> using something called MediaSite to view quarterly state library "webinars"
I should probably explain this better. A MediaSite webinar is similar to a YouTube video, with a few differences, which I will attempt to list. First, it's scheduled for a certain time and announced ahead of time, and you can visit the page ahead of time, but you can't watch it until the scheduled time. Second, the interface is significantly less user-friendly and as near as I can tell does not feature rewind or fast forward, though I could be missing something. Also, you can't watch it through a proxy, even a transparent proxy such as is commonly used for content filtering. You have to have Silverlight installed to view it. And instead of amusement, the primary intended purpose is pedagogical.
And yeah, the state library ought to know better. Why they can't just throw an MPEG up on a web server and link to it, I'll never know. They certainly have the bandwidth for that, especially since almost all of the viewers would be watching from within libraries that are connected to the internet via OPLIN, so the pipes to the outside world wouldn't even be involved in the overwhelming majority of cases. The T1 lines to each individual user retrieving the thing would be the bottleneck.
> The only time I've ever used Silverlight is when I've watched > SkyTV online in the UK as a media thingy for your browser.
Okay, that's the *third* use for Silverlight that I've heard of. The other two (playing around with the demo site Microsoft set up to advertise Silverlight, and using something called MediaSite to view quarterly state library "webinars" on the current state of library funding) didn't interest me either. So yeah, Silverlight is a solution in search of a problem. Is there anyone outside of Redmond who didn't know that?
Furthermore, even if Silverlight were awesomely useful and made whole new kinds of incredibly cool sites possible that were barely conceivable before, Moonlight would still be an astonishingly bad idea for the open-source community.
I mean, yeah, instead of protecting the open platform-neutral internet, let's actively help Microsoft decommoditize and proprietize it and make it part of the Windows-centric integrated Microsoft product offering stack, so that web developers all end up using Visual Studio and doing everything with functions provided by IIS and SQL Server and DirectX. That'll sure improve things for everyone.</sarcasm>
If you didn't realize before that Moonlight was a bad idea, think about the fact that Microsoft is making this promise not to sue, and what that implies. I don't mean Microsoft's motivation in making the promise, because that's debatable. No, I'm talking about the fact that anyone would even *wonder* whether they might sue Moonlight users, the fact that the promise is even necessary or meaningful, the fact that they would even think to make it. That is in itself a dead giveaway that relying on Moonlight for anything that matters is playing with fire. Frankly, if Microsoft released a closed-source Silverlight plugin for Linux systems, that would scare me less.
Actually, I consider the whole Mono project to be an inherently brain-damaged idea. I absolutely do not want Mono installed on any computer I use or administer. Ever.
> Of course, we lived with a couple of decades of terrorist > attacks ourselves long before Terrorism was the new bogeyman.
The UK only *thinks* they have terrorism problems. (The US too, for that matter.)
Spend a few months living in Israel, and you'll find out how it really is.
I'm not saying shooting the laptop was necessarily justified or the right thing to do. I don't know enough of the details of the situation to say that. But I will say that there are some security measures that Israel absolutely *has* to take, that would be viewed as unacceptably harsh in places with relatively low levels of terrorist violence. Israel really has very little choice in the matter. They're basically living in a war zone, all the time.
Did I mention that the entire middle-east is officially not on my "places to be sure to visit" list? Call me a wuss if you want, but I prefer to live in an area where the word "terrorist" generally calls to mind news stories from several years ago and several hours' drive away. Small-town USA is good. Our town doesn't make the news much, and we *like* it that way.
> twenty years ago all the scientists were saying we were > headed for a new ice age if we don't change our ways
That was thirty or forty years ago now.
Twenty years ago they were saying Florida would be completely underwater by the year 2000, and that global warming and the greenhouse effect were caused by large, halogen-containing compounds.
I think it was about ten or fifteen years ago they started whining about carbon dioxide.
> it seems like perl has slowly moved from an ubiquitous scripting > language to a fringe research project over the last few years
Perl is not an ubiquitous scripting language.
Perl is *the* ubiquitous scripting language. That hasn't really changed. There are a number of other scripting languages, some of them newer than Perl and being used more and more (notably, Ruby), and some of them older than Perl and being used less and less (notably sh), but what hasn't changed in the last fifteen years is that Perl is about as popular as the three next-most-popular ones combined. (I'm assuming here that Java and VB and C are not counted as scripting languages.)
As far as "fringe research project", there are certainly a number of fringe research projects associated with Perl, but the same could be said of any other major programming language.
> Not only that, but FreeBSD is a far more reliable and > higher-quality core than even Debian could ever hope to be.
If that's true, it's only because FreeBSD refuses to include anything in the core. Even extremely basic things like Perl and bash are ports-tree stuff and go in/usr/local. I'm sorry, but that's cheating.
Version 1.0? Meh. Emacs is up to version 23.1. Heck, way back when I was still using Windows with Cygwin and I compiled my own Emacs for the very first time, it was version 21.0.105 way back then, and that was years ago.
Why should I use any software with a version number less than 20? Everyone should get with the times and upgrade to using Emacs for everything. If you aren't using Emacs to post to slashdot, your software is not up to date or stable enough. Demand better: switch to Emacs.
If that reasoning doesn't make sense, it could be because comparing version numbers from one piece of software to another is inherently meaningless.
(Note: I *do* use Emacs for a lot of stuff. But I also use other software sometimes.)
> Any ideas on why they need such a secret and stealthy UAV in Afghanistan for?
Recon, I suspect. One supposes they've probably used it in Iraq as well, and maybe elsewhere.
The advantage of an unmanned plane is obvious: lower risk of politically expensive casualties.
The advantage of a plane over a satellite are more complex but still straightforward: lower cost and much greater flexibility, both in terms of positioning and in terms of the gear you can load on it for any given mission. If you want to refit a satellite with different or updated gear, you have to put maintenance personnel into orbit. No such problem with a plane. A plane can more easily look at things from an angle other than straight ahead, as well. Not that we'll stop using satellites, obviously. For the big picture you want global coverage, and that means satellites. But for specific operational recon planes are a useful supplement. That's not new. What's new about these planes is that they're unmanned.
The article doesn't say whether this UAV is autonomous or remote-control, but I would guess probably the latter, on the grounds that it would be easier to implement, and probably more reliable, though I suppose you would want to build in an emergency autopilot in case the control signal gets jammed somehow; an obvious thing to have it do if it can't reestablish contact with the control channel is climb to altitude and return to base. Ultimately I suspect they'll want to go with a hybrid approach -- a pre-programmed flight plan that the autopilot falls back to if it loses the control channel. But I doubt whether the first deployed models had that, because they wouldn't have wanted to hold off on deploying the thing until they got that far. Logical course would have been to go ahead and deploy models that are basically just remote-control, and then work on developing progressively better emergency autopilot and other features later. With luck, it might even be possible to retrofit such new features onto existing models, maybe even with a firmware update. But I doubt whether the first ones had such amenities initially.
> Either that OR he wants to make sure that the censorship > law is congruent with the German constitution.
Germany: where the government routinely censors political ideas and historical facts without qualm, but censoring child pornography would be unconstitutional.
> Every modern multitasking operating system will, in a low-memory situation, > terminate background processes in favour of foreground processes.
Actually, they usually terminate the process that's using the most memory, or at least that has been my experience.
What the OS *should* do in a memory-starved scenario, IMO, is autosuspend/hibernate some processes to disk (assuming there's enough disk space, which is unlikely to be a problem on a modern system) so that they can be recovered later if more memory becomes available.
And they *sort* of already do this, with virtual memory and swapping, but the implementation could be improved significantly. Among other things, to cut down on thrashing, processes that are swapped out in a low-memory scenario should not get time slices or be swapped back in until memory becomes available (or until the user pulls that application to the foreground, if such can be determined; I don't think most current operating system kernels have a mechanism for knowing whether an app is in the foreground, which is... unfortunate; the window manager and/or shell should be providing the kernel with information about that). Additionally, automatically suspended/hibernated processes should probably be written to files on the filesystem and their mapped virtual memory released. Admittedly this means an extra copy operation (plus another one to put it back later), but it also means you don't have a bunch of inactive processes cluttering up your virtual memory space, which seems like a good trade-off to me. Alternately, additional swap space could be created on the fly (which a handful of kernels, notably NT, already do, available disk space permitting). That's not as aesthetically "clean" as moving suspended processes out of virtual memory entirely, but it might perform better.
> As it turns out, doing the experiment 15 times and taking the > majority (plugging 7 into S(n)) will give you the correct answer > 99.4% of the time. Doing things 35 times gets you to five nines > of accuracy... completely reasonable in my books.
If it were a standard electronic computer, you'd return it for a refund and never buy anything from that manufacturer again.
Five nines sounds good, because for an *uptime* figure, it is good.
But for computational accuracy, that's abysmal. At a speed of only 1 GHz, which isn't exactly a screaming speed demon these days, your computer would be making ten thousand mistakes per second. For traditional computing applications, that makes it so much useless scrap. If our computers make one computational error per *hour*, we complain about how horribly buggy the software is, because we have come to expect that our hardware won't do that to us.
The other poster's suggestion, of using the quantum computer to suggest candidate answers and a traditional electronic computer to check them, makes much better sense. There are whole categories of computing problems where checking the answers is trivial, if you can in the first place get a candidate answer that's at all likely to be correct.
That, and obviously this is a research prototype and the accuracy could potentially be improved in the future, just as we expect that the performance level will be improved by further research.
> It won't matter anyhow, Microsoft just ignorantly invoked the Streisand effect.
Maybe.
> Note to everyone out there faced with a "leak": The best thing to do is NOTHING.
That depends on your goal.
> By trying to have something removed, it will only be spread more widely.
True.
> If Microsoft had simply ignored the incident, Cryptome would have hosted > it and the vast majority of people would have never even heard of COFEE.
Also true.
> Now, tons of people are downloading it just BECAUSE > of the reports of their takedown campaign.
Yes. But, speaking as a cynic, how do you know that this isn't why it was leaked in the first place? How do you know Microsoft didn't *plan* to invoke this reverse-psychology failed suppression effect to create buzz?
There's an old saying: there's no such thing as bad publicity. This isn't strictly true in the absolute sense, but it's much *closer* to true than many people realize. If it weren't for bad publicity, Microsoft would hardly have any publicity at all and yet, somehow, they seem to be doing okay, fiscally speaking.
Maybe I'm giving Microsoft too much credit, but it seems possible to me that someone within the organization arranged this on purpose, unofficially of course, as a form of marketing.
> Then you compromise a binary that has opted out of strict NX, > such as a Java virtual machine that needs to dynamically > recompile JVM bytecode to x86 bytecode.
Great, now we're all going to have to patch our befunge compilers.
London has one of the most active Perl Mongers groups on the planet. I think they're responsible for a double-digit percentage of the content on the CPAN. As a geek, you can't go to London and not visit the Perl Mongers group. It would be like going to the city where the Usenix conference is being held, during the week they're holding it, and not attending the conference. Totally not an option.
I think there's also some famous clock tower thing and some stuff.
> This thing has the potential IMHO to become a Synaptic for Windows!
There are other problems, though, most of them stemming from the fact that neither Windows nor the applications in question are designed for this.
For example, it is traditional, in the Windows universe, for developers to give their software the most completely insane default install options they can dream up. For instance, by default, IrfanView doesn't associate itself with most of the file types it supports, even important ones that you probably don't have any other software for (e.g., TIFF). Similarly, OO.o is afraid to associate itself with.doc by default, even though you don't have MS Office. (I guess they think someone might actually want to use WordPad, or maybe they were just on drugs.) Other software thinks its a good idea by default to associate itself with file types it doesn't even support properly and which you probably have other software for. Some software has lots of optional components, in which case the defaults for which ones to install are never what you would actually want. And so on. Everyone who's ever had to manage Windows systems knows the rule: never ever pick the default/easy/express install option. *Always* pick the advanced/custom option. This rule is much older than the "browser toolbars" that Ninite says it says no to.
Yeah, but does it associate them all with the correct file types? (IrfanView should be associated with all the raster image file types that don't have layers, Gimp with the ones that do have layers, Inkscape with vector image types it supports,...)
Most people, in my experience, can't discern jack diddly squat with their ears.
Most people can't hear the difference between a singer with years of vocal training versus a random person off the street who holds consonants, changes keys at random, and is a quarter tone flat half the time. Most people can't hear the difference between proper counterpoint and simple harmony. Heck, a lot of people can't hear the difference between melody alone versus melody with harmony. They just plain don't year the extra parts. Furthermore, most people can't tell the difference between a new CD versus a cassette tape that's been floating around the console and floor of a car for five years getting heated by the sun, chilled in winter, and dirty from salty-snow off of boots. Basically, hoi polloi can't tell the difference between music and noise.
But people with actual *ears* can tell. Music with mp3 compression sounds like jpeg images look: horrible. Ogg a little better, but still clearly inferior. FLAC, assuming your system isn't too heavily loaded, sounds just like WAV, very comparable in quality to a CD.
While we're at it, a lot of people can't see the difference between 16-bit and 24-bit color and can't see any difference between serif and sans-serif fonts (unless one's a lot bigger or bolder than the other), let alone stuff like kerning. A lot of people just plain never learn to pay attention to details.
This does not mean mp3 is a good format for music. It must means most people have no discernment, which, frankly, is nothing new.
> Are they only counting the places where people go > to the page and do a search or are they counting > all the 'embedded' searches which are snuck into > other apps like IE and Windows Live to boost numbers?
Don't be an idiot. This is Bing we're talking about, not Yahoo. Do you really think 10% of people go to it on purpose? Outside of extreme geekdom, nobody's even heard of it yet.
Basically what this means is IE8 has, mostly as a result of automatic updates, reached about 10% market share among people who think the browser's location bar is a search box and haven't bothered to express an opinion about what search engine it should use. IE8 ships with "Live Search", alias Bing, as the default; IE6 and IE7 used MSN Search as their default, so what we're seeing here is mostly new-version uptake.
There are also a few geeks using it on purpose to try it out, but even if 100% of the slashdot-reading population did that it wouldn't be anywhere near 1% market share, let alone 10%. And the single most popular search engine among the slashdot-reading geekdom is almost certainly still Google at this point.
No, the bulk of the 10% we're talking about here consists of people using the IE8 UI.
> Now that they are out of prison, German law states that they > can't be referred to by name in relation to the killings.
Great. I find new reasons to be worried about Germany every year.
We are talking about historical facts here, and the censorship thereof. That's *scary*.
On a related note, Germany also still restricts political speech. Apparently in the last hundred years of history they have learned... nothing. Germany tops my list of "most likely countries to cause World War III", not because there aren't plenty of other governments this retarded, but because the others are all located in third-world countries. Germany is a major western post-industrial economic powerhouse, and that makes their backwardness on this issue (freedom of political speech) very dangerous.
> You ever priced a Linux Guru? Cheap they ain't because there simply aren't many of them.
Well, a junior high or high school could almost certainly find some students capable of maintaining it (as long as these are non-sensitive computers we're talking about -- computer lab computers and such, not the one in the office that makes the report cards) and willing to do so in exchange for minor privileges (like, getting out of study hall). But a primary school might have a harder time of that, I guess.
Perhaps the most significant German advance in firearms during WWII was the "machine pistol", i.e., the assault rifle. (The name "machine pistol" make sit sound more like a submachine gun, along the lines of the Thompson. It wasn't.)
> Resource limits did play a part in it, but high tech was used by both sides of the > conflict, and it is very unlikely that USSR would've won if it went for quantity alone.
Another important factor is that Germany was fighting a multi-front war.
On top of that, Hitler diverted significant resources from the German war effort toward the fulfillment of his twisted political agenda.
So yeah, there are a lot of reasons why Germany lost.
Whereas, with Japan, there was really two reason. First, they were outmatched, plain and simple. Japan did not have adequate economic and industrial infrastructure to sustain a long-term conflict against the United States. Second, their allies lost, which left them going it alone, which is usually not too helpful.
> Things like radar were also highly advanced military technologies, > which the US/Brits had, and the Germans/Japanese did not.
Not to mention signals intelligence and machine-assisted cryptanalysis.
> In short, the Axis were only slightly more advanced than the Allies.
I would say they were only *arguably* more advanced overall, though they were definitely more advanced in certain areas.
> Where the level of military technology is more disparate, it > certainly can and does become an overwhelming advantage.
You mean like the US invasion of Iraq? (But, there was also a huge production and infrastructure imbalance there, which was probably an even bigger factor.)
Another thing that can make a huge difference is the level of training and readiness of the units that you do have. Any questions about that, look up the Six Day War. (Hint: it's called that for a reason. There was relatively little technology gap, but it was totally one-sided and over very fast; the basic outcome had become clear to all rational observers within twenty-four hours. Why? The guys on the one side knew what they were doing, and the guys on the other side very much didn't.)
And a really good general can make a pretty big difference too. The US civil war, all else being equal, should have been over in about three months; home-field advantage ("we're fighting for our homes" and all that) could have been expected to drag it out to maybe twice that long. But the south had most of the good generals, particularly Lee and Jackson, and darned if they didn't almost win, despite running out of ammunition, out of guns, out of gunpowder, out of food, out of cannons, out of important raw materials (e.g., iron), low on horses, and just about running out of fighting-age men.
Slashdot Mirror
> using something called MediaSite to view quarterly state library "webinars"
I should probably explain this better. A MediaSite webinar is similar to a YouTube video, with a few differences, which I will attempt to list. First, it's scheduled for a certain time and announced ahead of time, and you can visit the page ahead of time, but you can't watch it until the scheduled time. Second, the interface is significantly less user-friendly and as near as I can tell does not feature rewind or fast forward, though I could be missing something. Also, you can't watch it through a proxy, even a transparent proxy such as is commonly used for content filtering. You have to have Silverlight installed to view it. And instead of amusement, the primary intended purpose is pedagogical.
And yeah, the state library ought to know better. Why they can't just throw an MPEG up on a web server and link to it, I'll never know. They certainly have the bandwidth for that, especially since almost all of the viewers would be watching from within libraries that are connected to the internet via OPLIN, so the pipes to the outside world wouldn't even be involved in the overwhelming majority of cases. The T1 lines to each individual user retrieving the thing would be the bottleneck.
> The only time I've ever used Silverlight is when I've watched
> SkyTV online in the UK as a media thingy for your browser.
Okay, that's the *third* use for Silverlight that I've heard of. The other two (playing around with the demo site Microsoft set up to advertise Silverlight, and using something called MediaSite to view quarterly state library "webinars" on the current state of library funding) didn't interest me either. So yeah, Silverlight is a solution in search of a problem. Is there anyone outside of Redmond who didn't know that?
Furthermore, even if Silverlight were awesomely useful and made whole new kinds of incredibly cool sites possible that were barely conceivable before, Moonlight would still be an astonishingly bad idea for the open-source community.
I mean, yeah, instead of protecting the open platform-neutral internet, let's actively help Microsoft decommoditize and proprietize it and make it part of the Windows-centric integrated Microsoft product offering stack, so that web developers all end up using Visual Studio and doing everything with functions provided by IIS and SQL Server and DirectX. That'll sure improve things for everyone.</sarcasm>
If you didn't realize before that Moonlight was a bad idea, think about the fact that Microsoft is making this promise not to sue, and what that implies. I don't mean Microsoft's motivation in making the promise, because that's debatable. No, I'm talking about the fact that anyone would even *wonder* whether they might sue Moonlight users, the fact that the promise is even necessary or meaningful, the fact that they would even think to make it. That is in itself a dead giveaway that relying on Moonlight for anything that matters is playing with fire. Frankly, if Microsoft released a closed-source Silverlight plugin for Linux systems, that would scare me less.
Actually, I consider the whole Mono project to be an inherently brain-damaged idea. I absolutely do not want Mono installed on any computer I use or administer. Ever.
> Of course, we lived with a couple of decades of terrorist
> attacks ourselves long before Terrorism was the new bogeyman.
The UK only *thinks* they have terrorism problems. (The US too, for that matter.)
Spend a few months living in Israel, and you'll find out how it really is.
I'm not saying shooting the laptop was necessarily justified or the right thing to do. I don't know enough of the details of the situation to say that. But I will say that there are some security measures that Israel absolutely *has* to take, that would be viewed as unacceptably harsh in places with relatively low levels of terrorist violence. Israel really has very little choice in the matter. They're basically living in a war zone, all the time.
Did I mention that the entire middle-east is officially not on my "places to be sure to visit" list? Call me a wuss if you want, but I prefer to live in an area where the word "terrorist" generally calls to mind news stories from several years ago and several hours' drive away. Small-town USA is good. Our town doesn't make the news much, and we *like* it that way.
> twenty years ago all the scientists were saying we were
> headed for a new ice age if we don't change our ways
That was thirty or forty years ago now.
Twenty years ago they were saying Florida would be completely underwater by the year 2000, and that global warming and the greenhouse effect were caused by large, halogen-containing compounds.
I think it was about ten or fifteen years ago they started whining about carbon dioxide.
> it seems like perl has slowly moved from an ubiquitous scripting
> language to a fringe research project over the last few years
Perl is not an ubiquitous scripting language.
Perl is *the* ubiquitous scripting language. That hasn't really changed. There are a number of other scripting languages, some of them newer than Perl and being used more and more (notably, Ruby), and some of them older than Perl and being used less and less (notably sh), but what hasn't changed in the last fifteen years is that Perl is about as popular as the three next-most-popular ones combined. (I'm assuming here that Java and VB and C are not counted as scripting languages.)
As far as "fringe research project", there are certainly a number of fringe research projects associated with Perl, but the same could be said of any other major programming language.
> Not only that, but FreeBSD is a far more reliable and
/usr/local. I'm sorry, but that's cheating.
> higher-quality core than even Debian could ever hope to be.
If that's true, it's only because FreeBSD refuses to include anything in the core. Even extremely basic things like Perl and bash are ports-tree stuff and go in
Version 1.0? Meh. Emacs is up to version 23.1. Heck, way back when I was still using Windows with Cygwin and I compiled my own Emacs for the very first time, it was version 21.0.105 way back then, and that was years ago.
Why should I use any software with a version number less than 20? Everyone should get with the times and upgrade to using Emacs for everything. If you aren't using Emacs to post to slashdot, your software is not up to date or stable enough. Demand better: switch to Emacs.
If that reasoning doesn't make sense, it could be because comparing version numbers from one piece of software to another is inherently meaningless.
(Note: I *do* use Emacs for a lot of stuff. But I also use other software sometimes.)
> Any ideas on why they need such a secret and stealthy UAV in Afghanistan for?
Recon, I suspect. One supposes they've probably used it in Iraq as well, and maybe elsewhere.
The advantage of an unmanned plane is obvious: lower risk of politically expensive casualties.
The advantage of a plane over a satellite are more complex but still straightforward: lower cost and much greater flexibility, both in terms of positioning and in terms of the gear you can load on it for any given mission. If you want to refit a satellite with different or updated gear, you have to put maintenance personnel into orbit. No such problem with a plane. A plane can more easily look at things from an angle other than straight ahead, as well. Not that we'll stop using satellites, obviously. For the big picture you want global coverage, and that means satellites. But for specific operational recon planes are a useful supplement. That's not new. What's new about these planes is that they're unmanned.
The article doesn't say whether this UAV is autonomous or remote-control, but I would guess probably the latter, on the grounds that it would be easier to implement, and probably more reliable, though I suppose you would want to build in an emergency autopilot in case the control signal gets jammed somehow; an obvious thing to have it do if it can't reestablish contact with the control channel is climb to altitude and return to base. Ultimately I suspect they'll want to go with a hybrid approach -- a pre-programmed flight plan that the autopilot falls back to if it loses the control channel. But I doubt whether the first deployed models had that, because they wouldn't have wanted to hold off on deploying the thing until they got that far. Logical course would have been to go ahead and deploy models that are basically just remote-control, and then work on developing progressively better emergency autopilot and other features later. With luck, it might even be possible to retrofit such new features onto existing models, maybe even with a firmware update. But I doubt whether the first ones had such amenities initially.
> Either that OR he wants to make sure that the censorship
> law is congruent with the German constitution.
Germany: where the government routinely censors political ideas and historical facts without qualm, but censoring child pornography would be unconstitutional.
> Every modern multitasking operating system will, in a low-memory situation,
> terminate background processes in favour of foreground processes.
Actually, they usually terminate the process that's using the most memory, or at least that has been my experience.
What the OS *should* do in a memory-starved scenario, IMO, is autosuspend/hibernate some processes to disk (assuming there's enough disk space, which is unlikely to be a problem on a modern system) so that they can be recovered later if more memory becomes available.
And they *sort* of already do this, with virtual memory and swapping, but the implementation could be improved significantly. Among other things, to cut down on thrashing, processes that are swapped out in a low-memory scenario should not get time slices or be swapped back in until memory becomes available (or until the user pulls that application to the foreground, if such can be determined; I don't think most current operating system kernels have a mechanism for knowing whether an app is in the foreground, which is... unfortunate; the window manager and/or shell should be providing the kernel with information about that). Additionally, automatically suspended/hibernated processes should probably be written to files on the filesystem and their mapped virtual memory released. Admittedly this means an extra copy operation (plus another one to put it back later), but it also means you don't have a bunch of inactive processes cluttering up your virtual memory space, which seems like a good trade-off to me. Alternately, additional swap space could be created on the fly (which a handful of kernels, notably NT, already do, available disk space permitting). That's not as aesthetically "clean" as moving suspended processes out of virtual memory entirely, but it might perform better.
> As it turns out, doing the experiment 15 times and taking the
> majority (plugging 7 into S(n)) will give you the correct answer
> 99.4% of the time. Doing things 35 times gets you to five nines
> of accuracy... completely reasonable in my books.
If it were a standard electronic computer, you'd return it for a refund and never buy anything from that manufacturer again.
Five nines sounds good, because for an *uptime* figure, it is good.
But for computational accuracy, that's abysmal. At a speed of only 1 GHz, which isn't exactly a screaming speed demon these days, your computer would be making ten thousand mistakes per second. For traditional computing applications, that makes it so much useless scrap. If our computers make one computational error per *hour*, we complain about how horribly buggy the software is, because we have come to expect that our hardware won't do that to us.
The other poster's suggestion, of using the quantum computer to suggest candidate answers and a traditional electronic computer to check them, makes much better sense. There are whole categories of computing problems where checking the answers is trivial, if you can in the first place get a candidate answer that's at all likely to be correct.
That, and obviously this is a research prototype and the accuracy could potentially be improved in the future, just as we expect that the performance level will be improved by further research.
> It won't matter anyhow, Microsoft just ignorantly invoked the Streisand effect.
Maybe.
> Note to everyone out there faced with a "leak": The best thing to do is NOTHING.
That depends on your goal.
> By trying to have something removed, it will only be spread more widely.
True.
> If Microsoft had simply ignored the incident, Cryptome would have hosted
> it and the vast majority of people would have never even heard of COFEE.
Also true.
> Now, tons of people are downloading it just BECAUSE
> of the reports of their takedown campaign.
Yes. But, speaking as a cynic, how do you know that this isn't why it was leaked in the first place? How do you know Microsoft didn't *plan* to invoke this reverse-psychology failed suppression effect to create buzz?
There's an old saying: there's no such thing as bad publicity. This isn't strictly true in the absolute sense, but it's much *closer* to true than many people realize. If it weren't for bad publicity, Microsoft would hardly have any publicity at all and yet, somehow, they seem to be doing okay, fiscally speaking.
Maybe I'm giving Microsoft too much credit, but it seems possible to me that someone within the organization arranged this on purpose, unofficially of course, as a form of marketing.
> Then you compromise a binary that has opted out of strict NX,
> such as a Java virtual machine that needs to dynamically
> recompile JVM bytecode to x86 bytecode.
Great, now we're all going to have to patch our befunge compilers.
London has one of the most active Perl Mongers groups on the planet. I think they're responsible for a double-digit percentage of the content on the CPAN. As a geek, you can't go to London and not visit the Perl Mongers group. It would be like going to the city where the Usenix conference is being held, during the week they're holding it, and not attending the conference. Totally not an option.
I think there's also some famous clock tower thing and some stuff.
> This thing has the potential IMHO to become a Synaptic for Windows!
.doc by default, even though you don't have MS Office. (I guess they think someone might actually want to use WordPad, or maybe they were just on drugs.) Other software thinks its a good idea by default to associate itself with file types it doesn't even support properly and which you probably have other software for. Some software has lots of optional components, in which case the defaults for which ones to install are never what you would actually want. And so on. Everyone who's ever had to manage Windows systems knows the rule: never ever pick the default/easy/express install option. *Always* pick the advanced/custom option. This rule is much older than the "browser toolbars" that Ninite says it says no to.
There are other problems, though, most of them stemming from the fact that neither Windows nor the applications in question are designed for this.
For example, it is traditional, in the Windows universe, for developers to give their software the most completely insane default install options they can dream up. For instance, by default, IrfanView doesn't associate itself with most of the file types it supports, even important ones that you probably don't have any other software for (e.g., TIFF). Similarly, OO.o is afraid to associate itself with
How does Ninite cope with install options?
Yeah, but does it associate them all with the correct file types? (IrfanView should be associated with all the raster image file types that don't have layers, Gimp with the ones that do have layers, Inkscape with vector image types it supports, ...)
Yeah. Microsoft isn't the only culprit, though. Symantec, for instance, does the same thing.
Most people, in my experience, can't discern jack diddly squat with their ears.
Most people can't hear the difference between a singer with years of vocal training versus a random person off the street who holds consonants, changes keys at random, and is a quarter tone flat half the time. Most people can't hear the difference between proper counterpoint and simple harmony. Heck, a lot of people can't hear the difference between melody alone versus melody with harmony. They just plain don't year the extra parts. Furthermore, most people can't tell the difference between a new CD versus a cassette tape that's been floating around the console and floor of a car for five years getting heated by the sun, chilled in winter, and dirty from salty-snow off of boots. Basically, hoi polloi can't tell the difference between music and noise.
But people with actual *ears* can tell. Music with mp3 compression sounds like jpeg images look: horrible. Ogg a little better, but still clearly inferior. FLAC, assuming your system isn't too heavily loaded, sounds just like WAV, very comparable in quality to a CD.
While we're at it, a lot of people can't see the difference between 16-bit and 24-bit color and can't see any difference between serif and sans-serif fonts (unless one's a lot bigger or bolder than the other), let alone stuff like kerning. A lot of people just plain never learn to pay attention to details.
This does not mean mp3 is a good format for music. It must means most people have no discernment, which, frankly, is nothing new.
> Are they only counting the places where people go
> to the page and do a search or are they counting
> all the 'embedded' searches which are snuck into
> other apps like IE and Windows Live to boost numbers?
Don't be an idiot. This is Bing we're talking about, not Yahoo. Do you really think 10% of people go to it on purpose? Outside of extreme geekdom, nobody's even heard of it yet.
Basically what this means is IE8 has, mostly as a result of automatic updates, reached about 10% market share among people who think the browser's location bar is a search box and haven't bothered to express an opinion about what search engine it should use. IE8 ships with "Live Search", alias Bing, as the default; IE6 and IE7 used MSN Search as their default, so what we're seeing here is mostly new-version uptake.
There are also a few geeks using it on purpose to try it out, but even if 100% of the slashdot-reading population did that it wouldn't be anywhere near 1% market share, let alone 10%. And the single most popular search engine among the slashdot-reading geekdom is almost certainly still Google at this point.
No, the bulk of the 10% we're talking about here consists of people using the IE8 UI.
> You sir, are a fool or a very good troll
Whoosh.
It was a joke, obviously, albeit a somewhat macabre one.
Store? You went into a *store*?
I got mine from a deaf-blind hermit who assembles them from junk parts. I paid him with vegetables.
> Now that they are out of prison, German law states that they
> can't be referred to by name in relation to the killings.
Great. I find new reasons to be worried about Germany every year.
We are talking about historical facts here, and the censorship thereof. That's *scary*.
On a related note, Germany also still restricts political speech. Apparently in the last hundred years of history they have learned... nothing. Germany tops my list of "most likely countries to cause World War III", not because there aren't plenty of other governments this retarded, but because the others are all located in third-world countries. Germany is a major western post-industrial economic powerhouse, and that makes their backwardness on this issue (freedom of political speech) very dangerous.
> You ever priced a Linux Guru? Cheap they ain't because there simply aren't many of them.
Well, a junior high or high school could almost certainly find some students capable of maintaining it (as long as these are non-sensitive computers we're talking about -- computer lab computers and such, not the one in the office that makes the report cards) and willing to do so in exchange for minor privileges (like, getting out of study hall). But a primary school might have a harder time of that, I guess.
Perhaps the most significant German advance in firearms during WWII was the "machine pistol", i.e., the assault rifle. (The name "machine pistol" make sit sound more like a submachine gun, along the lines of the Thompson. It wasn't.)
> Resource limits did play a part in it, but high tech was used by both sides of the
> conflict, and it is very unlikely that USSR would've won if it went for quantity alone.
Another important factor is that Germany was fighting a multi-front war.
On top of that, Hitler diverted significant resources from the German war effort toward the fulfillment of his twisted political agenda.
So yeah, there are a lot of reasons why Germany lost.
Whereas, with Japan, there was really two reason. First, they were outmatched, plain and simple. Japan did not have adequate economic and industrial infrastructure to sustain a long-term conflict against the United States. Second, their allies lost, which left them going it alone, which is usually not too helpful.
> Things like radar were also highly advanced military technologies,
> which the US/Brits had, and the Germans/Japanese did not.
Not to mention signals intelligence and machine-assisted cryptanalysis.
> In short, the Axis were only slightly more advanced than the Allies.
I would say they were only *arguably* more advanced overall, though they were definitely more advanced in certain areas.
> Where the level of military technology is more disparate, it
> certainly can and does become an overwhelming advantage.
You mean like the US invasion of Iraq? (But, there was also a huge production and infrastructure imbalance there, which was probably an even bigger factor.)
Another thing that can make a huge difference is the level of training and readiness of the units that you do have. Any questions about that, look up the Six Day War. (Hint: it's called that for a reason. There was relatively little technology gap, but it was totally one-sided and over very fast; the basic outcome had become clear to all rational observers within twenty-four hours. Why? The guys on the one side knew what they were doing, and the guys on the other side very much didn't.)
And a really good general can make a pretty big difference too. The US civil war, all else being equal, should have been over in about three months; home-field advantage ("we're fighting for our homes" and all that) could have been expected to drag it out to maybe twice that long. But the south had most of the good generals, particularly Lee and Jackson, and darned if they didn't almost win, despite running out of ammunition, out of guns, out of gunpowder, out of food, out of cannons, out of important raw materials (e.g., iron), low on horses, and just about running out of fighting-age men.