Re:Virtual Reality must be Free and Decentralized
on
Metaverse Launched?
·
· Score: 1
Well, I would argue that most people's perception of their everyday surroundings focuses mostly on specific locations like home, work, school, shopping malls, etc. The commute between those places is boring and frequently deliberately blocked out (say listening to the radio rather than admiring the scenery). People have had little problems adapting to the web, which is nothing like the "learned experience" of a book or newspaper, so why does it have to be different for 3D spaces?
Well, we've been deliberately keeping a low profile -- Slashdot used to be notorious for posting free software projects that were just as much vaporware as the latest thing from Microsoft. We really don't want to do that, so we're trying to have a really complete application together before working on the "marketing" side of things. That said, we're quite far along in some ways and there are certainly plenty of other ways we could use some help:-) Also I doubt we have the bandwidth to stand up to a slashdotting...
Re:I can't see this flying.. centrally controlled
on
Metaverse Launched?
·
· Score: 1
See
my comment on this. I agree, commercial ventures are at best uninteresting and at worst highly suspect. Of all free projects that I've found (and I've looked around quite a lot) I think the Interreality project is by far the most advanced in terms of architechture, features, and vision for Virtual Reality on the Internet.
Virtual Reality must be Free and Decentralized
on
Metaverse Launched?
·
· Score: 2, Informative
The fundamental problem I have with the numerous attempts at building VR systems (aside from the fact that they are usually quite clunky and boring to use for anything besides basic chatting) is that these systems are almost universally based on large-scale central servers, rather than networks of small sites. Consider the model of the world wide web: anyone with a little bit of connectivity and bandwidth can host their own web site. Why shouldn't VR be the same way? Why do we chain ourselvers to monolithic, commercially controlled world servers rather than a community of interlinked VR rooms? The underlying technology has subtle social effects as well: would you rather have an autocratic dictatorship of a "planned" world or a democratic community where anyone can add their own pieces to the world?
The second problem (that has been noted by many people) is that 3DVR chatting has been done many, many times and is fairly uninteresting. To make this usefull, we need some big ideas about how we can bring existing applications into a 3D space and have them work more effectively, as well as completely new approaches which are only possible in an immersive environment. If VR is to be useful, it should be a ubiquitus application sitting in the background, one of many which helps you achive your tasks (work or play) rather than the elephant application that demands constant attention.
I have been leading my own free software project to build such a system: The Interreality Project and Virtual Object System (VOS). We have built a new protocol infrastructure to support distributed virtual objects (which are used to build virtual worlds) and a client program using the Crystal Space 3D engine. All of our software in available under the GPL or LGPL, and we are on our tenth public release (with a new release planned for the next couple weeks). The system has been in use at the University of Massachusetts, Amherst Laboratory for Perceptual Robotics for over a year, we have several outside developers, and there seems to be quite a bit of interest in our little project (the web site got 55,000 hits last month).
Our vision is a 3D companion to the web, but not in the (incredibly stupid) sense of putting 3D objects on web pages, but a highly interlinked ecosystem of small information resources -- which happen to convey information in 3D rather than mere 2D. Of course, multiuser support is also a fundamental part of this system -- if the world is dynamic, then any part can move and change and communicate, users, bots, agents and applicances. Perhaps the real key is that we must strive for a system that reflects the nature of the beast, the nature of the Internet, rather than trying to emulate the real world (badly).
The Interreality Project - Open Source P2P 3D
on
P2P Roaming Chat
·
· Score: 1
The Virtual Object System / Interreality Project is a free software (GPL/LGPL) effort to build the software infrastructure and applications for distributed peer-to-peer virtual reality. The basic model of enabling users to host their own virtual worlds is the same as BrendanWorld, but our software supports 3D, is open source, cross platform (GNU/Linux, MacOS X and Windows) and built upon a powerful, extensible generic base. We're pretty far along in some ways, but the reason you haven't heard of us is that we've been keeping a pretty low profile so far, at least until we get our documentation up to speed. We do welcome new developers, so please come check us out!
The one major impression that I get from using Galeon is that it's clearly a browser that the authors wrote with the intention of USING it. The interface is simple and efficient, very configurable, and the bookmark handling has a number of features I haven't seen anywhere else. Their slogan is "The web. Only the web." which pretty much sums up the apparent design philosophy: do one thing, and do it well. It's definately worth a try.
I actually work with the Wearable Computing Group at the University of Massachusetts Amherst. Most of our research right now is done on wearables from Xybernaut, the company that Hitachi is partnering with as mentioned in the press release. The Xybernaut systems we have now are regular p200s (192MB RAM, 4GB disk, serial/PS2/USB ports, pretty much all the stuff you'd find on a laptop) running Red Hat 6.2. When we got them, they cost about $4000.
The biggest gripe most people have about them (aside from their general bulk) is the awful Head Mounted Display (HMD) they use. It works by using a small 640x480 LCD display pointing away from you with uses a concave mirror to reflect the image back into your eye. There are a lot of problems with them - it's very hard to get the entire screen in focus and visible, you have this big arm holding the display in front of your face, and it's almost useless in sunlight. There's also the privacy issue of the fact that anyone can just look at the display itself (which, like I said, points away from you) and see a horizontally flipped image of whatever you're currently viewing.
A useful wearable device, almost more than anything, has to have a display that is easy to wear, small, and unobtrusive. I'm curious as whether this Hitachi device will achive that to a better extent than the current Xybernaut HMD.
Something very few people bring up is the fact that the university is also a research environment. There are lots of smart people here doing cutting-edge thinking about software and hardware, and by attending school you have a chance to talk to these people and get involved. Yes, you can go to college and just take the classes you need to get by the major, but if you take some time to make friends with the professors and get involved in what they are doing, you'll learn far more than if you had spent four years writing perl scripts for a web site.
You'll also have a chance, in addition to straight computer science to take a lot of higher math courses. Calculus, linear algebra, statistics and discrete mathematics (set theory, predicate logic, finite state autonoma among others) are all quite important to know for many software applications.
It's also an excelent opportunity to get involved in student groups that do totally non-technical things, like environmental activism, cultural groups, or even trying to legalize marijuana:-) You'll learn a lot about working with other people doing these sorts of things, and that's incredibly important once one gets out into the job market.
I'm rather amused by the two main sets of comments here on/. - the "if they could pull this off, it would be really cool" crowd and the "rewriting windows is an impossible venture" crowd.
I'm not entirely convinced either way. Idealistically it's probably a good thing, more free software, and just maybe they'll make it. On the other hand, technically, we're talking about a project that's going to take years. Three or four, and probably longer. You have to implement a multitasking kernel, GDI, C libraries, multimedia libraries, system tools, and a hundred other little and big things. Against a moving target, no less. Is this project useful?
This project will be useful to the programers on it. They'll probably learn something about programming and software engineering. Will this project be useful to other people? Maybe one day, but not for a long time. Don't hold your breath.
Twofish is not an asymmetric algorithm! RSA and ElGamal are. Otherwise you're correct, asymmetric algorithms use a key pair where one key can be used to encrypt and the other decrypt. RSA can use either the public or private key to encrypt and the opposite one decrypts, whereas ElGamal can only do one-way encryption. ElGamal is a bit faster than RSA and has been proven to be at least as strong as RSA (and perhaps a bit stronger) but suffers from a couple major drawbacks, the first being that the ciphertext is twice the size of the plaintext, the second being that because it is one way you need two keypairs, one for signing messages and another for decrypting incoming messages...
The basic problem with making X faster is that it is still essentially a drawing protocol. Most of the proposals to speed up X basically rest under the assumption of making the coupling between clients and display hardware much tighter, which is fine for local display but does nothing for network performance. Berlin takes a more modular approach by separating display components from their application so instead of each application having to specify all drawing operations over some network connection, they only need to communicate high-level results (eg, "button pushed", not "mouse move" "mouse over" "mouse clicked"). The widget components then have abstract interfaces that describe their functions, which can then be implemented by any number of apropriate widgets; applications need only know about the interfaces. This makes taking theming to a really absurd level actually quite easy. You just replace the components that perform those functions. These rendering components then can be tightly or loosely coupled with the actual drawing server as necessary without having to bind the whole application closly to the server, which is where Windows is and where I see X going. Berlin is similiar in spirit to display postscript, but because it's built on CORBA it's a lot more flexible.
It's hard to tell from this article exactly what Stephenson said in this speech, in particular the bullet-going-through-window-almost-hitting-child quote seems taken out of context. I'm sure Stephenson supports the idea of cryptography, or he wouldn't have written a book like Cryptonomicon. What I think he may have been trying to say, then, is that crypto/privacy people need to change their focus first from the largely mythical (in the US anyway) jackbooted government thugs and spooks who listen to your every word and IP packet or maybe break in, make an exact copy of your hard disk and leave without a trace.
The said fact of the matter is probably that many hackers don't actually have that much to hide except maybe for their involvemnt in developing cryptographic software:)
I think Stephenson is trying to challenge the crypto "orthodoxy" that pervasive end-to-end encryption can only be a good thing. He makes it abundantly clear in Cryptonomicon that many of the initial customers for Epiphyte's Crypt are rather shady characters. This is a very interesting point, since we now have to weigh the pros and cons of the situation. Cryptography does make it much harder for people's rights to be violated by whoever might be listening it - this includes both "good guys" like the FBI and "bad guys" like criminals intent on credit card fraud. It does make it easier for criminals to communicate without a trace, but I have two rebuttals to that: first, they could be sending cleartext email right now and no one would probably notice; second the constitution was written to protect us from bad laws. This means that if the only way they can possibly prove that some law was broken is by rampantly invading someone's privacy, it is probably a very bad law to begin with. For example, the Methamphetimine Anti-Proliferation Act going through congress right now seeks to ride roughshod over the first amendment by outlawing pretty much any speech related to effects, usage, or production of illegal drugs and paraphenalia. This is a very bad law, one which the constitution was designed to protect us from, and by protecting our right to free speech through encryption we are really promoting our first amendment rights.
Okay, I've strayed a bit from the topic, but I will admit that we probably don't need 128-bit encryption for dog grooming forums and whatnot. To that end the goal of crypto nuts is clearly to call less attention to themselves by making crypto the norm rather than a notable exception (imagine filtering email for "PGP" and then watching the names that come up for shady behavior - they must have something to hide!) Of course, the idea of privacy isn't really to prevent anyone from knowing things about you - it's to let exert some control over exactly what certain people know about you. I see nothing wrong with that, because we can do this in our daily lives by only telling certain things to certain people, lowering our voices, going somewhere else to talk - why should the internet be any different, our thoughts be an open book to anyone with a packet sniffer?
The only wrist problems I have ever had I think I have to attribute to using the mouse. I tended to rest my hand on it and weight of my arm then went on my wrist, and so sometimes got pain in my right wrist (my left, non-mousing wrist was fine) which went away in a few days. In my opinion the mouse is a very bad device from an ergonomics standpoint, especially the act of clicking places a lot of stress on the tendons.
These days I have a laptop with a touchpad and haven't had any problems, the amount of force required to move the pointer around on a touchpad is much less than than to move a mouse around.
I also use a dvorak keyboard layout, which places far less stress on my wrists for typing - the debate on whether it really is faster aside, dvorak is subjectivly a much more comfortable way to type.
One seemingly minor change that I have found very nice is moving the backspace key from its far-flung position in the right corner - which forces you to either contort you wrist or move your whole arm to reach it - to the alt key to the left of the spacebar. Now deleting text is just a matter of holding down the left alt key with my left thumb, and I never have to leave the home row - faster and more comfortable. I haven't seen too many people who do this but even with a QWERTY keyboard it is a really good idea, I think.
One thing that I haven't seen anyone suggest is maybe a keyboard with a built-in pointing device. Most common on laptops (for obvious reasons) but also available on full-sized keyboards are trackpoints and touchpads. Usually the trackpoint is located right on the keyboard and the touchpad right below it. With either the travel distance is much less than out to a mouse beside the keyboard, in fact with a trackpoint you don't even need to move your hands off the home row! For the efficiency-minded who are still forced to use a pointing device it may be the best solution. Then again, I also play Quake 3 with my laptop's touchpad:)
Okay, I'm not an expert but my understanding is that this HAS already happened, and been going on for some time. Since the original pentium intel processors at least have contains multiple cores of sorts which are then used for HEAVY pipelining. The pentium is in some sense two 486 cores welded together and madly piplined. Likewise, the PPro has four cores for piplining. This is why at equal mhz a pentium is about twice the speed of a 486 and a ppro/pII is twice again as fast as a pentium. So my Celeron-433 is roughly 4 times faster than my 200mhz pentium (twice for clock rate, twice again for chip architecture). That's my understanding of it, anyway.
I belive the GPL does allow for "in-house" use of software without requiring certain stipulations of the GPL be followed - for example if your company were to modify gcc to compile INTERCAL and ONLY used it in-house and did not distribute it AT ALL they are not required to release their changes to the source. The question here is then if a beta test could be considered in-house use. If so, then it is merely fuzzy whether they are breaking the GPL, if the beta test really is public distribution, then it is a clear violation of the GPL. I think Corel has some explaining to do.
Yep. I've been playing on a SMP p200 with a voodoo1 and it's decently playable, maybe 25-30 FPS. Granted, I have everything set on "fastest rendering" and play at 512x384, but it's good enough to rule the local Q3T server:)
On the other hand, that could be because I'm two 100baseT switches away from the server and am reported as having a 0 ping playing against people off the internet through a saturated 12 megabit line all pinging _at least_ 100 - 200;-)
If your FPS is above 20, you'll be OK. For your p233, I'd actually suggest just getting a cheap voodoo1, the voodoo2 will processor bottleneck long before it will fill-rate bottleneck... Lots of unused rendering power. Same problem with a voodoo1 on a pentium, for that matter;)
playing starcraft [which i have working flawlessly with wine (well as flawlessly as can be expected)]
What versions of Wine/Starcraft are you using? I finally got wine running SC (actually SC/Brood War) a few weeks ago but it was incredibly unstable - I managed to make it through the first few terran missions by saving every 5 minutes so I could restore my game when it inevitably locked up and I had to bail out with alt-sysRq-k...
As far as I know, they specifically hired a fairly experieced Linux coder. This isn't just getting a few of their internal engineers to figure out "this Linux thing", this is someone who knows what he is doing and pushing hard from the inside to get Creative to support ALL of their hardware on Linux. So I don't think the situation is as bad as you think.
justin was killed by a full nuclear modified mini imploding cluster smart missile from justin. How strange!
Hahahahahahah... I've been wondering about that sig and just got it. I've been playing xpilot recently:) So go play xpilot and tell everyone Q3A is for weenees:)
I have a dual-p200 with a voodoo1, and with graphics settings moved over to "fastest" (with the exception of bumping the res from 400x300 to 512x384) I get more or less 30 FPS. HOWEVER, this is only running around the levels, I haven't actually played multiplayer yet so I don't know what sort of performance hit will come from actualy firefights. Nevertheless, it looks to be decent and playable enough. Maybe not super-pretty, maybe not ultra-fast, but hell, it works. Then again, 30 FPS is pretty much what I get at 640x480 in Quake 2, and I used to play Doom on a 386dx/40, so I'm kind of used to a sucky framerates:P
The classes required of a CS major are similar at my school (University of Massachusetts at Amherst). First is intro to programming, then data structures. Those are in Java. Then architechture and assembly language, which is actually a fairly heavy dose of computer engineering with some assembly projects slapped on the side. The class uses a simulated assembly language called x16, although from what I hear it's modeled after the 68000, and there is an honors section done in x86 asm. Then there is programming language paradigms class where students get their first taste of functional programming in scheme, the honors section does ML and Prolog. There is also a class on discrete mathematics (proofs, set theory, FSAs, combinatorics, etc) which is the prereq for the algorithms course. Then there is software engineering in java, operating systems, and two out of four options (databases, AI, number theory, and compilers I belive). I may be missing some, but it seems to be a fairly comprehensive curriculum... Then again, UMass is supposed to be the top reseach department in the field of artificial intellegence, and all the professors I have met are really smart guys. I have no regrest so far at majoring in CS. Other universities might not be so good...
I doubt though without a certain background you are going to be able to read and understand Knuth, to name the most famous one.
This is an interesting point. I bought TAOCP 1-3 and read them over senior year of high school. Now that I am finishing off my first year of college and have taken a class is discrete math for CS majors as well as lots of calculus and some programming language theory, I find that Knuth makes a lot more sense now, particularly the mathy parts. They are really great books, but they require a significant amount of background to really get what he's talking about. Every time I refer back to them I catch a little more detail and learn a little bit more. While it's possible that I could have done this on my own, if I didn't take classes on these subjects, it's unlikely that I would have really needed to study about permutations and such things...
Slashdot Mirror
Well, I would argue that most people's perception of their everyday surroundings focuses mostly on specific locations like home, work, school, shopping malls, etc. The commute between those places is boring and frequently deliberately blocked out (say listening to the radio rather than admiring the scenery). People have had little problems adapting to the web, which is nothing like the "learned experience" of a book or newspaper, so why does it have to be different for 3D spaces?
Well, we've been deliberately keeping a low profile -- Slashdot used to be notorious for posting free software projects that were just as much vaporware as the latest thing from Microsoft. We really don't want to do that, so we're trying to have a really complete application together before working on the "marketing" side of things. That said, we're quite far along in some ways and there are certainly plenty of other ways we could use some help :-) Also I doubt we have the bandwidth to stand up to a slashdotting...
See my comment on this. I agree, commercial ventures are at best uninteresting and at worst highly suspect. Of all free projects that I've found (and I've looked around quite a lot) I think the Interreality project is by far the most advanced in terms of architechture, features, and vision for Virtual Reality on the Internet.
The fundamental problem I have with the numerous attempts at building VR systems (aside from the fact that they are usually quite clunky and boring to use for anything besides basic chatting) is that these systems are almost universally based on large-scale central servers, rather than networks of small sites. Consider the model of the world wide web: anyone with a little bit of connectivity and bandwidth can host their own web site. Why shouldn't VR be the same way? Why do we chain ourselvers to monolithic, commercially controlled world servers rather than a community of interlinked VR rooms? The underlying technology has subtle social effects as well: would you rather have an autocratic dictatorship of a "planned" world or a democratic community where anyone can add their own pieces to the world?
The second problem (that has been noted by many people) is that 3DVR chatting has been done many, many times and is fairly uninteresting. To make this usefull, we need some big ideas about how we can bring existing applications into a 3D space and have them work more effectively, as well as completely new approaches which are only possible in an immersive environment. If VR is to be useful, it should be a ubiquitus application sitting in the background, one of many which helps you achive your tasks (work or play) rather than the elephant application that demands constant attention.
I have been leading my own free software project to build such a system: The Interreality Project and Virtual Object System (VOS). We have built a new protocol infrastructure to support distributed virtual objects (which are used to build virtual worlds) and a client program using the Crystal Space 3D engine. All of our software in available under the GPL or LGPL, and we are on our tenth public release (with a new release planned for the next couple weeks). The system has been in use at the University of Massachusetts, Amherst Laboratory for Perceptual Robotics for over a year, we have several outside developers, and there seems to be quite a bit of interest in our little project (the web site got 55,000 hits last month).
Our vision is a 3D companion to the web, but not in the (incredibly stupid) sense of putting 3D objects on web pages, but a highly interlinked ecosystem of small information resources -- which happen to convey information in 3D rather than mere 2D. Of course, multiuser support is also a fundamental part of this system -- if the world is dynamic, then any part can move and change and communicate, users, bots, agents and applicances. Perhaps the real key is that we must strive for a system that reflects the nature of the beast, the nature of the Internet, rather than trying to emulate the real world (badly).
The Virtual Object System / Interreality Project is a free software (GPL/LGPL) effort to build the software infrastructure and applications for distributed peer-to-peer virtual reality. The basic model of enabling users to host their own virtual worlds is the same as BrendanWorld, but our software supports 3D, is open source, cross platform (GNU/Linux, MacOS X and Windows) and built upon a powerful, extensible generic base. We're pretty far along in some ways, but the reason you haven't heard of us is that we've been keeping a pretty low profile so far, at least until we get our documentation up to speed. We do welcome new developers, so please come check us out!
The one major impression that I get from using Galeon is that it's clearly a browser that the authors wrote with the intention of USING it. The interface is simple and efficient, very configurable, and the bookmark handling has a number of features I haven't seen anywhere else. Their slogan is "The web. Only the web." which pretty much sums up the apparent design philosophy: do one thing, and do it well. It's definately worth a try.
I actually work with the Wearable Computing Group at the University of Massachusetts Amherst. Most of our research right now is done on wearables from Xybernaut, the company that Hitachi is partnering with as mentioned in the press release. The Xybernaut systems we have now are regular p200s (192MB RAM, 4GB disk, serial/PS2/USB ports, pretty much all the stuff you'd find on a laptop) running Red Hat 6.2. When we got them, they cost about $4000.
The biggest gripe most people have about them (aside from their general bulk) is the awful Head Mounted Display (HMD) they use. It works by using a small 640x480 LCD display pointing away from you with uses a concave mirror to reflect the image back into your eye. There are a lot of problems with them - it's very hard to get the entire screen in focus and visible, you have this big arm holding the display in front of your face, and it's almost useless in sunlight. There's also the privacy issue of the fact that anyone can just look at the display itself (which, like I said, points away from you) and see a horizontally flipped image of whatever you're currently viewing.
A useful wearable device, almost more than anything, has to have a display that is easy to wear, small, and unobtrusive. I'm curious as whether this Hitachi device will achive that to a better extent than the current Xybernaut HMD.
Something very few people bring up is the fact that the university is also a research environment. There are lots of smart people here doing cutting-edge thinking about software and hardware, and by attending school you have a chance to talk to these people and get involved. Yes, you can go to college and just take the classes you need to get by the major, but if you take some time to make friends with the professors and get involved in what they are doing, you'll learn far more than if you had spent four years writing perl scripts for a web site.
:-) You'll learn a lot about working with other people doing these sorts of things, and that's incredibly important once one gets out into the job market.
You'll also have a chance, in addition to straight computer science to take a lot of higher math courses. Calculus, linear algebra, statistics and discrete mathematics (set theory, predicate logic, finite state autonoma among others) are all quite important to know for many software applications.
It's also an excelent opportunity to get involved in student groups that do totally non-technical things, like environmental activism, cultural groups, or even trying to legalize marijuana
I'm rather amused by the two main sets of comments here on /. - the "if they could pull this off, it would be really cool" crowd and the "rewriting windows is an impossible venture" crowd.
I'm not entirely convinced either way. Idealistically it's probably a good thing, more free software, and just maybe they'll make it. On the other hand, technically, we're talking about a project that's going to take years. Three or four, and probably longer. You have to implement a multitasking kernel, GDI, C libraries, multimedia libraries, system tools, and a hundred other little and big things. Against a moving target, no less. Is this project useful?
This project will be useful to the programers on it. They'll probably learn something about programming and software engineering. Will this project be useful to other people? Maybe one day, but not for a long time. Don't hold your breath.
Twofish is not an asymmetric algorithm! RSA and ElGamal are. Otherwise you're correct, asymmetric algorithms use a key pair where one key can be used to encrypt and the other decrypt. RSA can use either the public or private key to encrypt and the opposite one decrypts, whereas ElGamal can only do one-way encryption. ElGamal is a bit faster than RSA and has been proven to be at least as strong as RSA (and perhaps a bit stronger) but suffers from a couple major drawbacks, the first being that the ciphertext is twice the size of the plaintext, the second being that because it is one way you need two keypairs, one for signing messages and another for decrypting incoming messages...
The basic problem with making X faster is that it is still essentially a drawing protocol. Most of the proposals to speed up X basically rest under the assumption of making the coupling between clients and display hardware much tighter, which is fine for local display but does nothing for network performance. Berlin takes a more modular approach by separating display components from their application so instead of each application having to specify all drawing operations over some network connection, they only need to communicate high-level results (eg, "button pushed", not "mouse move" "mouse over" "mouse clicked"). The widget components then have abstract interfaces that describe their functions, which can then be implemented by any number of apropriate widgets; applications need only know about the interfaces. This makes taking theming to a really absurd level actually quite easy. You just replace the components that perform those functions. These rendering components then can be tightly or loosely coupled with the actual drawing server as necessary without having to bind the whole application closly to the server, which is where Windows is and where I see X going. Berlin is similiar in spirit to display postscript, but because it's built on CORBA it's a lot more flexible.
It's hard to tell from this article exactly what Stephenson said in this speech, in particular the bullet-going-through-window-almost-hitting-child quote seems taken out of context. I'm sure Stephenson supports the idea of cryptography, or he wouldn't have written a book like Cryptonomicon. What I think he may have been trying to say, then, is that crypto/privacy people need to change their focus first from the largely mythical (in the US anyway) jackbooted government thugs and spooks who listen to your every word and IP packet or maybe break in, make an exact copy of your hard disk and leave without a trace.
The said fact of the matter is probably that many hackers don't actually have that much to hide except maybe for their involvemnt in developing cryptographic software :)
I think Stephenson is trying to challenge the crypto "orthodoxy" that pervasive end-to-end encryption can only be a good thing. He makes it abundantly clear in Cryptonomicon that many of the initial customers for Epiphyte's Crypt are rather shady characters. This is a very interesting point, since we now have to weigh the pros and cons of the situation. Cryptography does make it much harder for people's rights to be violated by whoever might be listening it - this includes both "good guys" like the FBI and "bad guys" like criminals intent on credit card fraud. It does make it easier for criminals to communicate without a trace, but I have two rebuttals to that: first, they could be sending cleartext email right now and no one would probably notice; second the constitution was written to protect us from bad laws. This means that if the only way they can possibly prove that some law was broken is by rampantly invading someone's privacy, it is probably a very bad law to begin with. For example, the Methamphetimine Anti-Proliferation Act going through congress right now seeks to ride roughshod over the first amendment by outlawing pretty much any speech related to effects, usage, or production of illegal drugs and paraphenalia. This is a very bad law, one which the constitution was designed to protect us from, and by protecting our right to free speech through encryption we are really promoting our first amendment rights.
Okay, I've strayed a bit from the topic, but I will admit that we probably don't need 128-bit encryption for dog grooming forums and whatnot. To that end the goal of crypto nuts is clearly to call less attention to themselves by making crypto the norm rather than a notable exception (imagine filtering email for "PGP" and then watching the names that come up for shady behavior - they must have something to hide!) Of course, the idea of privacy isn't really to prevent anyone from knowing things about you - it's to let exert some control over exactly what certain people know about you. I see nothing wrong with that, because we can do this in our daily lives by only telling certain things to certain people, lowering our voices, going somewhere else to talk - why should the internet be any different, our thoughts be an open book to anyone with a packet sniffer?
The only wrist problems I have ever had I think I have to attribute to using the mouse. I tended to rest my hand on it and weight of my arm then went on my wrist, and so sometimes got pain in my right wrist (my left, non-mousing wrist was fine) which went away in a few days. In my opinion the mouse is a very bad device from an ergonomics standpoint, especially the act of clicking places a lot of stress on the tendons.
These days I have a laptop with a touchpad and haven't had any problems, the amount of force required to move the pointer around on a touchpad is much less than than to move a mouse around.
I also use a dvorak keyboard layout, which places far less stress on my wrists for typing - the debate on whether it really is faster aside, dvorak is subjectivly a much more comfortable way to type.
One seemingly minor change that I have found very nice is moving the backspace key from its far-flung position in the right corner - which forces you to either contort you wrist or move your whole arm to reach it - to the alt key to the left of the spacebar. Now deleting text is just a matter of holding down the left alt key with my left thumb, and I never have to leave the home row - faster and more comfortable. I haven't seen too many people who do this but even with a QWERTY keyboard it is a really good idea, I think.
One thing that I haven't seen anyone suggest is maybe a keyboard with a built-in pointing device. Most common on laptops (for obvious reasons) but also available on full-sized keyboards are trackpoints and touchpads. Usually the trackpoint is located right on the keyboard and the touchpad right below it. With either the travel distance is much less than out to a mouse beside the keyboard, in fact with a trackpoint you don't even need to move your hands off the home row! For the efficiency-minded who are still forced to use a pointing device it may be the best solution. Then again, I also play Quake 3 with my laptop's touchpad :)
Okay, I'm not an expert but my understanding is that this HAS already happened, and been going on for some time. Since the original pentium intel processors at least have contains multiple cores of sorts which are then used for HEAVY pipelining. The pentium is in some sense two 486 cores welded together and madly piplined. Likewise, the PPro has four cores for piplining. This is why at equal mhz a pentium is about twice the speed of a 486 and a ppro/pII is twice again as fast as a pentium. So my Celeron-433 is roughly 4 times faster than my 200mhz pentium (twice for clock rate, twice again for chip architecture). That's my understanding of it, anyway.
I belive the GPL does allow for "in-house" use
of software without requiring certain stipulations
of the GPL be followed - for example if your
company were to modify gcc to compile INTERCAL
and ONLY used it in-house and did not distribute
it AT ALL they are not required to release their
changes to the source. The question here is then
if a beta test could be considered in-house use.
If so, then it is merely fuzzy whether they are breaking the GPL, if the beta test
really is public distribution, then it is a clear
violation of the GPL. I think Corel has some
explaining to do.
Actually, I think they're using WINE.
After Q3Test _hardlocked_ my SMP machine twice (with an ne2000 no less) I was getting a bit concerned as well. This bugfix is very encouraging :)
Yep. I've been playing on a SMP p200 with a voodoo1 and it's decently playable, maybe 25-30 FPS. Granted, I have everything set on "fastest rendering" and play at 512x384, but it's good enough to rule the local Q3T server :)
;-)
;)
On the other hand, that could be because I'm two 100baseT switches away from the server and am reported as having a 0 ping playing against people off the internet through a saturated 12 megabit line all pinging _at least_ 100 - 200
If your FPS is above 20, you'll be OK. For your p233, I'd actually suggest just getting a cheap voodoo1, the voodoo2 will processor bottleneck long before it will fill-rate bottleneck... Lots of unused rendering power. Same problem with a voodoo1 on a pentium, for that matter
playing starcraft [which i have working flawlessly with wine (well as flawlessly as can be expected)]
What versions of Wine/Starcraft are you using? I finally got wine running SC (actually SC/Brood War) a few weeks ago but it was incredibly unstable - I managed to make it through the first few terran missions by saving every 5 minutes so I could restore my game when it inevitably locked up and I had to bail out with alt-sysRq-k...
As far as I know, they specifically hired a fairly experieced Linux coder. This isn't just getting a few of their internal engineers to figure out "this Linux thing", this is someone who knows what he is doing and pushing hard from the inside to get Creative to support ALL of their hardware on Linux. So I don't think the situation is as bad as you think.
justin was killed by a full nuclear modified mini imploding cluster smart missile from justin. How strange!
Hahahahahahah... I've been wondering about that sig and just got it. I've been playing xpilot recently :) So go play xpilot and tell everyone Q3A is for weenees :)
I have a dual-p200 with a voodoo1, and with graphics settings moved over to "fastest" (with the exception of bumping the res from 400x300 to 512x384) I get more or less 30 FPS. HOWEVER, this is only running around the levels, I haven't actually played multiplayer yet so I don't know what sort of performance hit will come from actualy firefights. Nevertheless, it looks to be decent and playable enough. Maybe not super-pretty, maybe not ultra-fast, but hell, it works. Then again, 30 FPS is pretty much what I get at 640x480 in Quake 2, and I used to play Doom on a 386dx/40, so I'm kind of used to a sucky framerates :P
The classes required of a CS major are similar at my school (University of Massachusetts at Amherst). First is intro to programming, then data structures. Those are in Java. Then architechture and assembly language, which is actually a fairly heavy dose of computer engineering with some assembly projects slapped on the side. The class uses a simulated assembly language called x16, although from what I hear it's modeled after the 68000, and there is an honors section done in x86 asm. Then there is programming language paradigms class where students get their first taste of functional programming in scheme, the honors section does ML and Prolog. There is also a class on discrete mathematics (proofs, set theory, FSAs, combinatorics, etc) which is the prereq for the algorithms course. Then there is software engineering in java, operating systems, and two out of four options (databases, AI, number theory, and compilers I belive). I may be missing some, but it seems to be a fairly comprehensive curriculum... Then again, UMass is supposed to be the top reseach department in the field of artificial intellegence, and all the professors I have met are really smart guys. I have no regrest so far at majoring in CS. Other universities might not be so good...
I doubt though without a certain background you are going to be able to read and understand Knuth, to name the most famous one.
This is an interesting point. I bought TAOCP 1-3 and read them over senior year of high school. Now that I am finishing off my first year of college and have taken a class is discrete math for CS majors as well as lots of calculus and some programming language theory, I find that Knuth makes a lot more sense now, particularly the mathy parts. They are really great books, but they require a significant amount of background to really get what he's talking about. Every time I refer back to them I catch a little more detail and learn a little bit more. While it's possible that I could have done this on my own, if I didn't take classes on these subjects, it's unlikely that I would have really needed to study about permutations and such things...