Suppose that depending on the value of a byte (random, so that any value has the same probability of happening) you have to do some work. You can write a switch with 256 choices. This will raise the time needed to do all the required tests linearly.
Actually, this is incorrect at least in gcc with optimization enabled. While a switch statement is conceptually equivalent to a long if-else if-else chain, the compiler produces a jump table so that the branch reduces to an array lookup of an address, which is O(1). That, by the way, is why you can't have complicated comparisons in a switch - the jump table trick wouldn't work anymore, so it would be no better than an if chain. Sparse possibility spaces are an exception, since in those cases the jump table would take an excessive amount of space.
I'm sure there probably isn't anything to worry about, but isn't there a chance of problems if you put magnetic things near storage media?
It's bad to put a strong magnet near storage media because storage media ARE magnetic; that is, data is stored by magnetizing portions of the disk surface, and so a sufficiently strong magnet can wipe out all of the contents of the disk. That said, it takes a *very* strong magnet to do this (or, alternatively, one millimeters away from the disk platter), and in this case, we're talking about the equivalent of putting two hard disks next to eachother. No issue at all.
The best solution, IMO, is to do it quick and dirty, sacrificing all your design principles *except* for modularity. That way the dirtiness is contained, and you can clean it up later; plus, the correct and proper solution won't take as long as it would normally because of what you learned from doing the dirty version (in the same way that it helps to do a rapid prototype). Then, when doing maintenance, your approach to adding any feature and to fixing any bug should be to refactor. And I cannot overemphasize the importance of refactoring at every opportunity after you've written a dirty program.
I see you didn't read the article. Nvidia is actually detecting 3dmark and substituting in more efficent renderers and dropping the back buffer clearing at certain points to get higher FPS scores.
Something else that may shock you: it appears that ATI is doing the same thing, although to a much lesser extent.</BLOCKQUOTE>
ATI got caught in a much worse cheat a while back: they lowered the texture quality in Quake 3 when it was set to maximum. Skipping a back buffer clear should, in theory, have no effect on the image. If ATI is doing it "to a much lesser extent", then it's only because of the backlash they got when they got caught that time.
One of the basic consequences of Shannon's Law, a fundamental tenet of information theory, that in order to increase your bandwidth and transmission rate, with a given noise level (which we can't reduce beyond a certain point, due to inherent cosmic background noise, not to mention many other manmade factors), you have to increase your transmission power to compensate.
No, it doesn't; in fact, it says that there is a maximum possible bandwidth for a given power and noise level, which current technologies are far short of. Therefore, it is possible to increase bandwidth without increasing power, to a point. I might also add that the FCC limits transmission power on all parts of the spectrum.
With all this RF energy floating about amidst space, I am sort of concerned that if ultra high-speed wireless becomes ubiquitous, without the right studies being done, this may cause negative impact to health. While I am not a physician or molecular biologist, I think that we need to investigate this before jumping too quickly.
I recall an article from Skeptical Inquirer awhile back which investigated the claims of some who claimed to suffer from 'electrosensitivity'. The finding was that a visible non-transmitting antenna or wire would produce the supposed symptoms, while a concealed active one had no measurable effect; therefore, the supposed symptoms were entirely psychological. In light of the number of crackpots claiming that EM radiation affects them, and the thoroughness with which they have been debunked, I don't think it necessary to do any further research.
When it says "Life + 70", that is life of the corporation which took out the copyright, which may be infinite, right?
No. It is life+70 if the author is an individual; if it is a corporation (which could live very long or die quickly), it is simply 90 years.
Re:Someone should start a BSD C/C++ compiler proje
on
RMS Turns 50
·
· Score: 3, Informative
My background is math, not CS, but I'm led to believe that writing a compiler (or at least the core of one) is a standard thing to do for undergrad CS students... some enterprising hacker should write a bare-bones C compiler and release it under the BSD license. It seems to me that if it were well-designed, plenty of hackers would be glad to help out with the optimizer, writing backends for other CPUs, etc... and perhaps after a few years, the compiler would be solid enough for the *BSDs to switch to as their default compiler.
Write a C compiler is easy. Writing a C++ compiler is hard. Writing an *optimizing* compiler is very, very hard. gcc may not produce the fastest code for any one processor, but it supports just about every processor you can name, and then some (for example, TI calculators; see my sig), and it optimizes well for all of them. And why should hackers choose to leave gcc for some upstart compiler? It would need some remarkable technical merits, and a BSD license only debatably counts as a merit at all.
The question is, does this release fix the "DAMN this game is IMPOSSIBLE!!" bug, or the "Holy crap, I keep dying for no reason." Bug?
Actually, it does. NetHack 3.4.0 was much harder than 3.3.1 because it made an AI change which allows spellcasting monsters to cast spells when not in melee range, causing priests (especially at the minetown altar) to go from pushover to deadly, with their insect-summoning, and all monsters which cast the Summon Nasties spell (titans, * liches, most demons, and Rodney) suddenly became absurdly dangerous, because they could summon other summoners. 3.4.1 fixes both of these, making the late-game as easy (in four ascenscions, I never lost a character past the castle) as it was in 3.3.1.
People here are talking about how it's WotC's right to charge for their data, and about the "3.5th ed" of AD&D. But no one has actually brought up the reason why this is important: it's all happened before, and it didn't end well.
D&D used to be published by a company called TSR (originally Tactical Simulation Rules, then the acronym was dropped). While the company was under the control of Gary Gygax, all was good; but when Gygax left, he was replaced by typical business types. They decided that by publishing new books, they could make more money from their existing players; so, they published an "advanced" version, AD&D, which would live alongside D&D. Some time after that, they published a 2nd edition AD&D and discontinued both D&D and AD&D. The new editions were improvements, but people weren't happy with re-purchasing and re-learning the same things repeatedly.
Somewhere along the line, someone at TSR found out about the World Wide Web, and they weren't happy with what they found: TSR's copyrighted material, even complete texts, on personal web pages. They decided to crack down, but they came for not only infringing sites, but also legitimate fan sites. They made a public statement explaining what fan sites could and couldn't use. They made two grave mistakes; first, they tried to claim ownership over the term "hit points", by then already used universally in nearly every RPG around, and proposed a lousy substitute. Second, they said that fan sites could use monsters which come from mythology (their example: a Hydra), but not ones which are creations of TSR (Drow elves). Problem was, an edition of Webster's Unabridged Dictionary published several years before D&D had an entry under "drow", which described TSR's drow elves perfectly.
At the same time, TSR was milking the market for all it was worth. They published "2.5th Ed." of AD&D, which was 2nd Ed. in a slightly different presentation (more pictures), and they published dozens and dozens of unnecessary, low-quality, repetitive and inconsistent rulebooks. In short, they made a mess of 2nd edition AD&D, and earned themselves a dismal reputation as "T$R". The backlash killed them, and TSR was bought out by Wizards of the Coast.
Third edition was a symbolic fresh start; it discarded the mess of rulebooks created in 2nd edition, simplified things, and used the D20 license to show that, unlike TSR, WotC was committed to openness. Does a 3.5th edition and stopping one copyright infringement mean that WotC is reverting into TSR? Of course not; but it's a step in that direction, and could become the start of something more.
An erroneous lookup results in getting directed at an advert, instead of getting told you're in error, more like.
Not exactly. It looks at the query and decides whether it thinks you want a non-English domain, and if so, directs you to a page to get an IE plugin which adds support for international URLs. A very dirty hack and not in any way part of the DNS standard, but not advertising.
I disagree about the difficulty in propagating the worm under IPv6. It might slow it down, but I was online when the worm hit and it was almost instant the way it consumed the backbones. I'd estimate that within 5-10 minutes the worm went from one end of the world to the other.
I used the word "astronomical" for a reason. Allow me to do some quick math. There are 2^128 possible IPv6 addresses. Assume there are 2^32 nodes on the network (many times the number that exist today), plus 2^24 infected nodes (assume that the worm's designer targetted them for infection to seed the worm; again, a much larger number than is really realistic). Thus, if they divide the work equally and there are no duplicate scans, each infected host will need to scan 2^72 nodes. Assume that a scan takes only 2^3 bytes (the size of the address to send to and nothing else), and that each of the infected hosts is connected to a 2^48bps (256 terabit) uplink. It will then take 2^24 seconds, or about 13 years, to find a single host. And mind you, that host probably isn't even vulnerable to the virus.
The scary thought for IPv6 to me is that it might slow down random IP propagation, but that would probably be inconsequential when compared with the increased number of spammers that would find new life and longevity in hiding amongst the exponentionally larger IP space.
Blocking subnets was a bad and controversial solution anyways, since innocents were frequently caught in the crossfire. An infinite number of subnets won't make any difference if no upstream provider will touch them with a ten foot pole, which is the way spam is properly fought.
No one's laid blame on it, but I think that the real way to get rid of these worms is to transition the net to IPv6. Slammer, Code Red, Code Red 2... all of them work by brute-force IP scanning. That only works because the IPv4 addres space is so densely populated; with IPv6, a worm would never be able to spread itself that way because the odds against a random hit are astronomical. I'm not saying that this should be a substitute for keeping servers up to date, but all the patching in the world doesn't help when the problem is that some faraway node is crushed under the traffic created by a worm, and IPv6 is good for many other reasons as well.
Everytime IE encounters (for the first time in each session) a non-IIS server, it promptly connects to MSN Search and submits the website address....
Either this depends on a specific option being set or using a specific version of IE other than the one I have, or this is an outright lie. I just watched a packet dump of IE making an HTTP request, and there were no suspicious extra IP addresses involved in the transfer (just myself, the nameserver, and the server requested).
I don't remember just how large the key was for the last RSA challenge, but it certainly wasn't more than a kilobyte. If we assume it was a kilobyte, the difference in processing power required is a factor of 2^2048/2^1024 = 2^1024 ~= 1.8x10^308. So unless they've reduced the effective key size by a lot, there probably isn't enough matter in the universe to make the computers that would be required to break that key.
Some time after getting Unreal Tournament 2003, I set out to appraise its AI. I decided to set up a game in which it couldn't cheat; I made a one-on-one game, on the map DM-Gael (a small, open map, so while the bot may always know the player's location, also vise versa), and with rocket launchers only (so that the bot couldn't do some simple trig to always hit). I set the bot to its highest difficulty, and played.
The bot had some notable weaknesses (it kept getting killed going for the powerup in the center, or while coming up a lift, and never seemed to learn from these mistakes), but did fairly well overall. In the end I won with a substantial, but not overwhelming, margin.
So, I said, the AI had failed the test: given a fair match, on its most difficult settings, it lost. But then I realized, I had a lot of fun administering it. Then I realized that the point of an AI isn't to beat the player, but to be fun to play against; whether it wins or loses really doesn't matter.
Plaguarism for karma is extremely dishonest. You didn't change a single word of YE's post on the previous version of this story. Looking at your posting history, this would seem to be a new low for you.
What is the performance like when BSD is emulating linux which is emulating windows?
No emulation is occuring; WineX implements the Windows API, and this article says someone implemented some Linux API calls on BSD. It doesn't involve any translation of machine code, which is the slow part of emulation; theoretically, it should run at the same speed as the Windows version, if the device drivers are of equal quality.
No. Common misconception. The danger is at the end of the eclipse: your eyes are accustomed to the dark, your pupils are dilated, and suddenly the first Bailey's Bead appears. The Bailey's Beads can be very bright compared to the eclipse itself: the corona is instantly invisible again.
I disagree; Bailey's Beads aren't that bright, and they appear gradually enough for you to look away. If you keep looking after that without putting on some protective gear, well, it's your own fault.
The download server's not responding, but I'd say it's probably 2.02 megs because it features a bloated installer to apply the changes. If you read the changelog, it's pretty trivial.
I can only think one thing, being the typical paranoid slashdotter... E. T. Phone Home! And then get caught by the cops! <BR><BR> I think it's just a few "fixes" to make people download it, then it phones home.</blockquote>
That's because you didn't follow the link and RTFA. This isn't from iD, and all it's changing are config files.
Sorry Kip Knight, but I find it hard to believe that you have come up with anything new. By refusing to disclose your invention, you have kept those who would tell you how wrong you are from doing so. For a one-time key to be provably secure, the key must be as the sum of all messages sent using it; any other method, such as re-using key space on the assumption that said re-use will not provide enough information to break the code, specifying a source of future key space in a message (since this is equivalent to using a key shorter than the message, assuming finite numbers of commonly-accessible sources of data), or manipulating the message in advance (to make it less recognizable, or to reduce size and to save key space), is obvious and/or dangerous. Also consider that the one-time pad comes from math, and may not be patentable due to the ban on patenting mathematical formulas.
Furthermore, any variation on the one-time pad is rather useless, since it ignores the practical problem in cryptography, which is key exchange. All modern cryptography works on the assumption that an analyst can observe *everything* you send and receive, including keys, and the solution to this is public/private key crypto. If you have to meet in person to exchange keys, it's easy to exchange large ones (CD-Rfulls of key), so re-using a key doesn't provide much real benefit.
If anything I'm saying is news to you, then you should forget the whole thing, since your invention is probably worthless. If you are aware of all this and still think that you have a useful and patentable invention, then you should find someone very knowledgeable in cryptography to talk to, get an NDA, and discuss what you have.
I'm comparing a console with a $3000 PC being marketed as a high-end game machine.
That doesn't mean that people buying it will use it exclusively to play games. You're counting the full $3000 price, when you should be counting the difference between that price and a mid-range computer.
Again, I agree with you. However, it's been my experience that in the rush to get products to market, game companies frequently ship defective products.
As a frequent dabbler in beta, demo, freeware and otherwise expected-to-be-bad software, I have to say that my experience is quite to the contrary. Back in the dark DOS when games had to interface directly with the hardware this was certainly true, but today they only deal with abstraction layers. It is true that companies can be somewhat sloppy because of patches, but (a) this is going to happen with any multiplayer console games, for the same reason, and (b) this allows for faster releases and thus a time advantage in games over consoles.
Subscription fees for multi-player gaming are virtually pure profit once the servers are in place.
As opposed to multiplayer PC games, which are (with the exception of MMORPGS) never subscription. This eats up more of the price difference you were arguing in point (1). As for the keyboard, there certainly is a reason why console makers won't add one: they don't want to turn their consoles into general-purpose PCs, and most people don't have furniture setups suitable for keyboards around their consoles.
Why would anyone pay three thousand dollars for a PC to play games?? Earlier this week - Slashdot ran a thread about PCs Losing Out as a Gaming Platform?. Game companies are developing for consoles first. PC games typically involve an hour or more (if you're lucky) of patch / configuration hell vs. load and go with console games. Soon most consoles will support broadband and multiplayer games. They might be able to sell a $3000 box to a business for CAD applications, but as a gaming box; high end PC's are dead.
I don't have moderator points today; if I did, I would mark this a troll. You can't compare a $3000 PC to a $200 console on price because they're entirely different things; the hardware in a console is several generations behind the hardware in the PC, and PCs do many (non-game) things that consoles don't. As for your claim that PC gaming is losing out to console gaming, well, people have been saying that for at least five years and it's never been true. And any game that involves "an hour or more (if you're lucky) of patch / configuration hell" is defective. Spend ten minutes while the auto-updater runs, set up your controls (or just use the defaults), and go.
That consoles will "soon" support broadband and multiplayer is irrelevant, firstly because it has been "soon" for some time now and has never really materialized, and second because consoles lack keyboards, and therefore automatically lose the very important social aspect of online gaming.
The only justification I can see for a serious gamer NOT having a high-end PC is that middle-end PCs run current games well enough. Howver, this will cease to be the case when ID releases the next Doom, and serious gamers use up the performance cranking detail settings anyways.
Nowadays, you see all of these benchmarks on chips/chipsets/memory, and unless you're talking inSANE resolutions and color depths, AMD/Intel vs nVidia/ATI really doesn't matter much. It's just personal preference.
Resolutions and color depths have nothing to do with the chips/chipsets/memory; the component most affected by that is the video card. And if you think video cards are fast enough that choosing between nVidia and ATI is just personal preference, odds are you aren't doing anything which deserves more than an old 4-meg video card.
It was my understanding that the major bottleneck of any system is the DISK. So no matter how fast your ram is, if you still have to swap to the slow-ass disk, your system will be slow.
While starting up programs, yes, the disk is the bottleneck as files are loaded the first time. But for the tasks where speed *really* matters - compiling programs, long simulations, games - the speed of CPU and RAM are critical. (Disk can be made important if RAM is lacking in quantity, but with RAM prices as they are these days, that is inexcusable.)
Screen real estate is extremely valuable, but the way to recover it is certainly not to move things onto a miniature secondary screen. Virtual desktops are a much better solution. Most Linux UIs support them, and if you use Litestep, so can Windows. The concept is simple: have several "virtual" screens and switch between them. That way, you can put all your "overhead" windows (ICQ, Winamp/XMMS, tickers, file manager) on one screen, Mozilla on another, etc, and switch between them with a mouse click. You can also gain space by getting rid of the stuff that wastes it, like adware, or by keeping it minimized.
Of course, you can also gain more screen real estate by investing in hardware, but if you're going to do that it makes far more sense to get a bigger or second monitor than a black+white LCD screen.
Slashdot Mirror
It's bad to put a strong magnet near storage media because storage media ARE magnetic; that is, data is stored by magnetizing portions of the disk surface, and so a sufficiently strong magnet can wipe out all of the contents of the disk. That said, it takes a *very* strong magnet to do this (or, alternatively, one millimeters away from the disk platter), and in this case, we're talking about the equivalent of putting two hard disks next to eachother. No issue at all.
The best solution, IMO, is to do it quick and dirty, sacrificing all your design principles *except* for modularity. That way the dirtiness is contained, and you can clean it up later; plus, the correct and proper solution won't take as long as it would normally because of what you learned from doing the dirty version (in the same way that it helps to do a rapid prototype). Then, when doing maintenance, your approach to adding any feature and to fixing any bug should be to refactor. And I cannot overemphasize the importance of refactoring at every opportunity after you've written a dirty program.
No. It is life+70 if the author is an individual; if it is a corporation (which could live very long or die quickly), it is simply 90 years.
People here are talking about how it's WotC's right to charge for their data, and about the "3.5th ed" of AD&D. But no one has actually brought up the reason why this is important: it's all happened before, and it didn't end well.
D&D used to be published by a company called TSR (originally Tactical Simulation Rules, then the acronym was dropped). While the company was under the control of Gary Gygax, all was good; but when Gygax left, he was replaced by typical business types. They decided that by publishing new books, they could make more money from their existing players; so, they published an "advanced" version, AD&D, which would live alongside D&D. Some time after that, they published a 2nd edition AD&D and discontinued both D&D and AD&D. The new editions were improvements, but people weren't happy with re-purchasing and re-learning the same things repeatedly.
Somewhere along the line, someone at TSR found out about the World Wide Web, and they weren't happy with what they found: TSR's copyrighted material, even complete texts, on personal web pages. They decided to crack down, but they came for not only infringing sites, but also legitimate fan sites. They made a public statement explaining what fan sites could and couldn't use. They made two grave mistakes; first, they tried to claim ownership over the term "hit points", by then already used universally in nearly every RPG around, and proposed a lousy substitute. Second, they said that fan sites could use monsters which come from mythology (their example: a Hydra), but not ones which are creations of TSR (Drow elves). Problem was, an edition of Webster's Unabridged Dictionary published several years before D&D had an entry under "drow", which described TSR's drow elves perfectly.
At the same time, TSR was milking the market for all it was worth. They published "2.5th Ed." of AD&D, which was 2nd Ed. in a slightly different presentation (more pictures), and they published dozens and dozens of unnecessary, low-quality, repetitive and inconsistent rulebooks. In short, they made a mess of 2nd edition AD&D, and earned themselves a dismal reputation as "T$R". The backlash killed them, and TSR was bought out by Wizards of the Coast.
Third edition was a symbolic fresh start; it discarded the mess of rulebooks created in 2nd edition, simplified things, and used the D20 license to show that, unlike TSR, WotC was committed to openness. Does a 3.5th edition and stopping one copyright infringement mean that WotC is reverting into TSR? Of course not; but it's a step in that direction, and could become the start of something more.
Blocking subnets was a bad and controversial solution anyways, since innocents were frequently caught in the crossfire. An infinite number of subnets won't make any difference if no upstream provider will touch them with a ten foot pole, which is the way spam is properly fought.
No one's laid blame on it, but I think that the real way to get rid of these worms is to transition the net to IPv6. Slammer, Code Red, Code Red 2... all of them work by brute-force IP scanning. That only works because the IPv4 addres space is so densely populated; with IPv6, a worm would never be able to spread itself that way because the odds against a random hit are astronomical. I'm not saying that this should be a substitute for keeping servers up to date, but all the patching in the world doesn't help when the problem is that some faraway node is crushed under the traffic created by a worm, and IPv6 is good for many other reasons as well.
I don't remember just how large the key was for the last RSA challenge, but it certainly wasn't more than a kilobyte. If we assume it was a kilobyte, the difference in processing power required is a factor of 2^2048/2^1024 = 2^1024 ~= 1.8x10^308. So unless they've reduced the effective key size by a lot, there probably isn't enough matter in the universe to make the computers that would be required to break that key.
Some time after getting Unreal Tournament 2003, I set out to appraise its AI. I decided to set up a game in which it couldn't cheat; I made a one-on-one game, on the map DM-Gael (a small, open map, so while the bot may always know the player's location, also vise versa), and with rocket launchers only (so that the bot couldn't do some simple trig to always hit). I set the bot to its highest difficulty, and played.
The bot had some notable weaknesses (it kept getting killed going for the powerup in the center, or while coming up a lift, and never seemed to learn from these mistakes), but did fairly well overall. In the end I won with a substantial, but not overwhelming, margin.
So, I said, the AI had failed the test: given a fair match, on its most difficult settings, it lost. But then I realized, I had a lot of fun administering it. Then I realized that the point of an AI isn't to beat the player, but to be fun to play against; whether it wins or loses really doesn't matter.
Plaguarism for karma is extremely dishonest. You didn't change a single word of YE's post on the previous version of this story. Looking at your posting history, this would seem to be a new low for you.
I disagree; Bailey's Beads aren't that bright, and they appear gradually enough for you to look away. If you keep looking after that without putting on some protective gear, well, it's your own fault.
The download server's not responding, but I'd say it's probably 2.02 megs because it features a bloated installer to apply the changes. If you read the changelog, it's pretty trivial.
Sorry Kip Knight, but I find it hard to believe that you have come up with anything new. By refusing to disclose your invention, you have kept those who would tell you how wrong you are from doing so. For a one-time key to be provably secure, the key must be as the sum of all messages sent using it; any other method, such as re-using key space on the assumption that said re-use will not provide enough information to break the code, specifying a source of future key space in a message (since this is equivalent to using a key shorter than the message, assuming finite numbers of commonly-accessible sources of data), or manipulating the message in advance (to make it less recognizable, or to reduce size and to save key space), is obvious and/or dangerous. Also consider that the one-time pad comes from math, and may not be patentable due to the ban on patenting mathematical formulas.
Furthermore, any variation on the one-time pad is rather useless, since it ignores the practical problem in cryptography, which is key exchange. All modern cryptography works on the assumption that an analyst can observe *everything* you send and receive, including keys, and the solution to this is public/private key crypto. If you have to meet in person to exchange keys, it's easy to exchange large ones (CD-Rfulls of key), so re-using a key doesn't provide much real benefit.
If anything I'm saying is news to you, then you should forget the whole thing, since your invention is probably worthless. If you are aware of all this and still think that you have a useful and patentable invention, then you should find someone very knowledgeable in cryptography to talk to, get an NDA, and discuss what you have.
As a frequent dabbler in beta, demo, freeware and otherwise expected-to-be-bad software, I have to say that my experience is quite to the contrary. Back in the dark DOS when games had to interface directly with the hardware this was certainly true, but today they only deal with abstraction layers. It is true that companies can be somewhat sloppy because of patches, but (a) this is going to happen with any multiplayer console games, for the same reason, and (b) this allows for faster releases and thus a time advantage in games over consoles.
As opposed to multiplayer PC games, which are (with the exception of MMORPGS) never subscription. This eats up more of the price difference you were arguing in point (1). As for the keyboard, there certainly is a reason why console makers won't add one: they don't want to turn their consoles into general-purpose PCs, and most people don't have furniture setups suitable for keyboards around their consoles.
That consoles will "soon" support broadband and multiplayer is irrelevant, firstly because it has been "soon" for some time now and has never really materialized, and second because consoles lack keyboards, and therefore automatically lose the very important social aspect of online gaming.
The only justification I can see for a serious gamer NOT having a high-end PC is that middle-end PCs run current games well enough. Howver, this will cease to be the case when ID releases the next Doom, and serious gamers use up the performance cranking detail settings anyways.
IHBT. IHL. HAND.
Screen real estate is extremely valuable, but the way to recover it is certainly not to move things onto a miniature secondary screen. Virtual desktops are a much better solution. Most Linux UIs support them, and if you use Litestep, so can Windows. The concept is simple: have several "virtual" screens and switch between them. That way, you can put all your "overhead" windows (ICQ, Winamp/XMMS, tickers, file manager) on one screen, Mozilla on another, etc, and switch between them with a mouse click. You can also gain space by getting rid of the stuff that wastes it, like adware, or by keeping it minimized.
Of course, you can also gain more screen real estate by investing in hardware, but if you're going to do that it makes far more sense to get a bigger or second monitor than a black+white LCD screen.