When you make claims like this its helpful to provide examples.
Examples? Nethack, Omega, XConq, FreeCiv, xblast, xtrek, xtb, and many others. Browse around the gaming-related USENET groups from the 1980's and 1990's. In fact, many of those wonderful commercial games you so much admire are copies (with more graphics but worse game play) of open source games.
Of course, in absolute numbers, there aren't as many of those OSS games around--while something like nethack gets maintained for 20 years, you might get a dozen releases from a commercial gaming company, all with roughly the same gameplay but new graphics.
Its also interesting that you compare all of one groups games versus only once genre
Uh, huh. We'll see in another decade or so. I predict we'll have a rich selection of high-quality OSS games in the currently hot commerical genres. It will start off as running "total conversions" and upgraded freely released commercial game content on new OSS engines and then gradually move to entirely new content.
It's possible that the rapid development of technology as we've seen it over the last century/centuries will allow humanity to meaningfully colonize Mars within the next two to three centuries. I think the space elevator is a piece of the puzzle since it will allow very low cost orbital deployment.
Yes, it may be technologically possible to put a colony on Mars even within a couple of centuries. But that simply has no bearing on where we send our planetary probes today. If Venus were scientifically more interesting, we should be sending probes there despite its tough conditions.
I don't see that happening. Commercial games typically take 18 months to complete. An open source project will take longer than that if it's worth playing. I can't see unpaid virtual teams sticking together for that long - it's too much effort for too little reward.
So? Many open source projects have been around for decades, not just 18 months. The ability to do development for the long haul is where open source has a huge advantage. And the different time scales are also why you are only now beginning to see open source equivalents of games like Quake: open source really is in it over the long run.
And games are much more of an artistic endeavour than a purely technical challenge of writing an operating system where there's often a provable "best way" of doing something. Operating systems are a means to an end - games exist only for the sake of the game, hence everyone wants things to work their way. Expect infighting, splinter groups and a lot of projects not getting finished.
I think you really misunderstand software if you think that there's a "provably 'best way'" of designing a word processor, kernel, or graphics editor.
In any case, if the game is designed well, that kind of dynamic will improve things: competing models, competing looks, multiple interacting worlds, etc.
In fact, there are plenty of open source games out there that are very popular and have stood the test of time. That's a lot more than can be said for many of those commercial FPS's out there.
The only reason you see so many commercial games with no apparent OSS counterpart is because commercial vendors push the limits of what hardware can do; OSS, for various reasons, is far more conservative, but no less inventive. And in terms of actual game play, most commercial games really aren't that hot.
Assuming those were good reasons to colonize other planets, the time frame of those issues is so long-term (as is the time-frame in which we will be technologically ready to actually colonize other planets in any meaningful way) that it really has no bearing on which planets to target by unmanned probes for the foreseeable future.
In this century and the next, we should pick our targets for maximum scientific benefit, not for the possibility of colonization.
You could ask the same about any game genre: why are the free, open-source first-person shooters lagging? The large amount of work that goes into any game means that they're not as easy to develop.
Neither is the Linux kernel or many other open source projects. Furthermore, there are plenty of "total conversions" of existing video games that involve completely new models of everything.
No, the reason why you don't see that stuff is cultural and sociological: the kind of people who do that sort of thing generally are not very much into open anything. Many of them come out of a shareware or "freeware" tradition that happens to be closed source and builds on commercial building blocks.
But it's going to change over time: as open source provides better tools engines and better graphics tools, you'll see more of that energy being poured into open source games of every genre.
There are also those who of course, believe that Mars is chiefly where we will dump those extra billions of people we are going to have in the next 100 years.
You're gonna ship African and Indian street kids to Mars by the billions? Because that's where those "extra billions" come from. It's not going to happen.
There are only two ways we will deal with the population explosion: family planning and social changes on the one hand, or disease and starvation on the other.
Opening up new horizons is part of fixing the problems on Earth. Not only is the space program generating research that is highly applicable to current Earthly problems*, [...] * Such as medicine, hydroponics, closed ecosystems, energy sources, micro-mechanization, robotics, etc.
What good do lessons about ecology do us when we don't apply them? We have more than enough technology to solve the problems here on earth--the obstacles are purely social and political. Colonizing Mars isn't going to solve that.
it is also providing for an eventual safety valve where disaffected members of society can go off to a Moon or Mars colony to start a new life rather than remaining on crowded Earth.
A six months interplanetary trip followed by enormously expensive life support is supposed to be a "safety valve"? Sorry, but you are naive.
An example of this can be seen in the early pioneering days of the United States. Sure it was a tough, difficult, and often deadly trip west but many people did it anyways in order to start off new. Many of those people had been feeling stifled in the eastern cities and so they went west, relieving the social caldrons which were beginning to boil over.
First of all, the US was, in many ways, a paradise: abundant wildlife, good climate, rich soils. Second, many people came lured by false advertising: despite nearly ideal conditions, life was indeed harsh (as life tends to be without tools and infrastructure). Third, emigration to the US hardly did much "relieving" for the countries where people emigrated from; quite to the contrary, after being a dumping ground for dopes and criminals initially, the US soon took away many of their most enterprising citizens, a "brain drain" that continues to this day. This continues to benefit the US at the cost of everybody else.
Of course, none of that is relevant to Mars: Mars is far less suitable for colonization than the Sahara desert, the top of Mount Everest, or the Antarctic. And in addition to that, it's much more costly to reach.
We either fix things here on earth or we die: colonization of other planets is not a viable alternative over the next couple of centuries at least.
In roughly the same sense that the moon can be, and lots of other bodies. But what's the point?
but first thing is first.
You're naive if you think that we are sending probes to Mars because we want to colonize it in the near future. Maybe Bush thinks that, maybe Dan Quayle thinks that, but it's just not going to happen any time soon.
The people who actually are making the exploration of Mars and the rest of the solar system happen don't want to colonize those bodies either. In fact, many astrobiologists would be horrified by the notion if it ever became a realistic possibility.
Colonizing Mars or Venus, let alone terraforming it, are fortunately so far out that nobody has to worry about it. The moon would be a much better place to "colonize" for the time being if we really wanted colonies in space. Of course, nobody has yet come up with a rational reason why we would want to.
As for Venus, well, right now, cooler worlds are just the more interesting ones: when so much of the budget is spent on military nonsense, scientists have to pick and choose their targets carefully.
Yes, film is pretty much doomed (except for niche applications). But Kodak has seen this coming and started preparing in time. I think among old companies that needed to transform themselves, Kodak has been doing pretty well: their digital camera lineup is decent, they have done some nifty stuff with OLED, and they still have lots of non-consumer products that probably make them money. They also were one of the first companies to actually sell digital cameras widely. Kodak isn't a hot company, but give the guys a break on this one--they haven't been blind and they have been trying to go for the new market.
What is really dragging Kodak down is their brand name--some companies have a brand name that stands for innovation, and they can put out any kind of garbage and people will think it's the latest and greatest thing. Kodak, on the other hand, can put out a really nifty digital camera and the stale odor of photographic fixing solution clings to it in the mind of buyers (yes, including my own).
I hope you do understand that it is the US that is pushing for this. Basically, the US is saying that the Europeans need to create biometric passports if they want to be able to continue to travel to the US under the visa waiver program.
What is particularly bad about this is that US citizens will apparently not be subjected to biometric identification, at least not yet. It actually seems rather doubtful that something like that is palatable domestically, although in the hysteria after 9/11, I suppose you never know.
What is also bad about it is that biometrics give people a false sense of security: they don't actually increase security very much (since they can be forged), but they tend to decrease human security checks--immigration officials trying to judge whether someone is behaving in odd ways, etc.
I hope that the Europeans are going to refuse and say "OK, kill your tourism market if you like"--the US has much more to lose from making it harder for Europeans to travel to the US than the other way around. But, politically, that doesn't look likely.
Call me crazy, but I think the US having the ability to rain down death and destruction on anyone who gets in our way does make the world a safer place.
Yes, you are crazy. If the US were to succeed, it would do anything from incur serious economic and political sanctions to increased anti-American sentiments and increased terrorism. One thing it wouldn't do is make America safer.
For Americans. And those are the only people our tax dollars should be protecting in the first place.
Except, of course, that the US tax payer isn't paying for this, the US is borrowing from other nations to pay for it.
Don't like it? Go get your own military for once.
Why? Another decade or two of this insane military spending and foreign nations will own anything that is worth owning in the US.
You've got to be kidding. Traditionally, few civilians got killed in warfare. Advanced military technology has led to an escalation in civilian deaths and a huge reduction in military deaths, in a perverse reversal of an already perverse means of settling disputes. And wars have become more frequent, not less frequent.
As for nuclear weapons being "proven deterrents", perhaps you forget how close we came to global nuclear holocaust on several occasions during the cold war. If either the US or the USSR leaders had gotten off on the wrong foot, we'd all be history. Nuclear weapons nearly killed us all.
The named argument lists and lack of oddball syntactic noise alone make Smalltalk syntax far more readable, for both experienced and novice programmers.
Ruby likes to pretend it's some kind of Smalltalk successor. It isn't. Ruby is a Perl successor that borrowed a few tidbits from Smalltalk.
Oh, dear, I did it too, referring to the platform as the "JVM". No, it's not sufficient for them to open source their JVM; if they want to make Java an open platform, at this point, they would have to open source their entire JDK 1.5 implementation, without conditions or strings attached. But, for the reasons I already mentioned, I don't think a truly open Java is even desirable. Java should just get replaced instead.
First, they need to open up the Java standard unconditionally, so that anybody who wants to implement it can do so without restrictions or interference from Sun; right now, the Java 2 platform specifications are covered by licenses, and conforming implementations would probably run into several Sun patents.
Second, because Java has become so bloated and Sun has effectively killed off competing efforts, it is far too late to hope for independent implementations of the Java platform, so if Java is ever going to be an "open standard" in any meaningful sense, Sun has to open source their JVM.
Of course, at this point, I hope they won't. I think technically, Java has become a lost cause. It's easier to replace it than to fix it.
I used to want Sun to open source Java, but they've actually been a pretty good steward and I quite like what they're doing with it. The Java Community Process seems to be working.
Working in what way? In the sense of producing a language that works for some people? Sure. But the same can be said for Microsoft and VisualBasic.
The real problem is that the Java core is heavily covered by Sun intellectual property (restrictions on the specifications, patents, copyrights). That means that all this wonderful free work that the JCP puts in around the periphery ends up effectively contributing only to a Sun-controlled platform.
I'm tired of Sun claiming that they are a friend of open source and getting away with it. Sun broke their promise to have Java standardized by an independent organization twice. Sun's licenses on the Java specifications and their Java implementations are reprehensible: if you only as much as look at them, they have legal claims over your Java-related work. Sun has threatened various open source projects and forced them to agree to Sun's licenses on the specifications. The JCP makes people work for free to benefit mostly Sun. And Sun keeps announcing that open source isn't good enough, that only Sun managed to turn Gnome into something decent, and that commmercial users will be flocking away from Linux to Solaris. Furthermore, technically, Java evolves at a snail's pace.
Sun used to be a decent company and there was much hope initially for Java. But almost a decade later, I think Sun and Java are bad news for open source. As Sun's fortunes decline (and they will because they are technically not very good anymore and way overpriced), they will become vicious and they'll start taking advantage of all the intellectual property they have on Java.
Children could be taught to program early in their school years.
Yes, and that's what we have languages like Logo for. Or, if you want something readable and more adult, Smalltalk (and the Squeak.org environment). Python, of course, originally also was designed for education.
Ruby is a nice language, but I don't think it comes out of an educational background. It syntax is cleaner than Perl's but doesn't seem like it would be all that intuitive to non-computer users either (both Smalltalk and Logo seem better in that regard).
Altogether, Ruby just seems like an odd choice when it comes to really caring about teaching people to program. Not a bad choice, but not clearly better than the more obvious choices either.
X10 did indeed exist. They changed the major version number to 11 because the protocol changed incompatibly. X10 was quite usable and very similar to X11, but it had a number of annoying limitations.
I don't know of any released or usable versions prior to X10 (keep in mind that, traditionally, MIT projects didn't use major and minor numbers for their software, but just integers, so 10 isn't that high a version number).
I doubt that there will be an X12. There doesn't seem to be any reason at this point to change the protocol incompatibly since X11 is quite general; most likely, all further changes to the protocol will be implemented as extensions or additions.
It definitely doesn't help when every conversation about how to improve X and fix its major flaws devolves into a bunch of zealots proclaiming how perfect it is and that they see no performance issues that might VASTLY hinder adoption of X as a desktop windowing system.
If you want Macintosh-like redraw behavior (and pay with Macintosh-like memory footprints), enable backing store. Problem solved.
Furthermore, you fail to distinguish between the X11 protocol design (which is quite good) and the XFree86 server implementation (which is quite messy and clunky).
What X11 lacked was a scalable drawing model and alpha compositing. That was a serious limitation given the expectation of modern desktop users and Keith has thankfully addressed it.
What X11 might also benefit from would be an optional server-side stored vector graphics extension, somewhat similar to the Macintosh, but only used by applications for which it makes sense.
However, if you or he or anybody else were to try to turn X11 into GDI+ or Quartz, you can expect lots of resistance: the X11 communications protocol and core design has turned out to be far more versatile and useful than either GDI+ or Quartz.
Okay, every time we talk about fixing it and proposing ideas, we get flamed out of existance.
It depends on the ideas you propose. See above. Contributing in one area doesn't give you the right to endanger functionality that many people rely on.
And if you were to start arguing about X11's evolution without taking into account its non-desktop uses, you would deserve to be flamed off any X11 project in my opinion: it's the scientific, engineering, and research uses of X11 that have kept the system alive for so long; don't even think about endangering those just to achieve a better desktop experience.
Actually, Quartz does not draw line by line. Quartz preserves a bitmap of each window, and when something damages the contents of the window, restores the image from that bitmap. This takes up tons of memory (need window buffers even if window is hidden) but eliminates any streaking.
Let me add to this that you can enable this behavior on most X11 servers as well (including XFree86); it's called "backing store". It's not enabled by default because, traditionally, X11 users preferred more graphics memory to faster redraws.
Furthermore, any distribution with Macintosh-like aspirations could just enable this feature by default if they liked; the user would never have to think about it. Maybe Mandrake and RedHat should do that so that newbies stop complaining about it.
To perform this test, start with a good web browser (firefox, mozilla, konqueror, galeon, whatever). [...] What happens is smearing. Gross. Ugly. Unacceptable. Call me picky, but I don't care how much hardware you support, or how popular you are, or whatever -- if your graphical system isn't good at *drawing graphics*, then it sucks.
What "sucks" is the toolkits used by those browsers, which all pretty much ignore X11's conventions and design decisions for redraw logic. Qt and Mozilla probably do that because they were designed with Windows in mind; I don't know what excuse Gtk+ has.
And this is what people notice when they first sit down in front of a linux machine. And it's killing us. Whatever the shortcomings of Windows and Macs, neither have this problem.
Both Windows and Mac "solve" this problem by storing enormous amounts of data in the display subsystem so that the display subsystem can avoid all that ugliness.
If you want the same behavior with XFree86, you can have that: just turn on backing store. In fact, applications like Mozilla should just give you the option to do it, given that that's the behavior they assume they are getting from the server anyway.
So this licensing issue is good news, if it can galvanize the community to pull more resources into developing alternatives to XFree86 (because it sucks!).
The XFree86 licensing issue is irrelevant. X11 is a solid, proven, well-defined standard that has been around for decades and it isn't going away because one server has licensing issues. It also isn't going to go away because applications like Mozilla use it incorrectly.
The one thing that X11 still needs is server-side stored vector graphics, but it's going to get that soon. With that, it will be functionally a superset of all the major window systems. The fact that people will continue to write lousy software running under X11 isn't X11's fault.
What's even more important IMHO, is that if the community wants to redesign or do a major change in the graphics subsystem layer, it should be done NOW, before Linux desktop becomes widely used. Just look at the serial and parallel ports at the back of your computer.
The Linux "graphics subsystem layer" is clearly specified by the X11 protocol and its defined extension (SHM, RENDER, etc.). That's not going to change. The existing protocol is quite good and there simply is no reasonable alternative around, not even a reasonable design for an alternative.
XFree86 is just one implementation of many of the X11 protocols. If the XFree86 implementation disappared (like it may if the license problems are real), then some other X11 server would replace it. The troubles of one implementation of X11 are not going to have much impact. That's, after all, one of the strengths of standards-based systems.
No really, XFree86 situation seems to be a mess at the moment, let's hope that interested parties (developers from KDE, GNOME, QT, Mandrake, RedHat, IBM etc.) will use it to reach a consensus on the whole desktop thing. It's now or never.
The consensus is here: it's X11 with RENDER, and it isn't going to change. So, that means: it's never. Whether you like it or not, you better get used to it.
Slashdot Mirror
Berkely DB is a transactional database; it will not necessarily work properly if multiple processes are accessing it at the same time
Ummm, working properly when "multiple processes [...] accessing it at the same time" is the whole point of having a "transactional database".
When you make claims like this its helpful to provide examples.
Examples? Nethack, Omega, XConq, FreeCiv, xblast, xtrek, xtb, and many others. Browse around the gaming-related USENET groups from the 1980's and 1990's. In fact, many of those wonderful commercial games you so much admire are copies (with more graphics but worse game play) of open source games.
Of course, in absolute numbers, there aren't as many of those OSS games around--while something like nethack gets maintained for 20 years, you might get a dozen releases from a commercial gaming company, all with roughly the same gameplay but new graphics.
Its also interesting that you compare all of one groups games versus only once genre
Some people find the oddest things interesting.
Arguement, failing. Straws...are all...that...are left...to grasp...
Uh, huh. We'll see in another decade or so. I predict we'll have a rich selection of high-quality OSS games in the currently hot commerical genres. It will start off as running "total conversions" and upgraded freely released commercial game content on new OSS engines and then gradually move to entirely new content.
It's possible that the rapid development of technology as we've seen it over the last century/centuries will allow humanity to meaningfully colonize Mars within the next two to three centuries. I think the space elevator is a piece of the puzzle since it will allow very low cost orbital deployment.
Yes, it may be technologically possible to put a colony on Mars even within a couple of centuries. But that simply has no bearing on where we send our planetary probes today. If Venus were scientifically more interesting, we should be sending probes there despite its tough conditions.
I don't see that happening. Commercial games typically take 18 months to complete. An open source project will take longer than that if it's worth playing. I can't see unpaid virtual teams sticking together for that long - it's too much effort for too little reward.
So? Many open source projects have been around for decades, not just 18 months. The ability to do development for the long haul is where open source has a huge advantage. And the different time scales are also why you are only now beginning to see open source equivalents of games like Quake: open source really is in it over the long run.
And games are much more of an artistic endeavour than a purely technical challenge of writing an operating system where there's often a provable "best way" of doing something. Operating systems are a means to an end - games exist only for the sake of the game, hence everyone wants things to work their way. Expect infighting, splinter groups and a lot of projects not getting finished.
I think you really misunderstand software if you think that there's a "provably 'best way'" of designing a word processor, kernel, or graphics editor.
In any case, if the game is designed well, that kind of dynamic will improve things: competing models, competing looks, multiple interacting worlds, etc.
In fact, there are plenty of open source games out there that are very popular and have stood the test of time. That's a lot more than can be said for many of those commercial FPS's out there.
The only reason you see so many commercial games with no apparent OSS counterpart is because commercial vendors push the limits of what hardware can do; OSS, for various reasons, is far more conservative, but no less inventive. And in terms of actual game play, most commercial games really aren't that hot.
Assuming those were good reasons to colonize other planets, the time frame of those issues is so long-term (as is the time-frame in which we will be technologically ready to actually colonize other planets in any meaningful way) that it really has no bearing on which planets to target by unmanned probes for the foreseeable future.
In this century and the next, we should pick our targets for maximum scientific benefit, not for the possibility of colonization.
You could ask the same about any game genre: why are the free, open-source first-person shooters lagging? The large amount of work that goes into any game means that they're not as easy to develop.
Neither is the Linux kernel or many other open source projects. Furthermore, there are plenty of "total conversions" of existing video games that involve completely new models of everything.
No, the reason why you don't see that stuff is cultural and sociological: the kind of people who do that sort of thing generally are not very much into open anything. Many of them come out of a shareware or "freeware" tradition that happens to be closed source and builds on commercial building blocks.
But it's going to change over time: as open source provides better tools engines and better graphics tools, you'll see more of that energy being poured into open source games of every genre.
There are also those who of course, believe that Mars is chiefly where we will dump those extra billions of people we are going to have in the next 100 years.
You're gonna ship African and Indian street kids to Mars by the billions? Because that's where those "extra billions" come from. It's not going to happen.
There are only two ways we will deal with the population explosion: family planning and social changes on the one hand, or disease and starvation on the other.
Opening up new horizons is part of fixing the problems on Earth. Not only is the space program generating research that is highly applicable to current Earthly problems*, [...] * Such as medicine, hydroponics, closed ecosystems, energy sources, micro-mechanization, robotics, etc.
What good do lessons about ecology do us when we don't apply them? We have more than enough technology to solve the problems here on earth--the obstacles are purely social and political. Colonizing Mars isn't going to solve that.
it is also providing for an eventual safety valve where disaffected members of society can go off to a Moon or Mars colony to start a new life rather than remaining on crowded Earth.
A six months interplanetary trip followed by enormously expensive life support is supposed to be a "safety valve"? Sorry, but you are naive.
An example of this can be seen in the early pioneering days of the United States. Sure it was a tough, difficult, and often deadly trip west but many people did it anyways in order to start off new. Many of those people had been feeling stifled in the eastern cities and so they went west, relieving the social caldrons which were beginning to boil over.
First of all, the US was, in many ways, a paradise: abundant wildlife, good climate, rich soils. Second, many people came lured by false advertising: despite nearly ideal conditions, life was indeed harsh (as life tends to be without tools and infrastructure). Third, emigration to the US hardly did much "relieving" for the countries where people emigrated from; quite to the contrary, after being a dumping ground for dopes and criminals initially, the US soon took away many of their most enterprising citizens, a "brain drain" that continues to this day. This continues to benefit the US at the cost of everybody else.
Of course, none of that is relevant to Mars: Mars is far less suitable for colonization than the Sahara desert, the top of Mount Everest, or the Antarctic. And in addition to that, it's much more costly to reach.
We either fix things here on earth or we die: colonization of other planets is not a viable alternative over the next couple of centuries at least.
Mars can be inhabited and explored by humans,
In roughly the same sense that the moon can be, and lots of other bodies. But what's the point?
but first thing is first.
You're naive if you think that we are sending probes to Mars because we want to colonize it in the near future. Maybe Bush thinks that, maybe Dan Quayle thinks that, but it's just not going to happen any time soon.
The people who actually are making the exploration of Mars and the rest of the solar system happen don't want to colonize those bodies either. In fact, many astrobiologists would be horrified by the notion if it ever became a realistic possibility.
Colonizing Mars or Venus, let alone terraforming it, are fortunately so far out that nobody has to worry about it. The moon would be a much better place to "colonize" for the time being if we really wanted colonies in space. Of course, nobody has yet come up with a rational reason why we would want to.
As for Venus, well, right now, cooler worlds are just the more interesting ones: when so much of the budget is spent on military nonsense, scientists have to pick and choose their targets carefully.
Yes, film is pretty much doomed (except for niche applications). But Kodak has seen this coming and started preparing in time. I think among old companies that needed to transform themselves, Kodak has been doing pretty well: their digital camera lineup is decent, they have done some nifty stuff with OLED, and they still have lots of non-consumer products that probably make them money. They also were one of the first companies to actually sell digital cameras widely. Kodak isn't a hot company, but give the guys a break on this one--they haven't been blind and they have been trying to go for the new market.
What is really dragging Kodak down is their brand name--some companies have a brand name that stands for innovation, and they can put out any kind of garbage and people will think it's the latest and greatest thing. Kodak, on the other hand, can put out a really nifty digital camera and the stale odor of photographic fixing solution clings to it in the mind of buyers (yes, including my own).
I hope you do understand that it is the US that is pushing for this. Basically, the US is saying that the Europeans need to create biometric passports if they want to be able to continue to travel to the US under the visa waiver program.
What is particularly bad about this is that US citizens will apparently not be subjected to biometric identification, at least not yet. It actually seems rather doubtful that something like that is palatable domestically, although in the hysteria after 9/11, I suppose you never know.
What is also bad about it is that biometrics give people a false sense of security: they don't actually increase security very much (since they can be forged), but they tend to decrease human security checks--immigration officials trying to judge whether someone is behaving in odd ways, etc.
I hope that the Europeans are going to refuse and say "OK, kill your tourism market if you like"--the US has much more to lose from making it harder for Europeans to travel to the US than the other way around. But, politically, that doesn't look likely.
Call me crazy, but I think the US having the ability to rain down death and destruction on anyone who gets in our way does make the world a safer place.
Yes, you are crazy. If the US were to succeed, it would do anything from incur serious economic and political sanctions to increased anti-American sentiments and increased terrorism. One thing it wouldn't do is make America safer.
For Americans. And those are the only people our tax dollars should be protecting in the first place.
Except, of course, that the US tax payer isn't paying for this, the US is borrowing from other nations to pay for it.
Don't like it? Go get your own military for once.
Why? Another decade or two of this insane military spending and foreign nations will own anything that is worth owning in the US.
You've got to be kidding. Traditionally, few civilians got killed in warfare. Advanced military technology has led to an escalation in civilian deaths and a huge reduction in military deaths, in a perverse reversal of an already perverse means of settling disputes. And wars have become more frequent, not less frequent.
As for nuclear weapons being "proven deterrents", perhaps you forget how close we came to global nuclear holocaust on several occasions during the cold war. If either the US or the USSR leaders had gotten off on the wrong foot, we'd all be history. Nuclear weapons nearly killed us all.
Ruby likes to pretend it's some kind of Smalltalk successor. It isn't. Ruby is a Perl successor that borrowed a few tidbits from Smalltalk.
Sun has to open source their JVM.
Oh, dear, I did it too, referring to the platform as the "JVM". No, it's not sufficient for them to open source their JVM; if they want to make Java an open platform, at this point, they would have to open source their entire JDK 1.5 implementation, without conditions or strings attached. But, for the reasons I already mentioned, I don't think a truly open Java is even desirable. Java should just get replaced instead.
Sun needs to do two things.
First, they need to open up the Java standard unconditionally, so that anybody who wants to implement it can do so without restrictions or interference from Sun; right now, the Java 2 platform specifications are covered by licenses, and conforming implementations would probably run into several Sun patents.
Second, because Java has become so bloated and Sun has effectively killed off competing efforts, it is far too late to hope for independent implementations of the Java platform, so if Java is ever going to be an "open standard" in any meaningful sense, Sun has to open source their JVM.
Of course, at this point, I hope they won't. I think technically, Java has become a lost cause. It's easier to replace it than to fix it.
I used to want Sun to open source Java, but they've actually been a pretty good steward and I quite like what they're doing with it. The Java Community Process seems to be working.
Working in what way? In the sense of producing a language that works for some people? Sure. But the same can be said for Microsoft and VisualBasic.
The real problem is that the Java core is heavily covered by Sun intellectual property (restrictions on the specifications, patents, copyrights). That means that all this wonderful free work that the JCP puts in around the periphery ends up effectively contributing only to a Sun-controlled platform.
I'm tired of Sun claiming that they are a friend of open source and getting away with it. Sun broke their promise to have Java standardized by an independent organization twice. Sun's licenses on the Java specifications and their Java implementations are reprehensible: if you only as much as look at them, they have legal claims over your Java-related work. Sun has threatened various open source projects and forced them to agree to Sun's licenses on the specifications. The JCP makes people work for free to benefit mostly Sun. And Sun keeps announcing that open source isn't good enough, that only Sun managed to turn Gnome into something decent, and that commmercial users will be flocking away from Linux to Solaris. Furthermore, technically, Java evolves at a snail's pace.
Sun used to be a decent company and there was much hope initially for Java. But almost a decade later, I think Sun and Java are bad news for open source. As Sun's fortunes decline (and they will because they are technically not very good anymore and way overpriced), they will become vicious and they'll start taking advantage of all the intellectual property they have on Java.
Children could be taught to program early in their school years.
Yes, and that's what we have languages like Logo for. Or, if you want something readable and more adult, Smalltalk (and the Squeak.org environment). Python, of course, originally also was designed for education.
Ruby is a nice language, but I don't think it comes out of an educational background. It syntax is cleaner than Perl's but doesn't seem like it would be all that intuitive to non-computer users either (both Smalltalk and Logo seem better in that regard).
Altogether, Ruby just seems like an odd choice when it comes to really caring about teaching people to program. Not a bad choice, but not clearly better than the more obvious choices either.
X10 did indeed exist. They changed the major version number to 11 because the protocol changed incompatibly. X10 was quite usable and very similar to X11, but it had a number of annoying limitations.
I don't know of any released or usable versions prior to X10 (keep in mind that, traditionally, MIT projects didn't use major and minor numbers for their software, but just integers, so 10 isn't that high a version number).
I doubt that there will be an X12. There doesn't seem to be any reason at this point to change the protocol incompatibly since X11 is quite general; most likely, all further changes to the protocol will be implemented as extensions or additions.
It definitely doesn't help when every conversation about how to improve X and fix its major flaws devolves into a bunch of zealots proclaiming how perfect it is and that they see no performance issues that might VASTLY hinder adoption of X as a desktop windowing system.
If you want Macintosh-like redraw behavior (and pay with Macintosh-like memory footprints), enable backing store. Problem solved.
Furthermore, you fail to distinguish between the X11 protocol design (which is quite good) and the XFree86 server implementation (which is quite messy and clunky).
What X11 lacked was a scalable drawing model and alpha compositing. That was a serious limitation given the expectation of modern desktop users and Keith has thankfully addressed it.
What X11 might also benefit from would be an optional server-side stored vector graphics extension, somewhat similar to the Macintosh, but only used by applications for which it makes sense.
However, if you or he or anybody else were to try to turn X11 into GDI+ or Quartz, you can expect lots of resistance: the X11 communications protocol and core design has turned out to be far more versatile and useful than either GDI+ or Quartz.
Okay, every time we talk about fixing it and proposing ideas, we get flamed out of existance.
It depends on the ideas you propose. See above. Contributing in one area doesn't give you the right to endanger functionality that many people rely on.
And if you were to start arguing about X11's evolution without taking into account its non-desktop uses, you would deserve to be flamed off any X11 project in my opinion: it's the scientific, engineering, and research uses of X11 that have kept the system alive for so long; don't even think about endangering those just to achieve a better desktop experience.
Actually, Quartz does not draw line by line. Quartz preserves a bitmap of each window, and when something damages the contents of the window, restores the image from that bitmap. This takes up tons of memory (need window buffers even if window is hidden) but eliminates any streaking.
Let me add to this that you can enable this behavior on most X11 servers as well (including XFree86); it's called "backing store". It's not enabled by default because, traditionally, X11 users preferred more graphics memory to faster redraws.
Furthermore, any distribution with Macintosh-like aspirations could just enable this feature by default if they liked; the user would never have to think about it. Maybe Mandrake and RedHat should do that so that newbies stop complaining about it.
To perform this test, start with a good web browser (firefox, mozilla, konqueror, galeon, whatever). [...] What happens is smearing. Gross. Ugly. Unacceptable. Call me picky, but I don't care how much hardware you support, or how popular you are, or whatever -- if your graphical system isn't good at *drawing graphics*, then it sucks.
What "sucks" is the toolkits used by those browsers, which all pretty much ignore X11's conventions and design decisions for redraw logic. Qt and Mozilla probably do that because they were designed with Windows in mind; I don't know what excuse Gtk+ has.
And this is what people notice when they first sit down in front of a linux machine. And it's killing us. Whatever the shortcomings of Windows and Macs, neither have this problem.
Both Windows and Mac "solve" this problem by storing enormous amounts of data in the display subsystem so that the display subsystem can avoid all that ugliness.
If you want the same behavior with XFree86, you can have that: just turn on backing store. In fact, applications like Mozilla should just give you the option to do it, given that that's the behavior they assume they are getting from the server anyway.
So this licensing issue is good news, if it can galvanize the community to pull more resources into developing alternatives to XFree86 (because it sucks!).
The XFree86 licensing issue is irrelevant. X11 is a solid, proven, well-defined standard that has been around for decades and it isn't going away because one server has licensing issues. It also isn't going to go away because applications like Mozilla use it incorrectly.
The one thing that X11 still needs is server-side stored vector graphics, but it's going to get that soon. With that, it will be functionally a superset of all the major window systems. The fact that people will continue to write lousy software running under X11 isn't X11's fault.
What's even more important IMHO, is that if the community wants to redesign or do a major change in the graphics subsystem layer, it should be done NOW, before Linux desktop becomes widely used. Just look at the serial and parallel ports at the back of your computer.
The Linux "graphics subsystem layer" is clearly specified by the X11 protocol and its defined extension (SHM, RENDER, etc.). That's not going to change. The existing protocol is quite good and there simply is no reasonable alternative around, not even a reasonable design for an alternative.
XFree86 is just one implementation of many of the X11 protocols. If the XFree86 implementation disappared (like it may if the license problems are real), then some other X11 server would replace it. The troubles of one implementation of X11 are not going to have much impact. That's, after all, one of the strengths of standards-based systems.
No really, XFree86 situation seems to be a mess at the moment, let's hope that interested parties (developers from KDE, GNOME, QT, Mandrake, RedHat, IBM etc.) will use it to reach a consensus on the whole desktop thing. It's now or never.
The consensus is here: it's X11 with RENDER, and it isn't going to change. So, that means: it's never. Whether you like it or not, you better get used to it.