most circumstances.
It's a tempting idea. In fact, I don't doubt that the OS market will be dominated by Open Source products within 5-10 years; maybe even certain other markets for common software that doesn't require particularly specialized expertise to write (office suites, anyone?).
But there will be two classes of software that will always support a healthy percentage of proprietary software, at least as long as capitalism is around:
1. Anything that requires some sort of rare, specialized knowledge. High-end scientific software, high-end accounting software, etc. The pool of developers that would be able to contribute to something in this category is just too small to make a purely open-source model workable.
2. Games. Users demand the latest and greatest and have repeatedly shown that they are willing to pay for it. As long as this is true, and game companies can keep up with those demands without going open-source, they're not going to do it. What's more, the usual open-source-related revenue streams just don't make sense for game companies (who's going to pay for service and support on Starcraft?:-). I wouldn't be surprised to see some companies releasing games commercially for a while, then once sales slow down, open the source. (Especially to gaming engines, since the company has as much or more to gain from advances there than anyone else.)
For cheap, no-frills access, check out some of the little ISPs in your area. Most of them are having enough trouble surviving that they'll be ecstatic to have your business, no matter what OS you're using. Many of them have $8-10 nothing-but-dialup dynamic IP accounts.
For what it's worth, most of the free access providers are losing a lot of money. (Most of them are publicly owned companies -- you can check this on Yahoo or the like.) I'm not surprised that any plans they may have to support Linux are being delayed.
And most universities don't get free T3's or anything like that. They do get big discounts, but my alma mater has an OC-12 (~150 Mb/s) pipe for which they pay tens of thousands of dollars per month. (Yes, any other organization would pay more, but it's far from free.) And most of these larger, cheaper connections are to so-called "Internet-II", which is just between research univerisities. They don't get nearly as much of a discount on their (usually smaller) regular Internet pipes.
Yeah, well. You know what I'm sick of? People whining that they can't get free internet access. It's a freaking utility. Do you expect to get electricity for free? Phone? Gas? So why should you expect that for internet access?
It makes sense to offer low-cost dial-up plans for people who are legitimately *really* poor, just like the phone company does...but whining that DSL and cable are just too expensive and about how the "elites" are keeping it all to themselves? Gimme a freaking break. It's expensive to lay the infrastructure for high-bandwidth connections, and contrary to popular belief, phone and cable companies are not making money hand-over-fist off these services.
(I know, I know, cry me a river...but these companies are in business to make money, and they've got to cover their costs like anyone else.)
Someday, high-speed access will be cheap/free, when bandwidth is no longer a scarce good. But by then, we'll have changed our idea of what "high-speed" is, so people will be whining about that, I'm sure.
Both points well taken. However, this is only true up to about 1600x1200 resolution (IIRC), and I expect it'll be soon enough that screen size and resolution will increase to the point where that's not enough any more.
Firewire (and AGP 8x) are obviously well ahead of the rest of the hardware used for most applications, but I wouldn't expect that to stay true.
Games always seem to keep pace. Very-high-res "real"-looking video could eat up most of what Firewire currently is capable of.
Maybe something that looks more vivid than real life. Then I won't have to go out in the Big Blue Room anymore. The savings on sunscreen alone should pay for my new Super-Ultra-Neato-Keen-AGP 32x video card...
But 2600 is itself (as much as Time-Warner, et al., might not like it) a journalistic source. If CNN can write about (and link to) the source, what's fundamentally different about 2600?
The hypocrisy here is Time-Warner's, not at all CNN's.
Slashdot Mirror
Isn't this mail fraud? I mean, the FBI adds mail (or wire) fraud to just about every case they prosecute; it's an incredibly broad statute. And I wouldn't be surprised if these letters were sent over state lines, so it's a federal case...
It's just that that's one heck of an "if", especially as long as sales of the P4 are as weak as they have been so far.
Gnome vs. KDE vs. etc. Why so many desktop environments, sure alternatives are good, but when work needs to be done, money is burnt by time spent figuring out whats what on Linux vs. point and click MS
Others have already noted the flaws with the "command line is too hard" attitude. As far as these two points go, what people fail to consider is that a company that's rolling out Linux on the desktop is going to choose ONE distribution for everyone, with ONE desktop environment. The end user doesn't need to be bothered with the multitude of choices.
In fact, I'd imagine that the most common situation would be where most desktops were just X terminals to a central server, giving tech support full control over desktop configurations. This would actually provide a greater degree of conformity and control than you can easily get under Windows without add-on products (like MS's SMS or Novell's ZENworks).
I can only assume that this is intentional blindness, as many of the people that advance these arguments are the same people that would not hesitate to establish corporate computing standards in other situations. It's really just a matter of planning.
(Note that I am not saying that Linux is actually ready for the corporate desktop -- that's a separate discussion, and I don't personally think it is outside of very specific situations. But the argument that the diversity of options available actually hurts Linux is at best silly.)
To get back to the contention of the original article, even if you only consider the corporate environment, I don't think we'll see the playing field narrowed to Caldera and Red Hat. I think some Debian-based distribution (possibly Debian itself) will very likely be in the mix, by virtue of the apt system (would make it trivial to install and upgrade new -- even custom -- packages on every Linux system in the company; just put up your own server with .deb packages and add that server to sources.list and there you are). This could change if someone created a similarly easy system for RPM, but the options out there now for RPM just aren't as good yet.
Citrix MetaFrame, while a nice solution to this problem (on the platforms it's available for), is mucho expensive. Not really appropriate for a probably budget-conscious school.
Ix
At least with a cell, there are hands-free kits and some limited voice-dialing functionality available; similar tech for computers, while it's been in the works for (it seems like) forever, just isn't up to the job yet. So you'll have Joe Public bending over to check the score of the hockey game from last night (or, worse, checking the porn he got in his e-mail last night *shudder*) and meeting a tree at 40 mph.
If they weren't so likely to take others along with them, I'd think it was a nice bit of auto-Darwination (pun not intended).
Ix
That said, I'm not sure where the author you're quoting got the idea that it's difficult to overclock a 1.2Ghz Athlon to 1.53Ghz. I've seen a 1.1Ghz Athlon clocked that high, and 1.2Ghz clocked significantly higher (1.7 isn't a stretch at all).
It might be more correct to say that without changing the FSB speed, it's very difficult to get 1.53 out of a 1.2 Ghz Athlon, and that would be a reasonable caveat to make in this case, since the Tyan board doesn't provide for changing the FSB speed easily, but that's not what the passage you quoted said.
Ix
Also, acquisitions usually are at a premium to the current stock price, so AMD would probably have to pay well over TMTA's current $3+ billion to acquire them. They just can't afford it.
It might well be a great long-term investment, but AMD just couldn't afford it without strangling themselves for the next several years.
Not to mention that AMD already is licensing the most important thing it could get from Transmeta -- the code-morphing tech so that they can simulate their upcoming chips.
If Transmeta had some fab capacity of their own, it might be different, but IBM produces their chips, and I don't think AMD's quite ready to buy IBM. :-)
Maybe in a couple years when AMD has an extra couple billion dollars sitting around and/or Transmeta's stock crashes, but for now, AMD is selling every Athlon they can produce without any help. There's just no good financial or technical reason for AMD to do this right now, not in their current position.
The Itanium, on the other hand, will run 32-bit software like a one-legged garden slug; it will debut no higher than 800 Mhz, and clock-for-clock will be terrible on 32-bit code (as in, much worse than any other Intel chip currently on the market). But if you must have a 64-bit chip now (for values of now equal to early next year), it's the only x86-ish game in town.
(Though given its performance shortfalls, that it will be a brand new chip -- with all the baggage that carries -- and the expense, I'm not sure why anyone who needs 64-bit now wouldn't go buy something from one of the big-box vendors...)
the difference is six figures.
Seven, actually. Even at quantities of 10,000, the price for Office 2000 is well over $100 each. That's a million right there, not counting Windows, vs. less than $100 for one boxed set of the Linux distribution of your choice.
(Yes, MS site licensing will change the numbers a bit, but trust me -- you'll end up paying over $1 million for 10,000 machines w/ Office and Windows.)
Fastest single processor depends heavily on what the problem is. Doing vector ops, a Cray's going to be a lot more impressive than if it's doing floating point. For some things, the AMD's 1.2 Ghz Athlon is almost certainly the fastest available (i.e., actually for sale as opposed to just in a chipmaker's labs). As for ASCI-White, according to IBM, each processor is a 375 MHz POWER3.
In most objects, you're not going to be using a great variety of different materials, and those you use are going to be arranged in (usually) fairly logical patterns. You've got a lump of aluminum over here, a strip of some polymer over there, etc. You usually wouldn't want a voxel of aluminum here, the next one is iron, the next one is plastic, randomly dispersed. (Unless the voxel resolution was very coarse.)
So, like with some 2-D graphical formats, adjacent voxels of the same color, material, etc., could get encoded together. To represent a 200x200x200 cube of aluminum, you don't need to specify each of the 8 million voxels; rather, a handful of coordinates will suffice. Obviously, this is a degenerate case, but even in more "real" cases, this would provide dramatic improvements over what they seemed to be discussing in that part of the story.
It also seems like some sort of "polyhedronal" (like polygonal, except in 3d) encoding, like is used by many video accelerators, might also provide some benefit.
Of course, this all seems too obvious for the companies working on this not to have thought of, so I'll just go back to my corner now...
The concept of market share is blind to all of that, and it's pretty clear that someone is using Linux on the desktop, and it's a growing someone.
(One could even make the argument that the people using Linux now are some of the most important in terms of market share: mostly geeks -> geeks are higher-paid on average -> have more expendable income to spend on software. Also, those Linux users working in IT/IS at some company will sometimes have some degree of influence over software choices...or at least are the people who will make those decisions in a few years. At the very least, they have more influence than "Joe Public".)
Ordered a Prius myself a month ago (delivery sometime in November -- they're massively backordered), and the warranty is incredible. Three years bumper-to-bumper everything, including normal tune-ups, EIGHT years on all of the electric parts (including the battery). And they estimate the battery should last quite a bit longer than that under most circumstances.
But there will be two classes of software that will always support a healthy percentage of proprietary software, at least as long as capitalism is around:
1. Anything that requires some sort of rare, specialized knowledge. High-end scientific software, high-end accounting software, etc. The pool of developers that would be able to contribute to something in this category is just too small to make a purely open-source model workable.
2. Games. Users demand the latest and greatest and have repeatedly shown that they are willing to pay for it. As long as this is true, and game companies can keep up with those demands without going open-source, they're not going to do it. What's more, the usual open-source-related revenue streams just don't make sense for game companies (who's going to pay for service and support on Starcraft? :-). I wouldn't be surprised to see some companies releasing games commercially for a while, then once sales slow down, open the source. (Especially to gaming engines, since the company has as much or more to gain from advances there than anyone else.)
Maybe I'm being trolled here, but don't ya think a grammar flame ought to be free of spelling errors?
For what it's worth, most of the free access providers are losing a lot of money. (Most of them are publicly owned companies -- you can check this on Yahoo or the like.) I'm not surprised that any plans they may have to support Linux are being delayed.
And most universities don't get free T3's or anything like that. They do get big discounts, but my alma mater has an OC-12 (~150 Mb/s) pipe for which they pay tens of thousands of dollars per month. (Yes, any other organization would pay more, but it's far from free.) And most of these larger, cheaper connections are to so-called "Internet-II", which is just between research univerisities. They don't get nearly as much of a discount on their (usually smaller) regular Internet pipes.
It makes sense to offer low-cost dial-up plans for people who are legitimately *really* poor, just like the phone company does...but whining that DSL and cable are just too expensive and about how the "elites" are keeping it all to themselves? Gimme a freaking break. It's expensive to lay the infrastructure for high-bandwidth connections, and contrary to popular belief, phone and cable companies are not making money hand-over-fist off these services.
(I know, I know, cry me a river...but these companies are in business to make money, and they've got to cover their costs like anyone else.)
Someday, high-speed access will be cheap/free, when bandwidth is no longer a scarce good. But by then, we'll have changed our idea of what "high-speed" is, so people will be whining about that, I'm sure.
Firewire (and AGP 8x) are obviously well ahead of the rest of the hardware used for most applications, but I wouldn't expect that to stay true.
This should make AGP 4x more affordable...
Maybe something that looks more vivid than real life. Then I won't have to go out in the Big Blue Room anymore. The savings on sunscreen alone should pay for my new Super-Ultra-Neato-Keen-AGP 32x video card...
Nah, it's too late by then. They're the media -- they know what you're thinking...
The hypocrisy here is Time-Warner's, not at all CNN's.