The disappointing thing about the article was that it didn't give any technical details on how they did it -- what settings they used on the laser cutter, power levels, etc. It's too bad, because these cutters aren't exactly hard to find (and relatively cheap if they're actually only $20k; that's less than a good vertical mill), if they had given the right info you could go to any machine shop that does a lot of prototyping and have them do it. Assuming it's not some sort of special feature of that particular model laser cutter that they're using.
But since they didn't give the settings for the cutter, I think you'd be taking a big risk at most shops since those machines are mostly used (in my experience) not for etching or engraving but for cutting, and you might end up with your Powerbook having a cool new form factor when you were done.
I had no idea what this article was all about, because I'd been logged into AIM continuously for the past day or so, and just now I reset the connection... lo and behold, there's a new group in my Gaim list: "AIM Bots."
It also popped up a message from the system, helpfully informing me that it had added these, and also telling me how to delete them (right-click, delete). Although why it didn't have something like "Type 1 to keep, type 2 to delete," I don't know. (Actually I suspect it's because a lot of people would have deleted them then, and won't now.)
There comes a point where you have to quit listening to theoretical arguments about how good a system could be, and face the cold, hard reality of what every implementation of it has been. Leftists will say that "x country isn't a good example of our system!" (where x is any country that has attempted Communism) As if somehow the people who tried to bring communism about in that country were somehow too stupid to do it right, or not properly motivated, or otherwise somehow defective. Naturally, that doesn't mean that they would fail if we decided to try it again now.
Communism doesn't scale -- period. It's been tried, over and over, and in every case it's either failed, or failing, or held together only by quasi-fascist dictatorships. And it's not as if the attempts have been half-hearted, either, or that anyone thought them doomed from the start; the Red takeover in Russia was hailed by American Communists as the beginning of a new utopia that would, once and for all, prove the superiority of their system. It didn't -- in fact it failed spectacularly. That alone should have been the last nail in that particular idea's coffin. But while success in Russia would have somehow meant to Leftists that Capitalism was doomed, they refuse to acknowledge the converse now that it's happened.
Instead we have people trying to drag the dead horse of Communism upright and make it walk around, time and time again, refusing to understand that the failures of the past might not have been because their predecessors in other countries were not ideogically pure enough, but because the idea is flawed from the very beginning.
Furthermore, while Communists in years past could have at least made a decent argument that their system was workable -- after all, it hadn't been tried -- and probably managed to convert many intelligent people to their side; today only a fool or an idiot would fall for their game, since thinking it can possibly succeed requires ignoring so much obvious evidence to the contrary. Thus the chances of success are far lower than in the past, since any halfway intelligent person can see that it's not sustainable.
While it would certainly be nice to put that idea to bed for once and for all, it has a certain seductive attraction to the lazy, the unmotivated, and the unproductive that is tough, if not impossible, to dispel.
Actually it's sounding more and more like it would be possible, if you read and look at the example further up in the thread with PostScript files. I think it still impractical though for most malicious users, because 1) the file size is so much bigger, meaning that the problem would be more computationally intensive, and 2) you're not trying to find a purely "weak" collision anymore (two arbitrary files with the same hash), instead you're trying to figure out what data you can pad your trojans with in order to make its hash equal the legit ISO's. I don't think the example code above will do that, at least not out of the box. I haven't examined it yet though.
This should be modded up. There is a rather large amount of focus put on the "root" nameservers in the U.S., somewhat disporportionally. If a country wants to do draconian things to its own internet, using its own root nameservers, it's more than free to do so. Nobody is forcing anyone else to stay connected or compatible, they are choosing to by their own volition because (one assumes) it benefits them to do so.
Or if you don't like U.S. policy on something, but don't want to unplug yourself from the rest of the network, you can use the space that's delegated to your own country: it's every nation's personal little sandbox, to build in or shit in, their choice. If showing 14-year-olds having sex with St. Bernards is legal in your country, you can serve it up from within your TLD; likewise if you want to censor everything your subjects can access, although I personally find it as detestable as the children-fucking-dogs business, that's where you can do it (and be shooting nobody but yourself in the foot by doing so).
If this guy actually got 161 governments to sit down and actually work together on something, I would be deeply, deeply concerned. Given that the average government official is probably in bed with so many different corporations and special interests (and it might not even be illegal or frowned upon in their countries, so it's not like we have much recourse) anything that they'd sit down and turn out is ultimately going to be terrible for users.
In my more morose moods I have this feeling that I'm going to some day be sitting around and telling my grandkids about how the Internet used to be, back in those wild, turn-of-the-century days, before everything was regulated and monitored to death; in the same way that I remember him telling me once about a time when you could buy a car and drive it around without a license to do so, or bolting a metal identification plate onto the bumper.
Governments are a sophisticated protection racket. You trade them some freedoms, in return they offer you some protection against our more cruel and brutish impulses and in theory allow us to live more pleasant lives. But with the Internet, there's currently nothing that we need protecting from and if we allow it to be regulated, we will have just given something away for nothing -- and it's not something we're ever likely to get back.
Although I don't necessarily agree with them per se (the folks who think that Linux should be the One True OS for Everyone), I'm not sure that you can convince me that the reverse of the current situation regarding Windows and Linux would be worse or even as bad as it is now. The inherently fragmented, decentralized nature of Linux wouldn't allow it to be leveraged by one company in the same way that Microsoft uses Windows. If some company decided to start charging a few hundred bucks a processor for their Linux-based OS, plus a few hundred more for "Linux Terminal Services," plus a few more for the client system's OS, it would only take one reasonably deep-pocketed client one time to say 'hell with this' and fork the code. Perhaps over time, some company could add enough clean-room-developed pieces of proprietary code (or one totally proprietary 'killer app') to get people locked in to their system, but I think it would take a new generation of users to not be able to see through that play.
What I do agree with is that rule by Apple might concievably be almost as bad as rule by Microsoft (and given that I'm writing this on a Power Mac which I'm very fond of, it pains me to say it), I'm not sure that Linux has enough of a strong governing body to exert what power it might gain if it ever became dominant.
That said, it's probably not going to be a situation we'll have to deal with, at least in the forseeably near future, so the whole argument does fall a bit deeply into the 'academic wanking' category.
Yeah the Blue Gene running Linux thing is a bit overstated. According to this article from Newsfactor, "only 1,000 of the nodes will be powered by Linux, while the rest will use specialized software to power the machine." I expect that the 'specialized software' is z/OS or something similar, or custom code specific to the application or problem being solved.
Fairly good question. I'm not sure where you start calling something a "computer" and where you fall off into the grey area of "computational network" or "cluster" or "grid computing system." After all, isn't SETI@Home a pretty massive computer? By some (very loose) definition it should be.
I think most people consider a computer to be something that, at some level, runs a single operating system (which then can abstract other OSes on top of itself), or perhaps is capable of addressing a single logical range of main memory (although this might not be a good definition either).
I haven't read the article yet to see if they give their definition, but it does seem as if the line between 'this is a computer' and 'this is a bunch of computers working together' is fairly blurry, and perhaps where one draws it is completely arbitrary.
Although I'm not sure that I would go so far as to call the math "wrong," since it looks to me as if it's factually correct at least in the narrowest mathematical sense, I do agree with you though that it represents an example which isn't representative of many car buyers. Obviously it's not representative of you (although you admit you're atypical), but it's not really typical of anyone, except in the short term while you're paying off the car.
Personally I think he made a mistake in messing around with car payments to begin with. What should have been compared is the apples-to-apples, upfront MSRP (or KBB invoice) prices of the different cars to start with, and then an estimate of the ownership costs per mile including fuel, maintainance estimates. I think that would have created a far more useful model for most people who are choosing between two different cars. Also, it would be pretty easy to add into that the effects of depreciation / resale value, since they're widely considered to be mileage-dependent. Also you could even add in a time-dependency into the fuel cost if you wanted to work in 3 dimensions and see how far up the gas price would have to go to make the hybrid pay off in the amount of time you normally keep cars for.
I have created graphs just like this to compare gasoline versus diesel cars (actually it was the VW Beetle vs. the Beetle TDI); and found them very helpful in figuring out how many miles I theoretically need to put on the car in order to make up the initial investment of the alternative engine. (IIRC it was around 100,000 miles, not counting the difference in depreciation and this was back when diesel was cheaper than gasoline, which is no longer true.)
I think he made things overly complicated by trying to compare the month-to-month payments, when he could have begun with a simple "cost of ownership versus mileage" graph and then added on additional factors as necessary.
In that case, wouldn't they be better from a marketing standpoint just to drop all the explanations of threads and cores and say "Sun introduces new 38.4GHz Chip! So bloody fast it'll make your head explode!"
Good point, but just think, if you argued the price of the hybrid down to within $1400 of the normal Civic price (by which I assume you mean MSRP) then just think of how hard you could have chiseled them on the standard one. There is a lot of demand for hybrid models and limited supply -- thus they cost more. I've heard that the Priuses are going for several hundred dollars above MSRP right now because of demand: the dealers add a "market price adjustment" onto the bottom line, which is all profit for them.
You say that nobody pays MSRP anymore, but then you compare the final price you paid after negotiation to the MSRP of a different model to show how much of a good deal yours was. That doesn't make a lot of sense. To be fair you'd have to somehow come up with what the price after negotiation of the standard Civic would have been.
Personally I really like hybrids but I've run the numbers myself as well, using my own mileage amounts and local gas prices, interest rates, etc., and they don't make sense quite yet. There are certain situations where they might (all city driving, where they get 300% better mileage than some other cars) but for an interstate commuter there's still too much of a premium being charged for the technology.
It's still an "early adopter" technology. Lots of people buy things that aren't ready for the mainstream market, just because they can -- and that's not something that should be discounted.
The tables look suspiciously like they could be Excel. Maybe they're OpenOffice or some stats program the output of which I've not seen yet, but they're well within the range of what you can get out of Excel. What makes them unusually nice is that he actually used a decent, not-too-obtrusive color scheme which actually enhances the legibility of the material, instead of obscuring it (which is what most people do) by using some obnoxious combination of contrasting colors. At any rate, it's just a nicely-done spreadsheet.
The graphs are fairly nice though, and don't say Excel so much. I think they could be GNUplot or some derivative product...?
But I agree, the citations are nice. A step above the usual vaporware press releases or hype journalism, anyway.
Today, military insignificant
on
A Flu Pandemic?
·
· Score: 1
I know you were being humorous, but it's worth pointing out that the number of troops in Iraq is so small as to be insignificant compared to the population moving around as a result of World War One. Both in terms of sheer numbers (there were battles in WWI that involved more people than the total number of combat troops in Iraq, leading me to wonder if we've relaxed the criteria for a "war" recently, but that's a different issue) and of the percentage of the global travelling population at the time. In 1918, if half a million people went from one place to another it was a huge event; today it's routine.
The problem is really the civilian travellers, because they're much harder to control and track, short of just shutting down the air routes completely and freezing every one in place. At least with the military you can pretty much verify where John Doe came from and was going to and who he might have been with -- good luck doing that with some random passenger off of a 747 going from Tokyo to LAX.
Yes, but you have to ask yourself, why did it kill so many people in 1918? People had been having the "achy go to bed for a few days" flu for generations then as well, and it didn't help them when the new version of the virus started to spread.
The problem with influenza is that it mutates like crazy -- what other things do you need to get a new vaccination for every year? Most of the time the mutations are small and relatively insignificant, so that if you contract it your body isn't too far behind in catching up and beating it, but every once in a while you get a really significant mutation and a lot of people are SOL.
The reason people are concerned now, is not because they think the 1918 virus strain will come back (it's probably still around, somewhere) because if it did you're right, most of us would probably be just fine, but that some new strain will come out which will hit us in the same way that people in 1918 got hit by their strain. The immunity we developed in past generations stops helping us after a certain period, and the question is whether we're at that point yet.
Just thought I'd toss out some information on Ebola Reston, since it's fairly relevant here and something I've read up on. It actually was not limited to Reston, VA -- there were outbreaks in imported primates in Texas and Pennsylvania as well.
But EBOR is some scary stuff: all the fun of Ebola, but airborne; its saving grace was that it was seems to be restricted to monkeys. The evidence for this is that monkeys from one shipment were held in one room, connected to other rooms only by a shared air supply, and eventually other shipments' monkeys started dying. It infects humans -- several people who were exposed had seroconverted -- but is asymptomatic.
The thing which has always kept me intrigued about Ebola is that we honestly don't know what its natural reservoir is, or where the different strains have come from. There are several different human strains with different lethalities, plus Reston which is airborne but only kills monkeys (and came from the Philippines), and at the end of the day nobody really has a good idea of what the common thread is. Other than that it's probably something in the jungle in Africa, and almost certainly something which doesn't experience any of the symptoms of Ebola that primates or humans do.
I don't have the background to say whether Ebola is more or less of a risk of starting an epidemic than influenza (or Soviet weaponized smallpox, or SARS, any other disease-of-the-week), but I certainly think it's something that we should try to keep not too far back from the front of our collective consiousness, if for no other reason than as a reminder that we haven't figured it all out yet.
Personally the best part of the whole Ebola Reston story: According to Wikipedia, "After the sanitation of the monkey house [in Reston, VA] it was turned into a daycare center."
Mod parent up -- although I didn't run the numbers to check, I agree that it seems far more likely that there was a series of false-positive tests originally, especially given that the person is a gay man in a relationship with an infected individual, pretty much your stereotypical AIDS patient. It doesn't take too much imagination to think that perhaps they ran the first antibody test, it came back as a false positive, they ran another one, it came back false positive too (if each has a 10% FP rate than that's only a 1:100 chance, not wholly insignificant) and then just went "yep, you have HIV, thanks for playing."
Although I agree more tests definitely need to be done (and if this guy doesn't want to work with the same clinic that possibly fucked up the original test, fine -- go to a different hospital or clinic, that'll make the results more believable anyway), the odds of it being a bunch of false positives in a row, helped out by the 'confirmation bias' as a result of him being the typical case demographic, seem much higher than the chances he was actually thoroughly infected with the virus and managed to cure himself.
Agreed. I didn't really even understand that point of the article; it's not like setting up a VPN is that hard these days, even a fairly dumb/understaffed/"spread-thin" IT department could put together a VPN if that was a major security concern. After all, banning Skype is only going to do you any good until the next little application that has NAT traversal comes along (and there are an increasing number that do, especially communications apps) and then you're hosed again.
This whole article is just FUD, it reads like the intended audience is ignorant middle and upper-level managers and the intent is to instill some sort of primitive idea that "Skype = bad" without any convincing explanations why. The end result I think is that a lot of system and network admins are going to be fielding a lot of questions as to "shouldn't we ban that 'skipe' thing? I've heard it could wreck our network," and have to sit down and explain why that's not true.
I'm a Mac user, and also a fan of Skype -- they do something that iChat can't touch. In fact, I would say that they nailed the "Mac-like" user experience of voice chat, far better than iChat does itself. Possibly for no fault of its own (owed mostly to compatibility problems with AIM, I think) I've never gotten the voice features of iChat to work between me and a PC user. I've gotten video to work -- poorly -- but never audio. (Between two Macs it's great, but add an AIM client into the mix...ugh.) I've never felt mascohistic enough to try with a Linux machine, if any clients exist.
By not coupling itself onto the AIM network or any other existing one, you can be sure that if you see a person in your Skype buddy list, that you can chat with them. End of story. Also, the integration with landline telephones (SkypeOut) makes it very attractive from an economic perspective and is a big reason why many people who otherwise wouldn't ever think of their computers as voice devices have installed the software.
I'm a big fan of all things Mac, but sometimes you just have to sit back and realize that occasionally, people really hit the nail on the head and do something way better than every body else, including Apple. Skype just cleans the floor with every other voice chat program out there; you can go from a bare computer to talking with someone on a landline on the other side of the world (at $0.01/min or so) in five minutes with nothing additional but a $10 headset mic and a credit card. iChat has some great features, including Address Book integration that's just wonderful, and it's a pity Skype doesn't build them into the OS X version, but for most people there's no comparison.
And people accuse the U.S. of being creepy and censorious? Glad to hear that the field is being pioneered elsewhere.
Granted, I haven't been to a highschool in this country in a while, but I have it on good authority that there is not nearly that level of filtering or access control. I think most filtering here in schools is aimed at curbing file sharing, and there isn't any content filtration in any libraries that I'm aware of.
How about rsync? I'm no command-line genius, and I managed to set up my system to back up my home directory to a remote system every night, took me less than an hour to set up and that included time to read the documentation.
The remote system doesn't even need to be Ubuntu, either. It could be any Linux flavor, probably some commercial Unixes, even Mac OS X / Darwin (which is what I use). Not sure if Windows has it -- I know it's not there by default, but I suspect there's probably something one could download -- and I bet if you check around there's probably a GUI frontend for it on Linux, if you don't like the command line.
It's essentially the same functionality that a commercial product (Connected DataProtector) used by my company on our business machines provides, but without having to pay what I assume is big bucks for licenses and dedicate a backup server. At least for me it works really well.
I think your other suggestions are well taken, but the pre-installed backup capability offered by rsync on basically every Linux distro (and OS X / Darwin) is one of the reasons I think it's a stronger desktop OS than a default Windows system. With a few scripts and the help of cron, I can make all the computers in the house backup and syncronize data to a server, and then have that server sync with another server, all 'automagically.' And if I can do this, with some good documentation and maybe a GUI frontend, it shouldn't be out of the reach of the typical user who actually cares about having a backup strategy.
Yes I read the same thing. I think it was in an interview/FAQ with Shuttleworth that was linked from the Ubuntu site a few weeks ago. Or maybe it was in a Slashdot article (some days they are substantially similar).
Anyway, it's discussed on his page in the Ubuntu Wiki:
Our current plan is that the Dapper Drake (Ubuntu 6.04 if we hit our April 2006 release date goal) will be the last of this first "set" of releases. So post-Dapper we have the opportunity to define a new "feel" or overarching theme. It would be unlikely to be... blue. But it might be substantially different to the current Human theme. For the moment, let's stay focused on the road to Dapper, polish up the existing Human theme to the max for that, and then break new ground post-Dapper.
You're right, there is plenty of skilled US citizens that work for $10/hr.
There ARE plenty of (reasonably) skilled US citizens, in the form of college students, willing to work for $10 an hour. I'm sure you wouldn't have any trouble finding people to work the scanner at that rate at any large university. Especially if you offered flexible/non-daytime hours: the most popular campus jobs in my experience were always the ones that you could work in the evening or at night.
In fact I'm a little disappointed that they're only runnning what sounds like two shifts in Toronto -- I'm sure that the bottleneck right now is more machine time than operator time; if they doubled their operators they could keep them going 20 hours a day (since it sounds like right now they're doing 10 hrs/day in two 5-hr shifts). And like the supervisor guy said in the article, there's no shortage of content.
Slashdot Mirror
The disappointing thing about the article was that it didn't give any technical details on how they did it -- what settings they used on the laser cutter, power levels, etc. It's too bad, because these cutters aren't exactly hard to find (and relatively cheap if they're actually only $20k; that's less than a good vertical mill), if they had given the right info you could go to any machine shop that does a lot of prototyping and have them do it. Assuming it's not some sort of special feature of that particular model laser cutter that they're using.
But since they didn't give the settings for the cutter, I think you'd be taking a big risk at most shops since those machines are mostly used (in my experience) not for etching or engraving but for cutting, and you might end up with your Powerbook having a cool new form factor when you were done.
I had no idea what this article was all about, because I'd been logged into AIM continuously for the past day or so, and just now I reset the connection ... lo and behold, there's a new group in my Gaim list: "AIM Bots."
It also popped up a message from the system, helpfully informing me that it had added these, and also telling me how to delete them (right-click, delete). Although why it didn't have something like "Type 1 to keep, type 2 to delete," I don't know. (Actually I suspect it's because a lot of people would have deleted them then, and won't now.)
Well said. You should get modded up.
There comes a point where you have to quit listening to theoretical arguments about how good a system could be, and face the cold, hard reality of what every implementation of it has been. Leftists will say that "x country isn't a good example of our system!" (where x is any country that has attempted Communism) As if somehow the people who tried to bring communism about in that country were somehow too stupid to do it right, or not properly motivated, or otherwise somehow defective. Naturally, that doesn't mean that they would fail if we decided to try it again now.
Communism doesn't scale -- period. It's been tried, over and over, and in every case it's either failed, or failing, or held together only by quasi-fascist dictatorships. And it's not as if the attempts have been half-hearted, either, or that anyone thought them doomed from the start; the Red takeover in Russia was hailed by American Communists as the beginning of a new utopia that would, once and for all, prove the superiority of their system. It didn't -- in fact it failed spectacularly. That alone should have been the last nail in that particular idea's coffin. But while success in Russia would have somehow meant to Leftists that Capitalism was doomed, they refuse to acknowledge the converse now that it's happened.
Instead we have people trying to drag the dead horse of Communism upright and make it walk around, time and time again, refusing to understand that the failures of the past might not have been because their predecessors in other countries were not ideogically pure enough, but because the idea is flawed from the very beginning.
Furthermore, while Communists in years past could have at least made a decent argument that their system was workable -- after all, it hadn't been tried -- and probably managed to convert many intelligent people to their side; today only a fool or an idiot would fall for their game, since thinking it can possibly succeed requires ignoring so much obvious evidence to the contrary. Thus the chances of success are far lower than in the past, since any halfway intelligent person can see that it's not sustainable.
While it would certainly be nice to put that idea to bed for once and for all, it has a certain seductive attraction to the lazy, the unmotivated, and the unproductive that is tough, if not impossible, to dispel.
Actually it's sounding more and more like it would be possible, if you read and look at the example further up in the thread with PostScript files. I think it still impractical though for most malicious users, because 1) the file size is so much bigger, meaning that the problem would be more computationally intensive, and 2) you're not trying to find a purely "weak" collision anymore (two arbitrary files with the same hash), instead you're trying to figure out what data you can pad your trojans with in order to make its hash equal the legit ISO's. I don't think the example code above will do that, at least not out of the box. I haven't examined it yet though.
I've seen a lot of funny posts get moderated as "Insightful," even though they obviously weren't, just so that the poster would get some karma.
I think this is the first case when I've seen a truly Insightful post get moderated "Funny."
This should be modded up. There is a rather large amount of focus put on the "root" nameservers in the U.S., somewhat disporportionally. If a country wants to do draconian things to its own internet, using its own root nameservers, it's more than free to do so. Nobody is forcing anyone else to stay connected or compatible, they are choosing to by their own volition because (one assumes) it benefits them to do so.
Or if you don't like U.S. policy on something, but don't want to unplug yourself from the rest of the network, you can use the space that's delegated to your own country: it's every nation's personal little sandbox, to build in or shit in, their choice. If showing 14-year-olds having sex with St. Bernards is legal in your country, you can serve it up from within your TLD; likewise if you want to censor everything your subjects can access, although I personally find it as detestable as the children-fucking-dogs business, that's where you can do it (and be shooting nobody but yourself in the foot by doing so).
Exactly.
If this guy actually got 161 governments to sit down and actually work together on something, I would be deeply, deeply concerned. Given that the average government official is probably in bed with so many different corporations and special interests (and it might not even be illegal or frowned upon in their countries, so it's not like we have much recourse) anything that they'd sit down and turn out is ultimately going to be terrible for users.
In my more morose moods I have this feeling that I'm going to some day be sitting around and telling my grandkids about how the Internet used to be, back in those wild, turn-of-the-century days, before everything was regulated and monitored to death; in the same way that I remember him telling me once about a time when you could buy a car and drive it around without a license to do so, or bolting a metal identification plate onto the bumper.
Governments are a sophisticated protection racket. You trade them some freedoms, in return they offer you some protection against our more cruel and brutish impulses and in theory allow us to live more pleasant lives. But with the Internet, there's currently nothing that we need protecting from and if we allow it to be regulated, we will have just given something away for nothing -- and it's not something we're ever likely to get back.
Although I don't necessarily agree with them per se (the folks who think that Linux should be the One True OS for Everyone), I'm not sure that you can convince me that the reverse of the current situation regarding Windows and Linux would be worse or even as bad as it is now. The inherently fragmented, decentralized nature of Linux wouldn't allow it to be leveraged by one company in the same way that Microsoft uses Windows. If some company decided to start charging a few hundred bucks a processor for their Linux-based OS, plus a few hundred more for "Linux Terminal Services," plus a few more for the client system's OS, it would only take one reasonably deep-pocketed client one time to say 'hell with this' and fork the code. Perhaps over time, some company could add enough clean-room-developed pieces of proprietary code (or one totally proprietary 'killer app') to get people locked in to their system, but I think it would take a new generation of users to not be able to see through that play.
What I do agree with is that rule by Apple might concievably be almost as bad as rule by Microsoft (and given that I'm writing this on a Power Mac which I'm very fond of, it pains me to say it), I'm not sure that Linux has enough of a strong governing body to exert what power it might gain if it ever became dominant.
That said, it's probably not going to be a situation we'll have to deal with, at least in the forseeably near future, so the whole argument does fall a bit deeply into the 'academic wanking' category.
Yeah the Blue Gene running Linux thing is a bit overstated. According to this article from Newsfactor, "only 1,000 of the nodes will be powered by Linux, while the rest will use specialized software to power the machine." I expect that the 'specialized software' is z/OS or something similar, or custom code specific to the application or problem being solved.
Fairly good question. I'm not sure where you start calling something a "computer" and where you fall off into the grey area of "computational network" or "cluster" or "grid computing system." After all, isn't SETI@Home a pretty massive computer? By some (very loose) definition it should be.
I think most people consider a computer to be something that, at some level, runs a single operating system (which then can abstract other OSes on top of itself), or perhaps is capable of addressing a single logical range of main memory (although this might not be a good definition either).
I haven't read the article yet to see if they give their definition, but it does seem as if the line between 'this is a computer' and 'this is a bunch of computers working together' is fairly blurry, and perhaps where one draws it is completely arbitrary.
Although I'm not sure that I would go so far as to call the math "wrong," since it looks to me as if it's factually correct at least in the narrowest mathematical sense, I do agree with you though that it represents an example which isn't representative of many car buyers. Obviously it's not representative of you (although you admit you're atypical), but it's not really typical of anyone, except in the short term while you're paying off the car.
Personally I think he made a mistake in messing around with car payments to begin with. What should have been compared is the apples-to-apples, upfront MSRP (or KBB invoice) prices of the different cars to start with, and then an estimate of the ownership costs per mile including fuel, maintainance estimates. I think that would have created a far more useful model for most people who are choosing between two different cars. Also, it would be pretty easy to add into that the effects of depreciation / resale value, since they're widely considered to be mileage-dependent. Also you could even add in a time-dependency into the fuel cost if you wanted to work in 3 dimensions and see how far up the gas price would have to go to make the hybrid pay off in the amount of time you normally keep cars for.
I have created graphs just like this to compare gasoline versus diesel cars (actually it was the VW Beetle vs. the Beetle TDI); and found them very helpful in figuring out how many miles I theoretically need to put on the car in order to make up the initial investment of the alternative engine. (IIRC it was around 100,000 miles, not counting the difference in depreciation and this was back when diesel was cheaper than gasoline, which is no longer true.)
I think he made things overly complicated by trying to compare the month-to-month payments, when he could have begun with a simple "cost of ownership versus mileage" graph and then added on additional factors as necessary.
In that case, wouldn't they be better from a marketing standpoint just to drop all the explanations of threads and cores and say "Sun introduces new 38.4GHz Chip! So bloody fast it'll make your head explode!"
Good point, but just think, if you argued the price of the hybrid down to within $1400 of the normal Civic price (by which I assume you mean MSRP) then just think of how hard you could have chiseled them on the standard one. There is a lot of demand for hybrid models and limited supply -- thus they cost more. I've heard that the Priuses are going for several hundred dollars above MSRP right now because of demand: the dealers add a "market price adjustment" onto the bottom line, which is all profit for them.
You say that nobody pays MSRP anymore, but then you compare the final price you paid after negotiation to the MSRP of a different model to show how much of a good deal yours was. That doesn't make a lot of sense. To be fair you'd have to somehow come up with what the price after negotiation of the standard Civic would have been.
Personally I really like hybrids but I've run the numbers myself as well, using my own mileage amounts and local gas prices, interest rates, etc., and they don't make sense quite yet. There are certain situations where they might (all city driving, where they get 300% better mileage than some other cars) but for an interstate commuter there's still too much of a premium being charged for the technology.
It's still an "early adopter" technology. Lots of people buy things that aren't ready for the mainstream market, just because they can -- and that's not something that should be discounted.
The tables look suspiciously like they could be Excel. Maybe they're OpenOffice or some stats program the output of which I've not seen yet, but they're well within the range of what you can get out of Excel. What makes them unusually nice is that he actually used a decent, not-too-obtrusive color scheme which actually enhances the legibility of the material, instead of obscuring it (which is what most people do) by using some obnoxious combination of contrasting colors. At any rate, it's just a nicely-done spreadsheet.
The graphs are fairly nice though, and don't say Excel so much. I think they could be GNUplot or some derivative product...?
But I agree, the citations are nice. A step above the usual vaporware press releases or hype journalism, anyway.
I know you were being humorous, but it's worth pointing out that the number of troops in Iraq is so small as to be insignificant compared to the population moving around as a result of World War One. Both in terms of sheer numbers (there were battles in WWI that involved more people than the total number of combat troops in Iraq, leading me to wonder if we've relaxed the criteria for a "war" recently, but that's a different issue) and of the percentage of the global travelling population at the time. In 1918, if half a million people went from one place to another it was a huge event; today it's routine.
The problem is really the civilian travellers, because they're much harder to control and track, short of just shutting down the air routes completely and freezing every one in place. At least with the military you can pretty much verify where John Doe came from and was going to and who he might have been with -- good luck doing that with some random passenger off of a 747 going from Tokyo to LAX.
Yes, but you have to ask yourself, why did it kill so many people in 1918? People had been having the "achy go to bed for a few days" flu for generations then as well, and it didn't help them when the new version of the virus started to spread.
The problem with influenza is that it mutates like crazy -- what other things do you need to get a new vaccination for every year? Most of the time the mutations are small and relatively insignificant, so that if you contract it your body isn't too far behind in catching up and beating it, but every once in a while you get a really significant mutation and a lot of people are SOL.
The reason people are concerned now, is not because they think the 1918 virus strain will come back (it's probably still around, somewhere) because if it did you're right, most of us would probably be just fine, but that some new strain will come out which will hit us in the same way that people in 1918 got hit by their strain. The immunity we developed in past generations stops helping us after a certain period, and the question is whether we're at that point yet.
Yeah I was kinda wondering about that, too...
I guess everyone has their personal end-of-the-world-as-we-know-it theory.
Just thought I'd toss out some information on Ebola Reston, since it's fairly relevant here and something I've read up on. It actually was not limited to Reston, VA -- there were outbreaks in imported primates in Texas and Pennsylvania as well.
But EBOR is some scary stuff: all the fun of Ebola, but airborne; its saving grace was that it was seems to be restricted to monkeys. The evidence for this is that monkeys from one shipment were held in one room, connected to other rooms only by a shared air supply, and eventually other shipments' monkeys started dying. It infects humans -- several people who were exposed had seroconverted -- but is asymptomatic.
The thing which has always kept me intrigued about Ebola is that we honestly don't know what its natural reservoir is, or where the different strains have come from. There are several different human strains with different lethalities, plus Reston which is airborne but only kills monkeys (and came from the Philippines), and at the end of the day nobody really has a good idea of what the common thread is. Other than that it's probably something in the jungle in Africa, and almost certainly something which doesn't experience any of the symptoms of Ebola that primates or humans do.
I don't have the background to say whether Ebola is more or less of a risk of starting an epidemic than influenza (or Soviet weaponized smallpox, or SARS, any other disease-of-the-week), but I certainly think it's something that we should try to keep not too far back from the front of our collective consiousness, if for no other reason than as a reminder that we haven't figured it all out yet.
Personally the best part of the whole Ebola Reston story: According to Wikipedia, "After the sanitation of the monkey house [in Reston, VA] it was turned into a daycare center."
Mod parent up -- although I didn't run the numbers to check, I agree that it seems far more likely that there was a series of false-positive tests originally, especially given that the person is a gay man in a relationship with an infected individual, pretty much your stereotypical AIDS patient. It doesn't take too much imagination to think that perhaps they ran the first antibody test, it came back as a false positive, they ran another one, it came back false positive too (if each has a 10% FP rate than that's only a 1:100 chance, not wholly insignificant) and then just went "yep, you have HIV, thanks for playing."
Although I agree more tests definitely need to be done (and if this guy doesn't want to work with the same clinic that possibly fucked up the original test, fine -- go to a different hospital or clinic, that'll make the results more believable anyway), the odds of it being a bunch of false positives in a row, helped out by the 'confirmation bias' as a result of him being the typical case demographic, seem much higher than the chances he was actually thoroughly infected with the virus and managed to cure himself.
Agreed. I didn't really even understand that point of the article; it's not like setting up a VPN is that hard these days, even a fairly dumb/understaffed/"spread-thin" IT department could put together a VPN if that was a major security concern. After all, banning Skype is only going to do you any good until the next little application that has NAT traversal comes along (and there are an increasing number that do, especially communications apps) and then you're hosed again.
This whole article is just FUD, it reads like the intended audience is ignorant middle and upper-level managers and the intent is to instill some sort of primitive idea that "Skype = bad" without any convincing explanations why. The end result I think is that a lot of system and network admins are going to be fielding a lot of questions as to "shouldn't we ban that 'skipe' thing? I've heard it could wreck our network," and have to sit down and explain why that's not true.
I'm a Mac user, and also a fan of Skype -- they do something that iChat can't touch. In fact, I would say that they nailed the "Mac-like" user experience of voice chat, far better than iChat does itself. Possibly for no fault of its own (owed mostly to compatibility problems with AIM, I think) I've never gotten the voice features of iChat to work between me and a PC user. I've gotten video to work -- poorly -- but never audio. (Between two Macs it's great, but add an AIM client into the mix...ugh.) I've never felt mascohistic enough to try with a Linux machine, if any clients exist.
By not coupling itself onto the AIM network or any other existing one, you can be sure that if you see a person in your Skype buddy list, that you can chat with them. End of story. Also, the integration with landline telephones (SkypeOut) makes it very attractive from an economic perspective and is a big reason why many people who otherwise wouldn't ever think of their computers as voice devices have installed the software.
I'm a big fan of all things Mac, but sometimes you just have to sit back and realize that occasionally, people really hit the nail on the head and do something way better than every body else, including Apple. Skype just cleans the floor with every other voice chat program out there; you can go from a bare computer to talking with someone on a landline on the other side of the world (at $0.01/min or so) in five minutes with nothing additional but a $10 headset mic and a credit card. iChat has some great features, including Address Book integration that's just wonderful, and it's a pity Skype doesn't build them into the OS X version, but for most people there's no comparison.
Wow.
And people accuse the U.S. of being creepy and censorious? Glad to hear that the field is being pioneered elsewhere.
Granted, I haven't been to a highschool in this country in a while, but I have it on good authority that there is not nearly that level of filtering or access control. I think most filtering here in schools is aimed at curbing file sharing, and there isn't any content filtration in any libraries that I'm aware of.
A good backup application
How about rsync? I'm no command-line genius, and I managed to set up my system to back up my home directory to a remote system every night, took me less than an hour to set up and that included time to read the documentation.
The remote system doesn't even need to be Ubuntu, either. It could be any Linux flavor, probably some commercial Unixes, even Mac OS X / Darwin (which is what I use). Not sure if Windows has it -- I know it's not there by default, but I suspect there's probably something one could download -- and I bet if you check around there's probably a GUI frontend for it on Linux, if you don't like the command line.
It's essentially the same functionality that a commercial product (Connected DataProtector) used by my company on our business machines provides, but without having to pay what I assume is big bucks for licenses and dedicate a backup server. At least for me it works really well.
I think your other suggestions are well taken, but the pre-installed backup capability offered by rsync on basically every Linux distro (and OS X / Darwin) is one of the reasons I think it's a stronger desktop OS than a default Windows system. With a few scripts and the help of cron, I can make all the computers in the house backup and syncronize data to a server, and then have that server sync with another server, all 'automagically.' And if I can do this, with some good documentation and maybe a GUI frontend, it shouldn't be out of the reach of the typical user who actually cares about having a backup strategy.
Anyway, it's discussed on his page in the Ubuntu Wiki:
So Dapper = last "brown" release.
You're right, there is plenty of skilled US citizens that work for $10/hr.
There ARE plenty of (reasonably) skilled US citizens, in the form of college students, willing to work for $10 an hour. I'm sure you wouldn't have any trouble finding people to work the scanner at that rate at any large university. Especially if you offered flexible/non-daytime hours: the most popular campus jobs in my experience were always the ones that you could work in the evening or at night.
In fact I'm a little disappointed that they're only runnning what sounds like two shifts in Toronto -- I'm sure that the bottleneck right now is more machine time than operator time; if they doubled their operators they could keep them going 20 hours a day (since it sounds like right now they're doing 10 hrs/day in two 5-hr shifts). And like the supervisor guy said in the article, there's no shortage of content.