I don't know if it can be done with the current system, but I do believe that votes can be counted to better than Poisson statistics (where the variance is equal to SQRT(N).)
However, I do not, and will not, believe any count of millions of votes is ever going to be truly accurate to a few votes, or even a few hundred votes. I am convinced that in both Florida, and the national popular election, the election was, as well as we are able to measure it, exactly a tie.
I wish we'd just pick one of the two guys and get on with it. All this talk about the "will of the People of (Florida | the USA)" is just so much hot air. Either guy has just under 50% of either population behind him. Both got more than half again as many votes as Bill Clinton did in 1992. I don't think we are capable of knowing which of the two guys "really" did slightly better than the other.
The only reason the system hasn't completely broken down before is becasue it was never this close. In cases where the popular vote was this close, the electoral vote was balance far enough one way to indicate a clear winner (e.g. Kennedy in 1960). We will be better off if we accept it was a tie, quickly pick one of the two guys on whichever technicality we want to, and move on. A lot of hemming and hawing about how to best count millions of votes down to the last vote is going to lead nowhere.
Ping latency of 400ms isn't bad only for gamers... it's bad for a lot of things.
I do a lot of work over my cable modem. This means that I use my home PC as an X server, and run terminals, editor sessions, etc., on my work computers, displaying to my home computer. Not to mention graphic displays (astronomical images, plots, etc.). When the ping gets up over 100ms, the "feel" and feedback of the terminal sessions and editor sessions starts to get annoying. If it's 400ms, it's to the point where I'd consider it unusable.
The bandwidth is crucial for when I'm displaying images (and, of course, when I'm cruising the web). But, I'd trade off some of that for better latency.
You know what this reminds me of more than anything? Back in 1990 or thereabouts, Amiga faithful would, in inevitable OS superiority arguements with MS-DOS/Windows 3.1 types, point at the Amiga's ability to multitask as one of the superior points of the Amiga. The inevitable response from the DOS/Windows crowd was, "who needs to multitask? I only run one thing on my computer at once anyway. That's a useless feature for a desktop OS."
I doubt you'd hear any Windows folks making that argument today.
The uptime argument is the same thing. If you're arguing that you don't need (multitasking, stability from crashing/rebooting, etc.), it's only because you're defensive about your own operating system and don't want to admit its faults. When you finally have it, you will realize that it's a great thing to have.
Third, the courts have found that reverse engineering is legal (as long as chunks of the original software code are not present in the resulting software).
Hasn't the DMCA effectively made reverse engineering illegal? (Or has the DMCA not yet really been tested in court? I thought the DeCSS case was an example of a judge indicating that we aren't going to get reasonable judgements about DMCA.) (A reasonable judgement meaning the trashing of the whole thing.)
On the bright side: consider that most of the recent highly publicized security breaches were associated with Microsoft Outlook's feature that it can automatically run executable attachments. This has been used to cause untold harm-- and to gain *access* to Microsoft's private source code, which was "effectively" protected by passwords and so forth.
To me, it looks like Microsoft Outlook is a circumvention technology. Somebody should sue Microsoft, and jail all of their programmers....
If it weren't for the 2nd amendment, wide ranging gun control would have been a reality in this country long ago. Think what you like about gun control, or even the "militia" interpretation of the 2nd amendment, and so forth, that amendment is a clause in the Constitution which gun control opponents can point to, giving them a powerful ally.
Unfortunately, there is no rule in the Constitution protecting our right to build and own technology-- devices or software. While shooting somebody with a gun is very illegal, owning the gun is not. You'd think that actually pirating, and violating copyright, is all that would need to be illegal. But, no, because we don't have a specific clause in the constitution protecting that right, even purchasing or building something which could be used for copyright violation is illegal. It's just crazy. It's so ironic I could just weep.
Perhaps the gun lobby is powerful enough that it could be enlisted as an ally? The parallels are clear; devices which themselves are not illegal, but which have readily apparent illegal uses. Perhaps the gun lobby could be convinced that the DMCA, once it stands, could be used as a precedent to weaken their case further.
It's really too bad that most of the population our country is not technically savvy enough to really understand these issues. As long as they can stick a DVD in their Windows machine and have it work, they're happy. And, they love to see the law crack down on all the dangerous pirates and hackers and similar scary people.
AT&T purchased TCI, so now in areas of the country where TCI offered @Home the service is billed as "AT&T/@Home". @Home remains an idependent entity, not considering that Excite and @Home merged some time ago.
OK, I stand corrected. All I really know is that the bills I get for my cable modem service come from "AT&T @Home", and that in this area, getting a cable modem means getting it from those folks. I was under the impression that cable modems everywhere were @Home, and I made the incorrect conclusion that since my @Home is AT&T, everyone's was.
You can say the same about Education, Healthcare, The War on Drugs ("Drugs are BAD!"), Welfare,
or any other program that the republicats push.
Very true. The trick is to figure out which ones are worth supporting, how much they're supporting, the degree to which government support of them mucks things up compared to private sector support of them, the degree to which private sector support fails.
Perhaps I'm painting myself as "part of the mindless mob" in saying this; a true libertarian feels that government should do nothing but mutual protection. I guess I'm not a true libertarian.
Yes, parents who value education will, under a completely free-market system, ensure that their kids get good education. Even if they are poor, I suspect that systems would evolve that would allow them to get a decent education. But it is my opinion that we *all* benefit the more that *everybody* is educated-- even the people whose parents won't pony up. I don't see it as strictly an entitlement; it's also "enlightened self interest."
Remember, @Home is owned by AT&T, those fun people who were talking about charging online merchants a fraction of any sale to a AT&T Broadband customer, and even for just "delivering" those customers to the merchant in the first place.
I believe that AT&T/@Home has a vritual monopoly in the US on the cable modem lines. (My cable modem is certinaly AT&T/@Home, and it makes me think I should be posting this anonymously... oh well.) If it weren't for DSL, it would be a virtual broadband monopoly. Is there a lesson in here somewhere? A lesson that's been learned over and over again in history?
If we had other cable companies to get our cable modem service from, we wouldn't have to endlessly bitch about @Home. From the other end, @Home would have to clean up their act to keep their customers. I know I've swicthed phone-in ISPs several times, and have found one which has service and capabilities that really match what I want. If we can't vote with our feet, but can only bitch against people who have weird "intellectual property" laws to stifle that bitching, we're hosed.
Before I say anything, I should note that I'm a scientist on the public dole, so I'm biased.
That being said, even though I have some libertarian tendencies I think that Browne's plan to end all government scientific funding is foolish. The reason: basic research is one of the best investments you can make. It is almost guaranteed to pay off. The problem is, you will invest in 1,000 research programs, and only have one program pay off. That one program will pay off to more than make up the investment for the other thousand-- and you will not be able to predict which one it was back at the beginning of the research program.
Some corporate funding of research has worked well in the past (Bell Labs?), but it just doesn't seem to be feasable today. Investing in basic scientific research is just too long term for most corporations. Never mind "five year plans" or even retirement times for top executives, you may not be able to fund enough projects to have any statistical confidence that any but the most applied of research programs may pay off for you. And the payoffs may be something unexpected, which you will have trouble reaping the benefits of anyway.
Scientific research is one of those things where everybody benefits (even if they don't realize it), and it is in everyone's interest to pool their resources to fund. But how to manage that? Well, isn't that what government is? My libertarian tendencies show themselves when I think that most people talk about government in the wrong way nowadays. The "we are your children" incident from one of the Clinton debates, which wasn't disupted by any of the candidates present, was a bad sign. To many of us see government as our parents, our protectors, those people who have control over us. They are our benign keepers. Yeah, they listen to us, and via voting we get to have some input into what we want done, but in the end many people in the USA see government as a particularly nice Big Brother.
Really, it should just be our way of acting collectively. The government should *be* us. It should be the way that we, as a society, perform the things that can only be done on a whole-societal level. My differences with libertarians come in as to what some of those things are. Scientific research is definitely one. Support of the arts is another-- rich individuals, and governments, are traditionally patrons of the arts. The arts have (mostly intangible) cultural value, but (with some very obvious exceptions) not much commercial value. Do we really want to let this part of our humanity go? Or is it worth some very small fraciton of our collective resources to support this endeavor? (When I say very small fraction, just compare arts funding in any government to defense, infrastructure, and sundry entitlements.)
The one question I have: won't this make it entirely possible and legal for somebody to pull what Microsoft tried with their "J++"? Microsoft got slapped for that. But, suppose that Suns JVM was GPL'ed; that would make it difficult for them to license the spec and slap Microsoft the way they did for J++. Then suppose that Microsoft releases a J++, also GPL'ed to keep it legal, but so tied into the Windows API that it will effecitvely only ever work on Windows. Microsoft then pushes it, hard, making IE only work with J++, trying to get all Windows users out there to move to this polluted version of Java.
Cross platform Java dies. Sure, if J++ is open sourced, people can eventually create emulators that will let it run elsewhere, but I'd much rather have a Java that was designed from the beginning to be cross platform, wouldn't you?
Mind you, I know this is a Chicken Little scenario, but it *would* have been possible of JVM was open sourced all along, I believe. Nowadays, Microsoft is going to try to push C# to replace Java, so Sun is opening itself up less to people like MS by fully open sourcing Java.
With luck, the fact that it's so bloody easy for everybody to hack thier device will teach them a lesson or two about paying really stupid, boneheaded patents.
Wouldn't it be beautiful if every company who both foisted and bought into stupid patents withered and died? Evolution in action. It would be nice. Amazon sitting in the graveyard next to Apple, with tourguides giving cautionary tales about stepping bast the bounds of common sense in the drive to abuse intellectual property laws.
A GIMP expert has to choose between "giving away" improvements that make GIMP better by patching the UI, or writing a book and making a quick buck.
Each GIMP expert should play to his strengths. Only sufficiently skilled programmers are capable of contributing to the code base itself. A non-programmer who's figured out the GIMP and is effective at communication best contributes by writing documentation. Note that while the authors of "Grokking the Gimp" may well make a quick buck on it, they have in fact put the text of the book online-- so, even if they do make money, you can't deny it's a contribution to users of the program!
I've got the "Gimp Essential Reference" myself, and find that it's by and large an excellent book. Before this book, I didn't really understand layers.... My one complaint is that the chapter on writing scripts and plugins is sketchy enough that it should be presented as more of an introduction. Additionally, with the intorudction of Gimp-Perl, I expect more people will want to use that than the Scheme language which is the default described in the book.
I bet it has a lot more to do with how you're cocking your head, with the quality of the sound reception, etc., than it does with anything that the phone itself is emitting.
I mean, heck, I get a headache after talking on any phone for more than a little while. And I get a headache if the neighbors are playing their stereo too loud. Even though it's not painfully loud at my house, all I hear is a rhythmic "booming" noise, just at the edge of my hearing. Drives me nuts.
Lots of stimuli can cause irritation and headaches; it doesn't take some sort of boogey man "radiation" to do it.
The leading cause of cellphone injury is probably car accidents.
-Rob
How will they know? And why won't they just block?
on
High-Speed Greed
·
· Score: 1
How is AT&T going to know what a "web retailer" is? I mean, sure, it's obvious for the big ones; you go to www.landsend.com, or www.amazon.com, or www.cdw.com, and AT&T's gonna know it's a retailer. But what of the million and one dinky stores out there that let you buy things over the web? How is AT&T going to tell the difference between a browser visiting a retailers site or a browser just browsing?
Second, if I ran a small outfit that marketed over the web, and suddnely AT&T was going to start *charting* me for every customer that they somehow claim to have redirected my way, you had better believe I'd start blocking all AT&T addresses from my site. Hate to lose the sales, but hate to be charged by AT&T for every browser they want to take credit for.
How could they even enforce this?
I sure hope this is just more noise, because of anything comes of it, it's going to be very ugly.
-Rob
Last year, two independent groups announced results from extremely distant supernovae that rule out the critical/no comsological constant universe to very high confidence. Unless there is some as-yet undetected systematic problem with the data, it probably doesn't make sense to keep reporting age estimates based on that universe. See http://www-supernova.lbl.gov
I had a huge headache at work a week and a half ago when everything died. I'm in a scientific group, and we've mostly used Solaris in the past. We've been ramping up our Linux usage, and on the whole Linux has been *more* stable than Solaris. However, we just recently started writing in bulk to a Linux disk NFS exported to a Solaris machine, and the nfs daemon *kept* *dying*. Very annoying.
I solved it with some Alan Cox patches that included H.J. Lu's latest knfsd. So Linux isn't as unreliable as it first looked. But there are a few places where Linux still does falter.
I think Thompson greatly overstates it, however. And, it is important qualifier on my NFS problems that there were patches out there I could apply that solved them.
Linux is riding the anti-Microsoft trend, but only in it's current state of getting popular. Most of us longterm Linux geeks were using Linux before it suddenly became fashionable. Calling the current Linux Phenomenon nothing more than Microsoft backlash might be reasonable. Calling Linux itself nothing more than Microsoft backlash is dumb, and clearly ignores a lot of the even recent history of Linux.
Slashdot Mirror
I don't know if it can be done with the current system, but I do believe that votes can be counted to better than Poisson statistics (where the variance is equal to SQRT(N).)
However, I do not, and will not, believe any count of millions of votes is ever going to be truly accurate to a few votes, or even a few hundred votes. I am convinced that in both Florida, and the national popular election, the election was, as well as we are able to measure it, exactly a tie.
I wish we'd just pick one of the two guys and get on with it. All this talk about the "will of the People of (Florida | the USA)" is just so much hot air. Either guy has just under 50% of either population behind him. Both got more than half again as many votes as Bill Clinton did in 1992. I don't think we are capable of knowing which of the two guys "really" did slightly better than the other.
The only reason the system hasn't completely broken down before is becasue it was never this close. In cases where the popular vote was this close, the electoral vote was balance far enough one way to indicate a clear winner (e.g. Kennedy in 1960). We will be better off if we accept it was a tie, quickly pick one of the two guys on whichever technicality we want to, and move on. A lot of hemming and hawing about how to best count millions of votes down to the last vote is going to lead nowhere.
-Rob
Err... I believe that they list Gore as "Win" simply because he won the state. It's for reference, not indicating who won each county.
I have no doubt whatsoever that Gore has carried California. I would have told you the same thing weeks (months? years?) ago.
-Rob
Ping latency of 400ms isn't bad only for gamers... it's bad for a lot of things.
I do a lot of work over my cable modem. This means that I use my home PC as an X server, and run terminals, editor sessions, etc., on my work computers, displaying to my home computer. Not to mention graphic displays (astronomical images, plots, etc.). When the ping gets up over 100ms, the "feel" and feedback of the terminal sessions and editor sessions starts to get annoying. If it's 400ms, it's to the point where I'd consider it unusable.
The bandwidth is crucial for when I'm displaying images (and, of course, when I'm cruising the web). But, I'd trade off some of that for better latency.
-Rob
Uptime is not a desktop client feature
You know what this reminds me of more than anything? Back in 1990 or thereabouts, Amiga faithful would, in inevitable OS superiority arguements with MS-DOS/Windows 3.1 types, point at the Amiga's ability to multitask as one of the superior points of the Amiga. The inevitable response from the DOS/Windows crowd was, "who needs to multitask? I only run one thing on my computer at once anyway. That's a useless feature for a desktop OS."
I doubt you'd hear any Windows folks making that argument today.
The uptime argument is the same thing. If you're arguing that you don't need (multitasking, stability from crashing/rebooting, etc.), it's only because you're defensive about your own operating system and don't want to admit its faults. When you finally have it, you will realize that it's a great thing to have.
-Rob
Dude, passage of laws like the DMCA indicate that there is already a measurable segment of humanity who aren't living on this planet.
-Rob
From the article:
Third, the courts have found that reverse engineering is legal (as long as chunks of the original software code are not present in the resulting software).
Hasn't the DMCA effectively made reverse engineering illegal? (Or has the DMCA not yet really been tested in court? I thought the DeCSS case was an example of a judge indicating that we aren't going to get reasonable judgements about DMCA.) (A reasonable judgement meaning the trashing of the whole thing.)
-Rob
On the bright side: consider that most of the recent highly publicized security breaches were associated with Microsoft Outlook's feature that it can automatically run executable attachments. This has been used to cause untold harm-- and to gain *access* to Microsoft's private source code, which was "effectively" protected by passwords and so forth.
To me, it looks like Microsoft Outlook is a circumvention technology. Somebody should sue Microsoft, and jail all of their programmers....
-Rob
If it weren't for the 2nd amendment, wide ranging gun control would have been a reality in this country long ago. Think what you like about gun control, or even the "militia" interpretation of the 2nd amendment, and so forth, that amendment is a clause in the Constitution which gun control opponents can point to, giving them a powerful ally.
Unfortunately, there is no rule in the Constitution protecting our right to build and own technology-- devices or software. While shooting somebody with a gun is very illegal, owning the gun is not. You'd think that actually pirating, and violating copyright, is all that would need to be illegal. But, no, because we don't have a specific clause in the constitution protecting that right, even purchasing or building something which could be used for copyright violation is illegal. It's just crazy. It's so ironic I could just weep.
Perhaps the gun lobby is powerful enough that it could be enlisted as an ally? The parallels are clear; devices which themselves are not illegal, but which have readily apparent illegal uses. Perhaps the gun lobby could be convinced that the DMCA, once it stands, could be used as a precedent to weaken their case further.
It's really too bad that most of the population our country is not technically savvy enough to really understand these issues. As long as they can stick a DVD in their Windows machine and have it work, they're happy. And, they love to see the law crack down on all the dangerous pirates and hackers and similar scary people.
-Rob
OK, I stand corrected. All I really know is that the bills I get for my cable modem service come from "AT&T @Home", and that in this area, getting a cable modem means getting it from those folks. I was under the impression that cable modems everywhere were @Home, and I made the incorrect conclusion that since my @Home is AT&T, everyone's was.
-Rob
You can say the same about Education, Healthcare, The War on Drugs ("Drugs are BAD!"), Welfare, or any other program that the republicats push.
Very true. The trick is to figure out which ones are worth supporting, how much they're supporting, the degree to which government support of them mucks things up compared to private sector support of them, the degree to which private sector support fails.
Perhaps I'm painting myself as "part of the mindless mob" in saying this; a true libertarian feels that government should do nothing but mutual protection. I guess I'm not a true libertarian.
Yes, parents who value education will, under a completely free-market system, ensure that their kids get good education. Even if they are poor, I suspect that systems would evolve that would allow them to get a decent education. But it is my opinion that we *all* benefit the more that *everybody* is educated-- even the people whose parents won't pony up. I don't see it as strictly an entitlement; it's also "enlightened self interest."
-Rob
Remember, @Home is owned by AT&T, those fun people who were talking about charging online merchants a fraction of any sale to a AT&T Broadband customer, and even for just "delivering" those customers to the merchant in the first place.
I believe that AT&T/@Home has a vritual monopoly in the US on the cable modem lines. (My cable modem is certinaly AT&T/@Home, and it makes me think I should be posting this anonymously... oh well.) If it weren't for DSL, it would be a virtual broadband monopoly. Is there a lesson in here somewhere? A lesson that's been learned over and over again in history?
If we had other cable companies to get our cable modem service from, we wouldn't have to endlessly bitch about @Home. From the other end, @Home would have to clean up their act to keep their customers. I know I've swicthed phone-in ISPs several times, and have found one which has service and capabilities that really match what I want. If we can't vote with our feet, but can only bitch against people who have weird "intellectual property" laws to stifle that bitching, we're hosed.
-Rob
Before I say anything, I should note that I'm a scientist on the public dole, so I'm biased.
That being said, even though I have some libertarian tendencies I think that Browne's plan to end all government scientific funding is foolish. The reason: basic research is one of the best investments you can make. It is almost guaranteed to pay off. The problem is, you will invest in 1,000 research programs, and only have one program pay off. That one program will pay off to more than make up the investment for the other thousand-- and you will not be able to predict which one it was back at the beginning of the research program.
Some corporate funding of research has worked well in the past (Bell Labs?), but it just doesn't seem to be feasable today. Investing in basic scientific research is just too long term for most corporations. Never mind "five year plans" or even retirement times for top executives, you may not be able to fund enough projects to have any statistical confidence that any but the most applied of research programs may pay off for you. And the payoffs may be something unexpected, which you will have trouble reaping the benefits of anyway.
Scientific research is one of those things where everybody benefits (even if they don't realize it), and it is in everyone's interest to pool their resources to fund. But how to manage that? Well, isn't that what government is? My libertarian tendencies show themselves when I think that most people talk about government in the wrong way nowadays. The "we are your children" incident from one of the Clinton debates, which wasn't disupted by any of the candidates present, was a bad sign. To many of us see government as our parents, our protectors, those people who have control over us. They are our benign keepers. Yeah, they listen to us, and via voting we get to have some input into what we want done, but in the end many people in the USA see government as a particularly nice Big Brother.
Really, it should just be our way of acting collectively. The government should *be* us. It should be the way that we, as a society, perform the things that can only be done on a whole-societal level. My differences with libertarians come in as to what some of those things are. Scientific research is definitely one. Support of the arts is another-- rich individuals, and governments, are traditionally patrons of the arts. The arts have (mostly intangible) cultural value, but (with some very obvious exceptions) not much commercial value. Do we really want to let this part of our humanity go? Or is it worth some very small fraciton of our collective resources to support this endeavor? (When I say very small fraction, just compare arts funding in any government to defense, infrastructure, and sundry entitlements.)
-Rob
The one question I have: won't this make it entirely possible and legal for somebody to pull what Microsoft tried with their "J++"? Microsoft got slapped for that. But, suppose that Suns JVM was GPL'ed; that would make it difficult for them to license the spec and slap Microsoft the way they did for J++. Then suppose that Microsoft releases a J++, also GPL'ed to keep it legal, but so tied into the Windows API that it will effecitvely only ever work on Windows. Microsoft then pushes it, hard, making IE only work with J++, trying to get all Windows users out there to move to this polluted version of Java.
Cross platform Java dies. Sure, if J++ is open sourced, people can eventually create emulators that will let it run elsewhere, but I'd much rather have a Java that was designed from the beginning to be cross platform, wouldn't you?
Mind you, I know this is a Chicken Little scenario, but it *would* have been possible of JVM was open sourced all along, I believe. Nowadays, Microsoft is going to try to push C# to replace Java, so Sun is opening itself up less to people like MS by fully open sourcing Java.
--Rob
With luck, the fact that it's so bloody easy for everybody to hack thier device will teach them a lesson or two about paying really stupid, boneheaded patents.
Wouldn't it be beautiful if every company who both foisted and bought into stupid patents withered and died? Evolution in action. It would be nice. Amazon sitting in the graveyard next to Apple, with tourguides giving cautionary tales about stepping bast the bounds of common sense in the drive to abuse intellectual property laws.
Hey, a guy can dream, can't he?
-Rob
A GIMP expert has to choose between "giving away" improvements that make GIMP better by patching the UI, or writing a book and making a quick buck.
Each GIMP expert should play to his strengths. Only sufficiently skilled programmers are capable of contributing to the code base itself. A non-programmer who's figured out the GIMP and is effective at communication best contributes by writing documentation. Note that while the authors of "Grokking the Gimp" may well make a quick buck on it, they have in fact put the text of the book online-- so, even if they do make money, you can't deny it's a contribution to users of the program!
-Rob
This page has a link to several Gimp books:
http://www.xach.com/gimp/books/
I've got the "Gimp Essential Reference" myself, and find that it's by and large an excellent book. Before this book, I didn't really understand layers.... My one complaint is that the chapter on writing scripts and plugins is sketchy enough that it should be presented as more of an introduction. Additionally, with the intorudction of Gimp-Perl, I expect more people will want to use that than the Scheme language which is the default described in the book.
-RobI bet it has a lot more to do with how you're cocking your head, with the quality of the sound reception, etc., than it does with anything that the phone itself is emitting.
I mean, heck, I get a headache after talking on any phone for more than a little while. And I get a headache if the neighbors are playing their stereo too loud. Even though it's not painfully loud at my house, all I hear is a rhythmic "booming" noise, just at the edge of my hearing. Drives me nuts.
Lots of stimuli can cause irritation and headaches; it doesn't take some sort of boogey man "radiation" to do it.
The leading cause of cellphone injury is probably car accidents.
-RobHow is AT&T going to know what a "web retailer" is? I mean, sure, it's obvious for the big ones; you go to www.landsend.com, or www.amazon.com, or www.cdw.com, and AT&T's gonna know it's a retailer. But what of the million and one dinky stores out there that let you buy things over the web? How is AT&T going to tell the difference between a browser visiting a retailers site or a browser just browsing? Second, if I ran a small outfit that marketed over the web, and suddnely AT&T was going to start *charting* me for every customer that they somehow claim to have redirected my way, you had better believe I'd start blocking all AT&T addresses from my site. Hate to lose the sales, but hate to be charged by AT&T for every browser they want to take credit for. How could they even enforce this? I sure hope this is just more noise, because of anything comes of it, it's going to be very ugly. -Rob
Last year, two independent groups announced results from extremely distant supernovae that rule out the critical/no comsological constant
universe to very high confidence. Unless there is some as-yet undetected systematic problem with the data, it probably doesn't make sense to keep reporting age estimates based on that universe. See http://www-supernova.lbl.gov
-Rob (who also plays cosmologist at work)
NFS.
I had a huge headache at work a week and a half ago when everything died. I'm in a scientific group, and we've mostly used Solaris in the past. We've been ramping up our Linux usage, and on the whole Linux has been *more* stable than Solaris. However, we just recently started writing in bulk to a Linux disk NFS exported to a Solaris machine, and the nfs daemon *kept* *dying*. Very annoying.
I solved it with some Alan Cox patches that included H.J. Lu's latest knfsd. So Linux isn't as unreliable as it first looked. But there are a few places where Linux still does falter.
I think Thompson greatly overstates it, however. And, it is important qualifier on my NFS problems that there were patches out there I could apply that solved them.
-Rob
Linux is riding the anti-Microsoft trend, but only in it's current state of getting popular. Most of us longterm Linux geeks were using Linux before it suddenly became fashionable. Calling the current Linux Phenomenon nothing more than Microsoft backlash might be reasonable. Calling Linux itself nothing more than Microsoft backlash is dumb, and clearly ignores a lot of the even recent history of Linux.
-Rob