...in other words, a ratio similar to non-software patents.
I strongly disagree. The small company I founded has some hardware patents in the structured-ASIC space. These patents attracted investors who otherwise probably would have passed us by. Our little space would not even have been explored, had it not been for hardware patents. Software patents generally have far less importance for small startups. Our hardware patents aren't stopping anyone else from innovating (the way software patents do), since no one else would even bother with the expense of exploring this space without a patent.
A programmer generally doesn't need software patents to attract investors. He can often build the first product version himself. As the article suggests, most innovation in software is a small incremental improvement on existing work, which typically can be implemented cheaply and quickly, without requiring outside investors, or the potential of owning a market for several years. Software patents often have the potential to restrict innovation, as it is quite likely that some random person will want to not only implement the innovation on their own, but improve upon it.
I'm not sure I agree... surely the popular works don't need long-term copyrights, but the vast majority of writers, musicians, and artists live on very low incomes, have no savings for retirement, and will need any tiny royalty that still comes in during their old age. However, the same argument in no way translates to software. Software naturally has a useful lifetime of only several years, and then it needs to be upgraded.
Split keyboards do help, but they aren't the only way... what they let you do is keep your fingers on the home row without turning your wrists. The other way to stop turning your wrists is to give up on the home row, and turn your hands at an angle to a normal flat keyboard. The positive effects for your hands are about the same either way. You can still rest your pointer fingers on the 'f' and 'j' keys, but with wrists turned, your other fingers drift into the next row.
But I basically agree with you. Getting a split keyboard is a very effective improvement that wont make you relearn how to type.
Agreed. Copyrights lasting 100 years is just silly. I can see reasonable arguments for copyrights lasting the lifetime of an artist or writer, but who in history has ever copyrighted a valuable work, and lived another 100 years? Why do programmers need many years of copyright protection? Since we update the stuff every year, the copyright's always renewed, and if we stop updating it, it's typically no longer worth money.
I agree... I don't tend to bash linux distros, but Novel has gone evil on us. The funny thing is that M$ is suing Novel anyway, through their newest front for patent suits.
Gee... I already have most of those cool new features, and ever time my OS ships "The greatest upgrade ever", it costs me nothing... man, I love linux, and Ubuntu/Debian in particular.
If he'd run Ubuntu instead of Suse, chance are the programs he wants would already be available through apt-get/Synaptic. Back when I used Fedora, I had to install tons of apps from source. Now that I use Ubuntu, it's only rare exceptions. You gotta love all the work those Debian guys did, even if we call it Ubuntu now days:-)
As a country develops it's own intellectual property, it becomes important to protect it. For example, India seems to be moving towards proper copyright protection. However, this article is specifically about software patents, not copyrights. Of course software copyrights need to be protected. The authors need to have the rights to do whatever they want with their software, and it should not be stolen from them. Software patents are different. They don't protect your software, but restrict me from writing equivalent software. I can't recall a single example in history where a software innovator was rewarded by their patent - there are probably a few, but they are vastly outnumbered by the software patents that have restricted innovation. Instead of rewarding innovation, software patents make it possible for M$ to attack Linux, while doing nothing to help innovators break into the software market dominated by M$. Don't get me wrong... I'm a big supporter of M$ in general, but lets face it: software patents exist to protect M$ and other huge calcified players, rather than reward innovators.
Jesus, guys! How is it that a bunch of super-smart slashdot geeks can be so incredibly in-the-dark about something very likely to affect them... It's kind of like sex/puberty and reading glasses/turning 40. No one seems to know/care what sex is about until puberty, and for some reason people don't understand that the lenses in their eyes solidify when they turn 40 (losing the ability to focus).
Emacs damages your unlar nervers (your "funny" bones). You can feel them in your elbows. They connect to your pinkies. Emacs wears them out. I blew mine, and had to program by voice for three years while they healed. During that time, even typing my password was painful, as was picking up the damned shampoo bottle in the shower, or a beer after work. Steering a car or holding my girl-friend's hand could be very painful.
Neo-conservative "scientists" keep on proving all kinds of great things: global warming is a liberal plot to control humanity; AIDS is not caused by HIV; second-hand smoke is not bad for you; and computer related hand injuries are a figment of your imagination. Do yourself a favor and do the following:
- When your hands hurt stop typing. Trust me on this. Please do it. - Switch the mouse to your off-hand for a while, but when your hands hurt, stop typing. - Use Dragon Naturally Speaking, and dictate your next paper. - Switch from Emacs to VI if your pinkies hurt. - Learn the right way to type... limp hands, small finger motions, never twist your wrists. You don't need a new keyboard, just a better way of typing. - Get a laptop. The keys don't hurt your hands as much.
Emacs trashed my hands, but without it I couldn't have controlled my computer by voice for three years. By the time I recovered, I had over 1,600 Emacs voice-macros for controlling everything on my computer. It trashed my hands, and then enabled me to keep my job. We programmers often suffer from ulnar nerve problems... we're lucky. CTS is more permanent, and just worse in general. Still, do yourself a favor, and lay off the keyboard when your hands hurt. I'm sorry for repeating it, but from experience, I know that none of you will listen to that advice, unless shouted at you night and day. We're just weird that way, both you and me.
I think being open for development isn't enough, you also need to spend a couple $100 million or so in marketing...
I think what you really need is best-of-class hardware, open development, and most of all - great software, which is where the iPhone really shines.
Given the importance of software, you'd think stupid Steve would be smart enough to allow the millions of hackers out here to help him expand his lead. Instead, we're gonna help the other guys tear him down. I smashed my iPhone to pieces at the Apple store after Steve borked it. He's seriously anti-hacker. My Neo1973 shows up today, and I'm gonna use spare cycles to make cool apps you can no longer run on iPhone - like an e-book reader, and maybe help out with the GPS apps, or dial-by-voice.
To really knock out iPhone, the default OS has to rock. I'd guess that it will only take a few dozen coders 6-9 months to pull this off... a team Google can easily afford. Here's my wish-list for OpenMoko. Same goes for gPhone. Core functionality like this would leave iPhone in the dust... and if it's open to hackers, it'll rock the world.
Have you seen the OSS out there? If you're worried about how your code will look, obviously, you haven't looked at what's out there. Unless your code is going to come alive on it's own, creep out of the computer, and strangle me in my sleep, it cannot possibly be the worst code I've read. Get over it, and post it:-)
Yes, it just covers multiple workspaces. So, if Gnome and KDE just drop the very-cool workspace switcher, the problem goes away. Also, this patent seems to expire this year (it was filed in 1987, and granted in 1991), so we would only be without our cool workspace switcher for a few months. Not much here, really.
I agree. Ever since Bush has taken office, the US has been a shinning example of democracy, human rights, and... hahahahaha!!!! Ok, I couldn't quite say that with a straight face:-)
Oh, come on... that's like not talking about God at Church. There's lots more insight for us all to gain, even if we are a bunch of calcified stuck-in-their-ways old farts:-)
Yes, I take it back. In fact, I've just purchased the OpenMoko Neo1973, and hope to be of some assistance in the open-source development community, which so far seems vibrant and friendlier than most:-)
Exactly. I did a lot of searching, and it seems that OpenMmoko is the only current significant effort at open-phone development. Apple and the rest run on *nix, but close up the phone so you can't do dick with it. Openmoko has some promise, but without wireless or a cell carrier in the US on-board, I'm not read to start hacking it. Ubuntu Mobile has potential, but the screen size currently has to be 4.8", and it looks more like a tablet PC OS at the moment than smartphone software.
There's some ultra-smart dudes at Google, at this point. The next major shift in computing will be smartphones. Only an open system with an excellent SDK for 3rd party applications has much chance of dominating, so the current players seem to be Google and Microsoft. I personally have disliked every version of Windows CE I've ever seen, though I hear good things about the latest version. Based on open-source GNU/Linux, Google's got a real shot at the largest new market on the horizon, IMO.
Yes, you are quite right. I was nearly dead broke on my last trip to England, and the cheapest items on the menu kept coming up as bangers and eggs! I've done fairly well in the years since, and on the trip to Italy, we were able to avoid the cheaper pizza bars, and experience the really good food Italy has to offer. Still... one only has their own experiences to draw on:-)
I am planning on starting a company of that sort next year in my country. I will let you know how it goes, if you want.
I'm always interested in how founders are doing, so please do keep me informed. You have to pass the challenge question (what color is the sky?), but my public e-mail is bill@billrocks.org. I founded a small company back in 2000, and I can't complain, though we're no Google or Yahoo. Actually, we're tiny, but it still delivers what I need and I still have big hopes and dreams. I think there's still tons of room for innovation, but business models need to keep changing. Areas like VoIP seem fertile for small businesses (see David Rowe's awesome Free Telephony Project). P2P has some gas left in it. In hopefully the not far distant future, we'll see the birth of self-replicating hardware, and I see that creating all kinds of need for designers. I also think the iPhone shows that in the future we will not be tied to M$ for mobile computing products, and there's lots of room for innovation in that direction. I'm anxiously awaiting a real OS on a smartphone, like Ubuntu Mobile.
My significant other switched from being CEO of a small company being a writer back in 2002, when the economy was in the tank. Now she's building another company around small-volume custom publishing, taking advantage of the changing trends in print media. She had fun being a pure writer while it lasted, but where's the money? She pays $25 for an article from 2nd-tier local writers, and $125 from the top-tier. She has unpaid interns who just want the experience so that one day they can be paid writers. Even as CEO of a small publishing company, she's making far less than she did as a regular employee at her previous high-tech job.
I feel for writers, but their not the only ones feeling the squeeze. This morning I came up with a fairly depressing argument about where the new startups are in high-tech: practically nowhere. If you want to do a startup making chips, forget it. If you build digital, then FPGAs, microcontrollers, and DSPs have it covered on the low-end, and digital high-end ASICs are too damned expensive. Analog is just too hard, though there's some room there. If you want to do a startup in software, you've got Microsoft dominating the market, and tons of free open-source to compete with. What's that leave? The web. The big successes that quickly come to mind for new high-tech companies over the last 15 years are Yahoo, Amazon, Ebay, and Google. Not software, not chips, but something else entirely. Since all those companies started back before the web bubble burst, what's left for us geeks now?
Oddly, it's kind of scary the parallels I can draw between sex and the Internet... *shudder*
That's what I love about shashdot... I'm not the only geek here who sees weird stuff like that. Anyway, I'm not gonna bash g-mail, and in fact have recently created my first account there. Yahoo stopped offering free POP download, so what do you do? G-mail is free, and seems to work just fine. As for privacy, it all gets piped to the NSA anyway. I don't think it would be possible to violate my privacy more without posting nude pictures of me on the net, and given the current state of my typical hacker-geek body, even that wouldn't count for much.
Spotted Dick has to be my all-time favorite popular British dish I have no desire to eat. Last time I was there I got bangers and eggs for breakfast, about as bland as it gets. So, for lunch I tried a nice sounding pita-bread thing which turned out to be sliced up bangers mixed with eggs in a pita. No matter how hard I tried to avoid ordering bangers and eggs, I kept on getting it. Now my trip to Italy... totally different! Yummy food everywhere, so long as I avoided pizza bars and tourist traps!
New ideas are almost universally shot down until you have it implemented in a product you own. Usually, I find that smart people grip about the most obvious potential flaws without looking into proposed solutions. Even I tend to fall into that trap. Those of us accustomed to routinely proposing new ideas we just come up with have to have very thick skins.
As for specific answers: There's no assumption about everyone running the same version. Each file would have a SHA1 code used to locate it, and it would be downloadable from others who have it, regardless of what their system looks like. It would be a bit like the git version control system that way - content addressable, with a SHA1 representing the exact state of the root directory and it's sub-dirs. Local changes would override the defaults, so for example your computer would have the correct drivers installed locally. When you execute XEmacs the first time, essentially the package would be installed at that time, and browsing it's lisp libraries would only download the corresponding directory entries, not the files themselves. However, installing XEmacs would happen lightning-fast, as only a tiny fraction of XEmacs is actually used most of the time, thus reducing one of XEmac's biggest problems: a huge disk footprint. Same think with X11 and KDE. The files are all virtually there, but contents download when accessed. If a user selected KDE as his default, it would very quickly download, as only the new desktop files and executables would be accessed. Most of the bloat would never need to be download.
Also, this is just one use case. I've listed three others on the web site. For example, BitTorrent is great for sharing huge files, and enables small-fry outfits to distribute many gigabytes of data to the whole world, almost for free. However, it sucks for collections of small files that tend to update over time. So, NetFS could be used for pubishing video blogs, or other often-changing content large enough to be a problem for one guy to upload, but too small and/or changing too fast for BitTorrent. Sites like sourceforge.net and gutenberg.org also require dotated mirrors to function effectively, and NetFS could potentially eliminate that need, while saving them money and improving download speeds.
Anyway, as I said, new ideas are generally shat upon. The reaction of the BitTorrent forums to the friends-downloading algorithm was a good example... The forums seemed to feel that there is something fundamentally wrong with having friends who can help you download faster. However, if implemented, BitTorrent users would generally be far better off. Another example of mine is DataDraw a CASE tool for automating high performance in-memory persistent database creation. It's in use at the last four companies I've worked at, and those who've used it give it glowing recommendations. However, getting smart guys to use someone else's CASE tools requires a world-wide marketing effort, and even then you typically fail.
I guess it's kind of like Linux... great stuff, the best in the world. However, don't expect the world to stop using Windows any time soon, no matter how good Linux is. In general I try not to worry about the rest of the world and focus on my own needs. Linux is the best operating system ever created for my needs, by far. I take great joy in using it, and no longer care what Joe Sixpack uses to download his porn and play games. Same thing with DataDraw - I develop faster, more readable code in less time, and work more effectively with my co-workers. If 99.99% of the hackers out there want to continue writing crap in raw C, or even worse - C++, so be it. To properly enjoy open-source creation, you kinda have to stop worrying about your project's true potential in the world, and focus on your own needs.
Slashdot Mirror
I think we basically agree. I think software should not have the same copywrites as art. We should allow them to be different.
I strongly disagree. The small company I founded has some hardware patents in the structured-ASIC space. These patents attracted investors who otherwise probably would have passed us by. Our little space would not even have been explored, had it not been for hardware patents. Software patents generally have far less importance for small startups. Our hardware patents aren't stopping anyone else from innovating (the way software patents do), since no one else would even bother with the expense of exploring this space without a patent.
A programmer generally doesn't need software patents to attract investors. He can often build the first product version himself. As the article suggests, most innovation in software is a small incremental improvement on existing work, which typically can be implemented cheaply and quickly, without requiring outside investors, or the potential of owning a market for several years. Software patents often have the potential to restrict innovation, as it is quite likely that some random person will want to not only implement the innovation on their own, but improve upon it.
I'm not sure I agree... surely the popular works don't need long-term copyrights, but the vast majority of writers, musicians, and artists live on very low incomes, have no savings for retirement, and will need any tiny royalty that still comes in during their old age. However, the same argument in no way translates to software. Software naturally has a useful lifetime of only several years, and then it needs to be upgraded.
Split keyboards do help, but they aren't the only way... what they let you do is keep your fingers on the home row without turning your wrists. The other way to stop turning your wrists is to give up on the home row, and turn your hands at an angle to a normal flat keyboard. The positive effects for your hands are about the same either way. You can still rest your pointer fingers on the 'f' and 'j' keys, but with wrists turned, your other fingers drift into the next row.
But I basically agree with you. Getting a split keyboard is a very effective improvement that wont make you relearn how to type.
Agreed. Copyrights lasting 100 years is just silly. I can see reasonable arguments for copyrights lasting the lifetime of an artist or writer, but who in history has ever copyrighted a valuable work, and lived another 100 years? Why do programmers need many years of copyright protection? Since we update the stuff every year, the copyright's always renewed, and if we stop updating it, it's typically no longer worth money.
I agree... I don't tend to bash linux distros, but Novel has gone evil on us. The funny thing is that M$ is suing Novel anyway, through their newest front for patent suits.
Gee... I already have most of those cool new features, and ever time my OS ships "The greatest upgrade ever", it costs me nothing... man, I love linux, and Ubuntu/Debian in particular.
If he'd run Ubuntu instead of Suse, chance are the programs he wants would already be available through apt-get/Synaptic. Back when I used Fedora, I had to install tons of apps from source. Now that I use Ubuntu, it's only rare exceptions. You gotta love all the work those Debian guys did, even if we call it Ubuntu now days :-)
As a country develops it's own intellectual property, it becomes important to protect it. For example, India seems to be moving towards proper copyright protection. However, this article is specifically about software patents, not copyrights. Of course software copyrights need to be protected. The authors need to have the rights to do whatever they want with their software, and it should not be stolen from them. Software patents are different. They don't protect your software, but restrict me from writing equivalent software. I can't recall a single example in history where a software innovator was rewarded by their patent - there are probably a few, but they are vastly outnumbered by the software patents that have restricted innovation. Instead of rewarding innovation, software patents make it possible for M$ to attack Linux, while doing nothing to help innovators break into the software market dominated by M$. Don't get me wrong... I'm a big supporter of M$ in general, but lets face it: software patents exist to protect M$ and other huge calcified players, rather than reward innovators.
Jesus, guys! How is it that a bunch of super-smart slashdot geeks can be so incredibly in-the-dark about something very likely to affect them... It's kind of like sex/puberty and reading glasses/turning 40. No one seems to know/care what sex is about until puberty, and for some reason people don't understand that the lenses in their eyes solidify when they turn 40 (losing the ability to focus).
Emacs damages your unlar nervers (your "funny" bones). You can feel them in your elbows. They connect to your pinkies. Emacs wears them out. I blew mine, and had to program by voice for three years while they healed. During that time, even typing my password was painful, as was picking up the damned shampoo bottle in the shower, or a beer after work. Steering a car or holding my girl-friend's hand could be very painful.
Neo-conservative "scientists" keep on proving all kinds of great things: global warming is a liberal plot to control humanity; AIDS is not caused by HIV; second-hand smoke is not bad for you; and computer related hand injuries are a figment of your imagination. Do yourself a favor and do the following:
- When your hands hurt stop typing. Trust me on this. Please do it.
- Switch the mouse to your off-hand for a while, but when your hands hurt, stop typing.
- Use Dragon Naturally Speaking, and dictate your next paper.
- Switch from Emacs to VI if your pinkies hurt.
- Learn the right way to type... limp hands, small finger motions, never twist your wrists. You don't need a new keyboard, just a better way of typing.
- Get a laptop. The keys don't hurt your hands as much.
Emacs trashed my hands, but without it I couldn't have controlled my computer by voice for three years. By the time I recovered, I had over 1,600 Emacs voice-macros for controlling everything on my computer. It trashed my hands, and then enabled me to keep my job. We programmers often suffer from ulnar nerve problems... we're lucky. CTS is more permanent, and just worse in general. Still, do yourself a favor, and lay off the keyboard when your hands hurt. I'm sorry for repeating it, but from experience, I know that none of you will listen to that advice, unless shouted at you night and day. We're just weird that way, both you and me.
I think what you really need is best-of-class hardware, open development, and most of all - great software, which is where the iPhone really shines.
Given the importance of software, you'd think stupid Steve would be smart enough to allow the millions of hackers out here to help him expand his lead. Instead, we're gonna help the other guys tear him down. I smashed my iPhone to pieces at the Apple store after Steve borked it. He's seriously anti-hacker. My Neo1973 shows up today, and I'm gonna use spare cycles to make cool apps you can no longer run on iPhone - like an e-book reader, and maybe help out with the GPS apps, or dial-by-voice.
To really knock out iPhone, the default OS has to rock. I'd guess that it will only take a few dozen coders 6-9 months to pull this off... a team Google can easily afford. Here's my wish-list for OpenMoko. Same goes for gPhone. Core functionality like this would leave iPhone in the dust... and if it's open to hackers, it'll rock the world.
I suspect Bill's guy forgot to offer an appropriate bribe. What does a typical crooked immigration guy expect from Bill Gates? $100? $1M?
Have you seen the OSS out there? If you're worried about how your code will look, obviously, you haven't looked at what's out there. Unless your code is going to come alive on it's own, creep out of the computer, and strangle me in my sleep, it cannot possibly be the worst code I've read. Get over it, and post it :-)
Yes, it just covers multiple workspaces. So, if Gnome and KDE just drop the very-cool workspace switcher, the problem goes away. Also, this patent seems to expire this year (it was filed in 1987, and granted in 1991), so we would only be without our cool workspace switcher for a few months. Not much here, really.
I agree. Ever since Bush has taken office, the US has been a shinning example of democracy, human rights, and... hahahahaha!!!! Ok, I couldn't quite say that with a straight face :-)
Oh, come on... that's like not talking about God at Church. There's lots more insight for us all to gain, even if we are a bunch of calcified stuck-in-their-ways old farts :-)
Yes, I take it back. In fact, I've just purchased the OpenMoko Neo1973, and hope to be of some assistance in the open-source development community, which so far seems vibrant and friendlier than most :-)
Exactly. I did a lot of searching, and it seems that OpenMmoko is the only current significant effort at open-phone development. Apple and the rest run on *nix, but close up the phone so you can't do dick with it. Openmoko has some promise, but without wireless or a cell carrier in the US on-board, I'm not read to start hacking it. Ubuntu Mobile has potential, but the screen size currently has to be 4.8", and it looks more like a tablet PC OS at the moment than smartphone software.
There's some ultra-smart dudes at Google, at this point. The next major shift in computing will be smartphones. Only an open system with an excellent SDK for 3rd party applications has much chance of dominating, so the current players seem to be Google and Microsoft. I personally have disliked every version of Windows CE I've ever seen, though I hear good things about the latest version. Based on open-source GNU/Linux, Google's got a real shot at the largest new market on the horizon, IMO.
Yes, you are quite right. I was nearly dead broke on my last trip to England, and the cheapest items on the menu kept coming up as bangers and eggs! I've done fairly well in the years since, and on the trip to Italy, we were able to avoid the cheaper pizza bars, and experience the really good food Italy has to offer. Still... one only has their own experiences to draw on :-)
Last I read, they'd created regenerating mice, super-long-lived mice, and fearless mice. Combining the together... yup: Mighty Mouse :-)
I'm always interested in how founders are doing, so please do keep me informed. You have to pass the challenge question (what color is the sky?), but my public e-mail is bill@billrocks.org. I founded a small company back in 2000, and I can't complain, though we're no Google or Yahoo. Actually, we're tiny, but it still delivers what I need and I still have big hopes and dreams. I think there's still tons of room for innovation, but business models need to keep changing. Areas like VoIP seem fertile for small businesses (see David Rowe's awesome Free Telephony Project). P2P has some gas left in it. In hopefully the not far distant future, we'll see the birth of self-replicating hardware, and I see that creating all kinds of need for designers. I also think the iPhone shows that in the future we will not be tied to M$ for mobile computing products, and there's lots of room for innovation in that direction. I'm anxiously awaiting a real OS on a smartphone, like Ubuntu Mobile.
My significant other switched from being CEO of a small company being a writer back in 2002, when the economy was in the tank. Now she's building another company around small-volume custom publishing, taking advantage of the changing trends in print media. She had fun being a pure writer while it lasted, but where's the money? She pays $25 for an article from 2nd-tier local writers, and $125 from the top-tier. She has unpaid interns who just want the experience so that one day they can be paid writers. Even as CEO of a small publishing company, she's making far less than she did as a regular employee at her previous high-tech job.
I feel for writers, but their not the only ones feeling the squeeze. This morning I came up with a fairly depressing argument about where the new startups are in high-tech: practically nowhere. If you want to do a startup making chips, forget it. If you build digital, then FPGAs, microcontrollers, and DSPs have it covered on the low-end, and digital high-end ASICs are too damned expensive. Analog is just too hard, though there's some room there. If you want to do a startup in software, you've got Microsoft dominating the market, and tons of free open-source to compete with. What's that leave? The web. The big successes that quickly come to mind for new high-tech companies over the last 15 years are Yahoo, Amazon, Ebay, and Google. Not software, not chips, but something else entirely. Since all those companies started back before the web bubble burst, what's left for us geeks now?
That's what I love about shashdot... I'm not the only geek here who sees weird stuff like that. Anyway, I'm not gonna bash g-mail, and in fact have recently created my first account there. Yahoo stopped offering free POP download, so what do you do? G-mail is free, and seems to work just fine. As for privacy, it all gets piped to the NSA anyway. I don't think it would be possible to violate my privacy more without posting nude pictures of me on the net, and given the current state of my typical hacker-geek body, even that wouldn't count for much.
Spotted Dick has to be my all-time favorite popular British dish I have no desire to eat. Last time I was there I got bangers and eggs for breakfast, about as bland as it gets. So, for lunch I tried a nice sounding pita-bread thing which turned out to be sliced up bangers mixed with eggs in a pita. No matter how hard I tried to avoid ordering bangers and eggs, I kept on getting it. Now my trip to Italy... totally different! Yummy food everywhere, so long as I avoided pizza bars and tourist traps!
New ideas are almost universally shot down until you have it implemented in a product you own. Usually, I find that smart people grip about the most obvious potential flaws without looking into proposed solutions. Even I tend to fall into that trap. Those of us accustomed to routinely proposing new ideas we just come up with have to have very thick skins.
As for specific answers: There's no assumption about everyone running the same version. Each file would have a SHA1 code used to locate it, and it would be downloadable from others who have it, regardless of what their system looks like. It would be a bit like the git version control system that way - content addressable, with a SHA1 representing the exact state of the root directory and it's sub-dirs. Local changes would override the defaults, so for example your computer would have the correct drivers installed locally. When you execute XEmacs the first time, essentially the package would be installed at that time, and browsing it's lisp libraries would only download the corresponding directory entries, not the files themselves. However, installing XEmacs would happen lightning-fast, as only a tiny fraction of XEmacs is actually used most of the time, thus reducing one of XEmac's biggest problems: a huge disk footprint. Same think with X11 and KDE. The files are all virtually there, but contents download when accessed. If a user selected KDE as his default, it would very quickly download, as only the new desktop files and executables would be accessed. Most of the bloat would never need to be download.
Also, this is just one use case. I've listed three others on the web site. For example, BitTorrent is great for sharing huge files, and enables small-fry outfits to distribute many gigabytes of data to the whole world, almost for free. However, it sucks for collections of small files that tend to update over time. So, NetFS could be used for pubishing video blogs, or other often-changing content large enough to be a problem for one guy to upload, but too small and/or changing too fast for BitTorrent. Sites like sourceforge.net and gutenberg.org also require dotated mirrors to function effectively, and NetFS could potentially eliminate that need, while saving them money and improving download speeds.
Anyway, as I said, new ideas are generally shat upon. The reaction of the BitTorrent forums to the friends-downloading algorithm was a good example... The forums seemed to feel that there is something fundamentally wrong with having friends who can help you download faster. However, if implemented, BitTorrent users would generally be far better off. Another example of mine is DataDraw a CASE tool for automating high performance in-memory persistent database creation. It's in use at the last four companies I've worked at, and those who've used it give it glowing recommendations. However, getting smart guys to use someone else's CASE tools requires a world-wide marketing effort, and even then you typically fail.
I guess it's kind of like Linux... great stuff, the best in the world. However, don't expect the world to stop using Windows any time soon, no matter how good Linux is. In general I try not to worry about the rest of the world and focus on my own needs. Linux is the best operating system ever created for my needs, by far. I take great joy in using it, and no longer care what Joe Sixpack uses to download his porn and play games. Same thing with DataDraw - I develop faster, more readable code in less time, and work more effectively with my co-workers. If 99.99% of the hackers out there want to continue writing crap in raw C, or even worse - C++, so be it. To properly enjoy open-source creation, you kinda have to stop worrying about your project's true potential in the world, and focus on your own needs.