Re:But does it have a useable file-save dialogue?
on
GNOME 2.16 Released
·
· Score: 1
So your problem is that Nautilus hides hidden directories too well? Snarky comment aside, there is an option to show hidden files, and I daresay the option is less hidden than in windows. Given that hidden files are primarily used to remove uninteresting (96 percent of the time) files from view, I'd say this isn't so bad. Far more offensive is that the List view can easily make the right click context menu, the one for making things like new folders, unaccessible if the directory already has enough entries in it to remove all blank space. It's still possible to do everything, but habits from previous window environments leads me to try that right click menu first:(
Maybe we should be looking at tablet pcs instead. They're a bit pricier, and rarer, but they don't suffer the horrible mouse input that other laptops do. But I've been struggling to find one with an nvidia card newer than a 6200 =(
Actually, the sophisticated software tricks are pretty much required to do anything neat in an embedded system. Multithreading in this case pretty much means interrupt handling. A decent compiler will remove the overhead of working in a slightly higher level language than machine code, but you can still access the registers as needed. Frankley, neglecting to teach students this will cost them in a job market where productivity is paramount to employability.
I work in a biomed research lab currently, and the guys that do all their programming in assembler are much less productive, finding themselves debugging stack pointer problems etc. It's okay for them because the students define a lot of the scope of any project, and their salary doesn't leave their bosses a lot of leeway to micromanage the research. If you make researching hell, it'll cost you assistants. These guys are still pretty bright folks, they just fall into obsessive complusive mode over the design. It's simply too much for a single guy to worry about the fabrication design, the part specs, the programming and testing to worry about every detail like how many bytes can be saved by careful consideration. Especially when doubling the size of EEPROM costs 20 cents plus the cost of making it a marginally larger device.
There's an old saying: "Never trust a programmer with a soildering iron, a hardware guy with a software fix, or a user with an idea."
Realistically, I don't see how welding is at all relevant to real time systems. If you're gonna build something that needs welds, hire a liscenced Professional Engineer to make sure you do the damn thing right.
My university holds a real time systems course that uses Mindstorms so that CS majors can build physical systems without needing to know things like welding and soildering and other machining things. But labview is a bit underpowered for us, when you want to talk about things like multithreaded applications undergoing realtime constraints. Traditionally we've used "Not Quite C", which is a c syntax with a few useful extensions and a library capable of manipulating some of the specific hardware devices on the lego command brick.
Not Quite C appears to have died out in 2005, is there anything in the works for this new stuff, or is it merely the same parts in different colors?
It might make more sense to develop a purpose specific card to push raytracing graphics, rather than an Intel chip. I've seen a university project to build one and they demonstrated an architecture that pretty much scaled up performance as they added more silicon. Which is to be somewhat expected; give each chip a region of view to render and they can cast their rays independently. The critical aspect of this system is memory, and memory contention. The good news is that rendering doesn't need to modify the shared geometry, and they should be rendering to different parts of the frame buffer, so it won't be a process synchronization hell (there will still be some problems when you need to modify that geometry when the frame is done, and figuring out how to ). Critics will tell you that this is going to destroy the cache, but the news is, what little cache coherency on raster cards that's left will continue to die out as textures increase in size, and as we introduce multiple texturing (bump mapping, textures, dirt maps, alpha maps, etc) per polygon. iD's new Quake wars is using some ridiculus texture size to reduce terrain fractals, and no matter how it's done it's gotta put the GPU memory at strain. GPUs have been pushing memory technology for a while now; GDDR3, common in cards for the last three or four years, is DDR2 improved to reduce heat production (and thus up the clock).
Compared to traditional raster based solutions, ray tracing hardware is underresearched. There's a whole host of possibilities with gaming in mind. You can do iterative improvement of the frame. You can dedicate more rays to the middle of the frame (presumably the object of interest). And that's just shit some dude who hasn't spent years of his life pondering this came up with. I'll be much more interested to see what's presented this September at IEEE's RT raytracing conference..
I really don't get what all this bitching about wireless is. "It should be as easy as Windows," people say. Fuck that. It needs to be easier than Windows. I don't want to dig around for an hour to find out that somehow the Windows Wireless Zero Configuration service was disabled, or find out that ZeroConf doesn't support my card at all.
I put in an Ubuntu Live CD the other day, and without a single question I had wireless set up and working. This might even be too easy, because I was connected to someone else's open point. But their networking system operates well for me. Yea, if you use gentoo, your configuration systems for wireless are pretty limited and shitty. But aren't you supposed to be editing leet/etc files anyways?
Each popular distro has a set of places to go for support. Some take a more hard line against things like ndiswrapper and mp3 support than others. Because each distro, while based on many of the same upstream sources, is still unique, each one requires its own unique set of common solutions. Ideally, the things Automatix fixes should be filed as bugs within Ubuntu rather than having to maintain a strict set of "you shouldn't do this if you live in the us or any other first world nation" guides and scripts. But I just hinted at what the real problem is. Unless you plan on purchasing and dedicating the various patents, I don't see how purchasing a liscence for all your users is a long term solution to encumbered software or open sourcing in general. This interview did a very piss poor job of getting ESR to answer the hard questions about what his advice means down the road.
Well, there's a couple of motivations behind that. First is the leveraged buyout, which was made rather popular in the 1980s before it degenerated into what's known as the "junk bond" scandal. Basically, you buy a company out, using the company you just bought as colateral. The financing behind it is much more complicated, but that's close to what it is underneath. Given the context of the situation (LBOs), this is mostly what they're referring to.
The other idea is that companies are supposed to seek above market returns. As long as the return their loan was spent on pays out better than the terms of the loan, it shouldn't not be an option. The problem is, as you've described, Microsoft is already making more cash than above market return ideas. The huge lump sum dividend is proof enough that they had no better place to invest it. And once you slash their R&D efforts and their practice of hiring good programmers just to keep them away from potential competitors, you'll have even more cash that you can't spend fast enough. Which may be what you meant when you said that the richest [public] company in the world needs to borrow. However, if part of the buyout means putting in new management with a much more aggressive set of ideas to pursue it might make sense. Primarily that would mean a lot more mergers and acquisitions, which is a really bad idea for a company that's already been subject to anticompetitive sanctions and still maintains a near monopoly.
I think it's along the lines of ISPs doing this without informing their customers what has happened. Their customers need not know that SBC extracted a heavy toll from YouTube or Google in order to deliver their video. And that even if you could know when your connection was tiered, no market offering would exist for an untiered connection. In other words, they're levying their massive subscriber base against people who profit from them having a decent internet connection, by holding it ransom. You'll note they aren't calling it anything like QoS, because that would imply that the offering has some level of reliability / quality.
I'm no lawyer, but this liscence seems awfully specific in all the wrong places. Lots of stuff specific to Microsoft, like suing them for patents etc. It's fine to sue users and derivate software distributers, just not MS. Moreover, it grants you liscence to use their copyrights and patents, so long as you grant a specific set of copyrights and patents ("you must grant all recipients the copyright and patent licenses in sections 2(A) & 2(B) for any file that contains code from the software"), likely including things you don't own. You don't have the authority to liscence Microsoft's patents to other people directly. I imagine that Microsoft intends that their offer is extended to anyone who might recieve this software down the chain, but it isn't made clear whether this is good enough. These two oddities combined basically means distributing the software, which MS inevitably holds a patent on, requires you to grant a liscence you don't own (and must seek from MS) and still allows them to sue to your users for infriging on their patent.
Try swinging your arm in circles for about a minute. Now try to justify how your arm, which is likely incredibly tired now, will be better able to hold up to playing Zelda sword fighting for 45 minutes to an hour. Sure, you can take short breaks, but: a) who wants to stop playing Zelda? b) zelda's design pratically requires 45 minutes of effort to accomplish anything significant. getting halfway through the water temple in ocarina for the first time, for example, was no mean feat of exploration and ingeniuity. c) we really have no idea exactly how the swordfighting will be done, but if it is going to be something twitchy, it's not what consumers will be expecting. I guess this is a case of getting what you wished for.
What will be even worse is when the physicists and chemists gang up to pummel computer scientists for using 'atom' to describe a kind of data. Whatever. Atoms were supposed to be indivisible, and they're still using the same word to describe something that contains both subatomic particles and sub-subatomic particles, to say nothing of what quarks might be made of:P
I don't think the grandparent was claiming that the majority of BT traffic was legal. Just that legitimate uses do exist. Unless you were trying to refute his claim that BT was legitimate. In which case, you did a pretty poor job. That argument is similar to blaming cars for the fact that the majority of highway traffic in urban areas breaks the speed limit. This brings up an interesting counter-question though: what does it mean for democracy when a majority of the population ignores a law?
I donno if Yoshi was all that bad. Down + B had just about the strongest priority in the game, which meant you could pwn just about anything from above. Certainly, doing anything other than blocking it was a feat of incredible timing.
This is just the continuation of the rhetorical split between GNU and Linux enthusiasts, of which ESR belongs to the linux camp. The cathedral and the bazaar, after all, was written primarily with GNU as the cathedral, and Linux as the bazaar. Of course, neither side's participants exclusively believe a certain way on the subjects around open source, but it's important to note that RMS's statement generally matches with Linus's, in that things like the nvidia binary driver, while not perfect, are allowed.
If you were truly one of the Annointed looking for the one true Free way, you'd already be familiar with this Linux situation and be running Hurd instead.
The real problem with flash drives is that you need to unmount them before you unplug them. Many, many people don't. And sometimes their data gets corrupted.
Thats not technically true. It's entirely possible to mount the drives as synchronous, so that writes are not buffered. Generally removing drives before requesting a buffer flush shouldn't be a problem unless you're copying a lot of information onto the drive. I suspect the same is true of your campus computer systems. In contrast most Linux distros I've seen prefer to mount drives async (turn on buffering) for the performance benefits on large writes. I've never seen data loss from surprise removal on Windows XP, and in fact the only time I've lost data is attempting to use the drive on a Mac OS 9's USB port that couldn't source the nessecary milliamps. The likelyhood of losing data from surprise removal in Windows seems pretty low unless your administrators have done something silly to the configurations.
The real kicker? They're replacing all the PCs in the campus labs with ones without floppy drives. So even those poor kids with only a few hundred KB of data will have to use a flash drive, and us student assistants will have to support them.
Frankly, this is a good thing. Floppies are shitty. They havent improved as a consumer technology since like 1987; they're prone to failure and painfully slow. Ideally, students using your campus labs should be using something like a network drive, which has the benefits of large storage, redundancy, protection against power failures, and periodic backups. But even if that isn't available, easily accomplished or part of student practices, USB drives are still a generally better alternative than floppies. I just bought a gigabyte of USB thumb drive for 20 dollars online. You can purchase 128M for under 10 dollars, and I suspect your campus bookstore could get a bulk discount to get them even cheaper (whether they decide to share that with students is best rated "unlikely"). Point is, for about the same price as a pack of disks you could get a hell of a lot more storage that runs faster, less prone to failure, and easier to carry around. And if your school wasn't backwards ass, students would have easy access to network storage, both from home and on campus.
Google searching suggests that there are removal tools for that u3 software stuff. I'm not sure if they work or if they'll work for you in an environment where no software is allowed to be installed. It might be easier to file a trouble report asking for the u3 drivers to be installed (assuming this isn't really device/vendor dependent). AFAIK, u3 only requires drivers for programs on the drive and password protected volumes.
I don't recall class sizes of 15 at my own well funded suburban school district, but that's neither here nor now. I realize teachers definately have a severe budget constraint, which is probably why giving free stuff pulls in a lot of clicks from teachers. Certainly, I'd expect the best public school teachers to be masters of getting something for nothing. But I think the author's banking on the idea that if a teacher finds the budget for it, they'll eventually use it as more than just a one-off activity. Certainly, my fourth grade teacher used some word search software for students on a weekly basis for spelling / vocabulary words.
But figuring out where these open source projects will come from is going to be a challenege. I think there's a lot of untapped potential here within schools. Most of these programs are simple to write if you can write software at all. It would be interesting to see high school students do an advanced / 2nd programming course duplicating the functionality of such a program at the request of a local teacher. Of course, it would require students to be taught a suitable language in the first place, so qbasic might be out, no matter how fun Gorillas and Nibbles are. You could even put the results on sourceforge and might get help from Edubuntu if they want to include the software.
Shit, you're right. I remember that too, now. All it takes to be moderated "+5, Informative" is attention to grammar and spelling, and long paragraphs, I guess. Still, it will be interesting to see how Nintendo's previous designs and considerations color !!M.
Bad news, we voted out the evolutionist school board. We're not stupid, we're just not ever vigilant at the ballot box. But you might try moving to Las Vegas: "God's Blind Spot."
I doubt Nintendo will offer an IM service to players. It's useful and lets players build communities, but the potential for Internet preditory practices is something Nintendo understandably wants to avoid, at the expense of building an online community. Look at the awkward systems on the DS. You can't communicate with people you haven't already met in the real world. And even if you do get their number through outside means like a gamefaqs forum, few games allow serious conversations. Mario kart provides no method of communication at all. Metroid Prime only allows you to talk with people you've met before, and only during the time for choosing game options.
What this embargo on interplayer communication means is team oriented play is nigh impossible. You'd have to do something similar to the iconographic messages used in Phantasy Star Online, which was primarily designed to let people of different languages play together. Building up a meaningful vocabulary like that will take a lot of investment from both the designer and the players.
Nevermind the occasional Linux ISO over BT. 20-30 gigabytes is chump change compared to the cost of a disk to store that data on. Instead, think Naruto. With over 2 thousand simultaneous users (and this is almost two weeks old!), it's likely a valuable net gain for them there. And if they go ahead with it, usage would spike even higher; many users find BT throttled or simply slow; if suddenly you were maxxing out that "6 mbps" line the cable company sold you for BT, I don't think you'd bother with DCC bots anymore. Scarywater used to host linux stuff on BT, but it's dwarfed by the sheer scale of their anime tracker.
But if ISPs are going to cache files like this, it pretty much eliminates the point of bittorrent.
Slashdot Mirror
So your problem is that Nautilus hides hidden directories too well? Snarky comment aside, there is an option to show hidden files, and I daresay the option is less hidden than in windows. Given that hidden files are primarily used to remove uninteresting (96 percent of the time) files from view, I'd say this isn't so bad. Far more offensive is that the List view can easily make the right click context menu, the one for making things like new folders, unaccessible if the directory already has enough entries in it to remove all blank space. It's still possible to do everything, but habits from previous window environments leads me to try that right click menu first :(
Maybe we should be looking at tablet pcs instead. They're a bit pricier, and rarer, but they don't suffer the horrible mouse input that other laptops do. But I've been struggling to find one with an nvidia card newer than a 6200 =(
Actually, the sophisticated software tricks are pretty much required to do anything neat in an embedded system. Multithreading in this case pretty much means interrupt handling. A decent compiler will remove the overhead of working in a slightly higher level language than machine code, but you can still access the registers as needed. Frankley, neglecting to teach students this will cost them in a job market where productivity is paramount to employability.
I work in a biomed research lab currently, and the guys that do all their programming in assembler are much less productive, finding themselves debugging stack pointer problems etc. It's okay for them because the students define a lot of the scope of any project, and their salary doesn't leave their bosses a lot of leeway to micromanage the research. If you make researching hell, it'll cost you assistants. These guys are still pretty bright folks, they just fall into obsessive complusive mode over the design. It's simply too much for a single guy to worry about the fabrication design, the part specs, the programming and testing to worry about every detail like how many bytes can be saved by careful consideration. Especially when doubling the size of EEPROM costs 20 cents plus the cost of making it a marginally larger device.
There's an old saying:
"Never trust a programmer with a soildering iron, a hardware guy with a software fix, or a user with an idea."
Realistically, I don't see how welding is at all relevant to real time systems. If you're gonna build something that needs welds, hire a liscenced Professional Engineer to make sure you do the damn thing right.
My university holds a real time systems course that uses Mindstorms so that CS majors can build physical systems without needing to know things like welding and soildering and other machining things. But labview is a bit underpowered for us, when you want to talk about things like multithreaded applications undergoing realtime constraints. Traditionally we've used "Not Quite C", which is a c syntax with a few useful extensions and a library capable of manipulating some of the specific hardware devices on the lego command brick.
Not Quite C appears to have died out in 2005, is there anything in the works for this new stuff, or is it merely the same parts in different colors?
I knew there was a reason I upgraded my processor last month!
There are plenty worse reasons to pick a school, like perhaps the performance of their varsity sports teams.
It might make more sense to develop a purpose specific card to push raytracing graphics, rather than an Intel chip. I've seen a university project to build one and they demonstrated an architecture that pretty much scaled up performance as they added more silicon. Which is to be somewhat expected; give each chip a region of view to render and they can cast their rays independently. The critical aspect of this system is memory, and memory contention. The good news is that rendering doesn't need to modify the shared geometry, and they should be rendering to different parts of the frame buffer, so it won't be a process synchronization hell (there will still be some problems when you need to modify that geometry when the frame is done, and figuring out how to ). Critics will tell you that this is going to destroy the cache, but the news is, what little cache coherency on raster cards that's left will continue to die out as textures increase in size, and as we introduce multiple texturing (bump mapping, textures, dirt maps, alpha maps, etc) per polygon. iD's new Quake wars is using some ridiculus texture size to reduce terrain fractals, and no matter how it's done it's gotta put the GPU memory at strain. GPUs have been pushing memory technology for a while now; GDDR3, common in cards for the last three or four years, is DDR2 improved to reduce heat production (and thus up the clock).
Compared to traditional raster based solutions, ray tracing hardware is underresearched. There's a whole host of possibilities with gaming in mind. You can do iterative improvement of the frame. You can dedicate more rays to the middle of the frame (presumably the object of interest). And that's just shit some dude who hasn't spent years of his life pondering this came up with. I'll be much more interested to see what's presented this September at IEEE's RT raytracing conference..
I really don't get what all this bitching about wireless is. "It should be as easy as Windows," people say. Fuck that. It needs to be easier than Windows. I don't want to dig around for an hour to find out that somehow the Windows Wireless Zero Configuration service was disabled, or find out that ZeroConf doesn't support my card at all.
/etc files anyways?
I put in an Ubuntu Live CD the other day, and without a single question I had wireless set up and working. This might even be too easy, because I was connected to someone else's open point. But their networking system operates well for me. Yea, if you use gentoo, your configuration systems for wireless are pretty limited and shitty. But aren't you supposed to be editing leet
Each popular distro has a set of places to go for support. Some take a more hard line against things like ndiswrapper and mp3 support than others. Because each distro, while based on many of the same upstream sources, is still unique, each one requires its own unique set of common solutions. Ideally, the things Automatix fixes should be filed as bugs within Ubuntu rather than having to maintain a strict set of "you shouldn't do this if you live in the us or any other first world nation" guides and scripts. But I just hinted at what the real problem is. Unless you plan on purchasing and dedicating the various patents, I don't see how purchasing a liscence for all your users is a long term solution to encumbered software or open sourcing in general. This interview did a very piss poor job of getting ESR to answer the hard questions about what his advice means down the road.
Well, there's a couple of motivations behind that. First is the leveraged buyout, which was made rather popular in the 1980s before it degenerated into what's known as the "junk bond" scandal. Basically, you buy a company out, using the company you just bought as colateral. The financing behind it is much more complicated, but that's close to what it is underneath. Given the context of the situation (LBOs), this is mostly what they're referring to.
The other idea is that companies are supposed to seek above market returns. As long as the return their loan was spent on pays out better than the terms of the loan, it shouldn't not be an option. The problem is, as you've described, Microsoft is already making more cash than above market return ideas. The huge lump sum dividend is proof enough that they had no better place to invest it. And once you slash their R&D efforts and their practice of hiring good programmers just to keep them away from potential competitors, you'll have even more cash that you can't spend fast enough. Which may be what you meant when you said that the richest [public] company in the world needs to borrow. However, if part of the buyout means putting in new management with a much more aggressive set of ideas to pursue it might make sense. Primarily that would mean a lot more mergers and acquisitions, which is a really bad idea for a company that's already been subject to anticompetitive sanctions and still maintains a near monopoly.
I think it's along the lines of ISPs doing this without informing their customers what has happened. Their customers need not know that SBC extracted a heavy toll from YouTube or Google in order to deliver their video. And that even if you could know when your connection was tiered, no market offering would exist for an untiered connection. In other words, they're levying their massive subscriber base against people who profit from them having a decent internet connection, by holding it ransom. You'll note they aren't calling it anything like QoS, because that would imply that the offering has some level of reliability / quality.
I'm no lawyer, but this liscence seems awfully specific in all the wrong places. Lots of stuff specific to Microsoft, like suing them for patents etc. It's fine to sue users and derivate software distributers, just not MS. Moreover, it grants you liscence to use their copyrights and patents, so long as you grant a specific set of copyrights and patents ("you must grant all recipients the copyright and patent licenses in sections 2(A) & 2(B) for any file that contains code from the software"), likely including things you don't own. You don't have the authority to liscence Microsoft's patents to other people directly. I imagine that Microsoft intends that their offer is extended to anyone who might recieve this software down the chain, but it isn't made clear whether this is good enough. These two oddities combined basically means distributing the software, which MS inevitably holds a patent on, requires you to grant a liscence you don't own (and must seek from MS) and still allows them to sue to your users for infriging on their patent.
Try swinging your arm in circles for about a minute. Now try to justify how your arm, which is likely incredibly tired now, will be better able to hold up to playing Zelda sword fighting for 45 minutes to an hour. Sure, you can take short breaks, but:
a) who wants to stop playing Zelda?
b) zelda's design pratically requires 45 minutes of effort to accomplish anything significant. getting halfway through the water temple in ocarina for the first time, for example, was no mean feat of exploration and ingeniuity.
c) we really have no idea exactly how the swordfighting will be done, but if it is going to be something twitchy, it's not what consumers will be expecting. I guess this is a case of getting what you wished for.
What will be even worse is when the physicists and chemists gang up to pummel computer scientists for using 'atom' to describe a kind of data. :P
Whatever. Atoms were supposed to be indivisible, and they're still using the same word to describe something that contains both subatomic particles and sub-subatomic particles, to say nothing of what quarks might be made of
I don't think the grandparent was claiming that the majority of BT traffic was legal. Just that legitimate uses do exist. Unless you were trying to refute his claim that BT was legitimate. In which case, you did a pretty poor job. That argument is similar to blaming cars for the fact that the majority of highway traffic in urban areas breaks the speed limit. This brings up an interesting counter-question though: what does it mean for democracy when a majority of the population ignores a law?
I donno if Yoshi was all that bad. Down + B had just about the strongest priority in the game, which meant you could pwn just about anything from above. Certainly, doing anything other than blocking it was a feat of incredible timing.
This is just the continuation of the rhetorical split between GNU and Linux enthusiasts, of which ESR belongs to the linux camp. The cathedral and the bazaar, after all, was written primarily with GNU as the cathedral, and Linux as the bazaar. Of course, neither side's participants exclusively believe a certain way on the subjects around open source, but it's important to note that RMS's statement generally matches with Linus's, in that things like the nvidia binary driver, while not perfect, are allowed.
If you were truly one of the Annointed looking for the one true Free way, you'd already be familiar with this Linux situation and be running Hurd instead.
The real problem with flash drives is that you need to unmount them before you unplug them. Many, many people don't. And sometimes their data gets corrupted.
Thats not technically true. It's entirely possible to mount the drives as synchronous, so that writes are not buffered. Generally removing drives before requesting a buffer flush shouldn't be a problem unless you're copying a lot of information onto the drive. I suspect the same is true of your campus computer systems. In contrast most Linux distros I've seen prefer to mount drives async (turn on buffering) for the performance benefits on large writes. I've never seen data loss from surprise removal on Windows XP, and in fact the only time I've lost data is attempting to use the drive on a Mac OS 9's USB port that couldn't source the nessecary milliamps. The likelyhood of losing data from surprise removal in Windows seems pretty low unless your administrators have done something silly to the configurations.
The real kicker? They're replacing all the PCs in the campus labs with ones without floppy drives. So even those poor kids with only a few hundred KB of data will have to use a flash drive, and us student assistants will have to support them.
Frankly, this is a good thing. Floppies are shitty. They havent improved as a consumer technology since like 1987; they're prone to failure and painfully slow. Ideally, students using your campus labs should be using something like a network drive, which has the benefits of large storage, redundancy, protection against power failures, and periodic backups. But even if that isn't available, easily accomplished or part of student practices, USB drives are still a generally better alternative than floppies. I just bought a gigabyte of USB thumb drive for 20 dollars online. You can purchase 128M for under 10 dollars, and I suspect your campus bookstore could get a bulk discount to get them even cheaper (whether they decide to share that with students is best rated "unlikely"). Point is, for about the same price as a pack of disks you could get a hell of a lot more storage that runs faster, less prone to failure, and easier to carry around. And if your school wasn't backwards ass, students would have easy access to network storage, both from home and on campus.
Google searching suggests that there are removal tools for that u3 software stuff. I'm not sure if they work or if they'll work for you in an environment where no software is allowed to be installed. It might be easier to file a trouble report asking for the u3 drivers to be installed (assuming this isn't really device/vendor dependent). AFAIK, u3 only requires drivers for programs on the drive and password protected volumes.
I don't recall class sizes of 15 at my own well funded suburban school district, but that's neither here nor now. I realize teachers definately have a severe budget constraint, which is probably why giving free stuff pulls in a lot of clicks from teachers. Certainly, I'd expect the best public school teachers to be masters of getting something for nothing. But I think the author's banking on the idea that if a teacher finds the budget for it, they'll eventually use it as more than just a one-off activity. Certainly, my fourth grade teacher used some word search software for students on a weekly basis for spelling / vocabulary words.
But figuring out where these open source projects will come from is going to be a challenege. I think there's a lot of untapped potential here within schools. Most of these programs are simple to write if you can write software at all. It would be interesting to see high school students do an advanced / 2nd programming course duplicating the functionality of such a program at the request of a local teacher. Of course, it would require students to be taught a suitable language in the first place, so qbasic might be out, no matter how fun Gorillas and Nibbles are. You could even put the results on sourceforge and might get help from Edubuntu if they want to include the software.
Shit, you're right. I remember that too, now. All it takes to be moderated "+5, Informative" is attention to grammar and spelling, and long paragraphs, I guess. Still, it will be interesting to see how Nintendo's previous designs and considerations color !!M.
Bad news, we voted out the evolutionist school board. We're not stupid, we're just not ever vigilant at the ballot box. But you might try moving to Las Vegas: "God's Blind Spot."
Was this nugget of insight any less valid reguarding the now defunct 939 slot, or the soon to be released AM3 slot?
I doubt Nintendo will offer an IM service to players. It's useful and lets players build communities, but the potential for Internet preditory practices is something Nintendo understandably wants to avoid, at the expense of building an online community. Look at the awkward systems on the DS. You can't communicate with people you haven't already met in the real world. And even if you do get their number through outside means like a gamefaqs forum, few games allow serious conversations. Mario kart provides no method of communication at all. Metroid Prime only allows you to talk with people you've met before, and only during the time for choosing game options.
What this embargo on interplayer communication means is team oriented play is nigh impossible. You'd have to do something similar to the iconographic messages used in Phantasy Star Online, which was primarily designed to let people of different languages play together. Building up a meaningful vocabulary like that will take a lot of investment from both the designer and the players.
Nevermind the occasional Linux ISO over BT. 20-30 gigabytes is chump change compared to the cost of a disk to store that data on. Instead, think Naruto. With over 2 thousand simultaneous users (and this is almost two weeks old!), it's likely a valuable net gain for them there. And if they go ahead with it, usage would spike even higher; many users find BT throttled or simply slow; if suddenly you were maxxing out that "6 mbps" line the cable company sold you for BT, I don't think you'd bother with DCC bots anymore. Scarywater used to host linux stuff on BT, but it's dwarfed by the sheer scale of their anime tracker.
But if ISPs are going to cache files like this, it pretty much eliminates the point of bittorrent.