Neither the police nor SGAE did anything. Why? Because it was legal.
Guess what, the SGAE didn't do anything illegal either, they just picked up the phone and asked his boss to get him to resign. So, both he and the SGAE each did something private and legal. What are you getting so upset about, then (rhetorical question)?
I'm sympathetic to his cause, but he went about this in an ineffective and stupid way. If he wanted to be a martyr to the cause, at least he should have done it in a way that resulted in them doing something actionable (like a wrongful termination).
It doesn't matter who came first. Eclipse is the free, vendor independent system, and it has by far the larger user community. Unlike Netbeans, Eclipse also runs on entirely free platforms, which is one of the reasons it can ship with Linux.
Netbeans may be "coming back with a vengeance", but it is still just an IDE, and Eclipse is still the IDE that plugins are written for.
Sun's competition with Eclipse is wasteful of their own resources and confusing to the market. And it's pointless: no matter how good a job Sun does on Netbeans, they have nothing lined up that is compelling enough to overcome the huge advantage that Eclipse has.
Well, you haven't told us what platform you run on (Windows?). Under X11, Swing sucks horribly, and not just in terms of performace. The worst part of Swing is that it almost looks like a native toolkit, but it behaves wrong in so many ways.
There are decent cross platform toolkits. There are even decent cross platform toolkits that do their own rendering. Swing is not one of them.
NetBeans may be a decent IDE, but is it substantially better than Eclipse? Does it have any specific, compelling advantages? If not, then what's the reason to use it or waste time on its development?
Eclipse not only is fully open source, it's what everybody is developing plug-ins for. And, unlike NetBeans, Eclipse actually runs on open source implementations of Java, which means that it ships with Linux distributions.
Why does Sun keep wasting resources on NetBeans? Don't they have anything better to do?
Sun has megabytes of unfixed bugs, but instead of focusing there, they are trying to compete with a highly successful, well-written free tool. This push for NetBeans ultimately comes from Sun's pathological desire to own and control everything: Sun absolutely hates the fact that Eclipse doesn't require their proprietary toolkit and that Eclipse can compile with open Java tools. Sun wants a desktop based on Sun Java, a server-side platform based on Sun Java, an office suite based on Sun Java, and an IDE based on Sun Java. It's really the same thing Microsoft is doing, only that resource constraints and public opinion constrain them a little more (e.g., they can't start a new GUI project from scratch, they just have to hack Sun Java into Gnome).
Eclipse is the only sensible choice for a Java IDE at this point: NetBeans may be a little better in some areas, and Eclipse in others, but those differences are minor. The deciding factor is that Eclipse has become the de-facto standard platform for plug-ins.
Actually, VNC is horrendously slow on OSX too, but it's very fast on X11 i agree, mainly because X is designed with remote displays in mind.
The fact that VNC is fast with X11 is only indirectly related to X11's network transparency. The real problem with implementing something like VNC or RDP is that you need to hook into the OS. X11 makes that easy because the X11 server is a user-mode process and is open source.
On Windows, there are hooked versions of X11 that are very fast; but they are tricky to install and are OS version dependent because there is no defined API for doing tihs.
On Macintosh, the display server is closed source, it doesn't have hooks either, and nobody has bothered to reverse engineer it. That's why VNC with the native Macintosh window system is slow as well.
Incidentally, exporting the primary screen via x11vnc is fairly slow on X11 as well, for the same reason it's slow on Windows and Macintosh: the hooks just aren't there yet. However, the Xdamage extension will provide the necessary hooks.
Quite to the contrary: X is actually quite high-level.
and the fact that it is "backwards" from the perspective of accessing a server. To get X working over NAT, you need something like X-forwarding with SSH
That's because X11 is a network transparent window system, while RDP is merely a remote desktop protocol. And, as you observe, there are simple, modular tools that give you the ability to switch the direction of the connection.
Not to mention that you *need* SSH or an equivelent because X doesn't have built-in encryption.
Geez, imagine that: a protocol that actually layers and separates different functionality. That kind of approach to software and protocol design must come as quite a shock to you!
Not to mention that GTK+ and QT both backbuffer their drawing, so you effectively are transmitting large bitmaps accross the link
Yeah, Gtk+ and Qt have crappy X11 bindings; what else is new? But they work well enough for enough people that most people don't really lose much sleep over it.
in some cases without any encryption.
I very much hope that neither Gtk+ nor Qt ever encrypt; that is not their function.
No, RDP isn't "new", but it's far better implemented than X.
X isn't an implementation, it's a protocol, and a damned good one at that. It has been around longer than Windows and Macintosh, it has remained stable, has dozens of independent, high-quality implementations, runs on dozens of operating systems, and has hundreds of toolkits supporting it. X survived even though people often had to uninstall their native window systems and install MIT X11 on their workstations in order to use it; it's survived because users demanded it. X will likely be around long after Microsoft and Apple have to rewrite their window systems again, and nothing Microsoft or Apple have ever done has had that kind of longevity.
RDP is usable over 1xRTT. Try that with X.
X11 runs fine at those speeds. But X11 is actually optimized for Ethernet speeds, something where RDP peforms poorly.
Unfortunately, you exemplify the same problem Microsoft frequently has: you don't even understand the problems. You take a superficial look at X11, you erroneously think you know what it does, and then you conclude that there is a quicker and simpler solution and that RDP does it better. That kind of ignorance is why Microsoft keeps producing such poor software, and why people keep buying it, even though it clearly is not a good solution to their problems (ignorance isn't bliss in this case, it just perpetuates the problem).
He proposed to demo what sort of legal uses one could make of copyrighted works from P2P networks, and informed the Spanish collecting society, the national police and the attorney general to let them know what he was up to.
Well, he was kind of asking for it. If the download was legal, he didn't have to inform the "collecting society" or the police about it. But he put it as a challenge, to both the collecting society and the police, and they rose to the challenge. From their point of view, what he announced he was going to do was arguably illegal, and they stopped it.
I'm not sure what he wanted to achieve, but whatever it was, this doesn't seem like this was the best way of going about it. At least, he could have forced the university to fire him, in which case he might as well have sued for wrongful termination, but he closed even that possibility.
If you don't see why encryption can solve this problem, then you don't have a technical mind.
And the problem is that you do. If there is no separate pin, you don't have to break encryption, a bad guy can simply carry out completely normal purchases with a normal credit card terminal through a normal, legitimate terminal while the card is still in the person's wallet. This has lots of potential for both criminal abuse and just simple problems, like unintended multiple charges. And since many credit cards are linked to debit cards now, you don't even have the usual purchase protections.
The acts of physically swiping a card and of signing a piece of paper both are important security features. Tampering with the process make it work less well.
Don't get me wrong folks, OSS is great, and I would love to see the beast from redmond defeated, but Terminal Services/Remote desktop is a solution done right, give credit where its due.
The credit belongs to the creators of X11; RDP is just your typical "me-too" product from Microsoft, a decade late and proprietary.
VNC is something different from both RDP and X11, but VNC is useful and "done right" in its own way.
Terminal Services/Remote desktop is a nice solution period. Its fast, easily an order of magnitude faster than even TightVNC.
That's an API issue specific to the Windows platform: Microsoft just doesn't make the right hooks available. An X11-based TightVNC server actually performs quite well.
Re:Nice Anti-Usian Propaganda, Now Some Facts
on
Cuba Switching to Linux
·
· Score: 2, Interesting
Hell, the US would be happy enough if they were just European socialists (you know, the type with taxes > 50%
Top tax rates in the US and many European countries are actually comparable. Perhaps the biggest difference is that European governments are fiscally more conservative than the US and still manage to provide more services with the money they do have available. How that reality translates into your myth of "European socialists" is a mystery to me.
The US gov't's problem with Cuba is narrowly limited to its practice of oppression.
That's total bullshit--the US has no qualms about doing business with far more oppressive regimes than Cuba. Furthermore, lots of other nations have serious concerns about US treatment of its own citizens, but that doesn't give them the right to launch assassination attempts against US heads of state or to orchestrate economic embargoes against the US.
The reality simply is that Cuba is run by a corrupt and incompetant military dictator
Nobody who manages to stay in power that long and survives numerous assassination attempts by the US secret service is completely incompetent. Furthermore, think of all the nations where the US has interfered militarily or economically; some of them have done better than Cuba, but a lot of them have not.
The complete mismanagement of the economy by his everlasting regime led to scarcity, and the spoils system inherent in any communist regime has led to a disparity whereby most Cubans live in abject poverty, but the priveledged few live in opulant comfort.
Cuba is not like North Korea; we actually have fairly good information about the country from many sources and that doesn't square with any information that's coming out of Cuba.
The most straightforward answer to your claim is the Gini index, a formal measure of income distribution. Cuba's Gini index seems to be comparable to that of egalitarian societies like Sweden and Norway and much lower than that of the US.
I hope Cuba will become democratic and that the Cuban people will get to choose freely what kind of economic system they want (which may well not be US-style capitalism, which is still fairly unique in the world). I also hope that political change in Cuba won't come at the hands of US assassins; the US should get out of the regime change business.
The idea is notnew, and there is even an open source application (and as an open source developer, let me say: it's about time that we can touch each other without proprietary protocols).
You're right--that's simpler than I thought. The web page is a bit confusing, because the install-from-CD instructions are tacked onto the description, followed by a big, major section with what turns out to be old install instructions.
People invest so much time in figuring out how to run this sort of thing from a USB stick, and then they fail to make it trivial to install it on a stick.
Please, instead of lots of DOS commands, create a little self-contained application (e.g., in FLTK) that pulls over all the necessary files, finds the memory stick, copies everything over, and makes the thing bootable.
That's particularly important given one of the likely user communities for these kinds of Linux distributions: people who want to start experimenting with Linux without devoting a whole machine to it.
Apple? The company that started look-and-feel lawsuits and wanted to keep everybody in the world from using GUIs? Apple patents a lot, and they try to enforce their patents--they are no better in this regard than Microsoft, just smaller.
If you want to do something against these sorts of patent system abuses, use (and contribute to) open source software on generic hardware.
Patents are written in legalese, not in english. Unless you are a patent lawyer, I don't think you can really judge them.
Patents are supposed to be written by practitioners for practitioners. If only lawyers can understand what a patent means, then the patent is a bad patent no matter what it covers (and it may also be invalid).
This is not a patent for autocomplete. It is much more specific.
Yes, it is more specific: it is autocompletion for Email addresses, which is a straightforward extension of autocompletion for everything else.
We did innovate in this space in MacOE.
No, they did not. The fact that they thought they innovated is just a testament to their ignorance.
David Reed doesn't know what he is talking about. Interference is very real and well defined, independent of the particular receiver, once you have decided on frequency and bandwidth allocations for different users.
Historically, because of technological limitations, receivers have gotten more bandwidth than they needed for their tasks, which allowed them to be more robust to interference, but the interference was still there.
Somehow, I had hoped that the Presidio would be home to something with a bit more class and style than Lucas and his company. It's just such a beautiful and unique spot.
If anything is going to kill Linux and the open-source movement, it's the presence of certifiable lunatics in the ranks representing the users.
Here is a clue: there are certifiable lunatics present in almost any random group of people that has more than maybe a hundred people in it. Human beings deal as best as they can with that reality through things like the police and the courts. If someone made death threats against O'Gara, it's a matter for the police and has nothing to do with Linux.
Unless the community gets a handle on this, grows up, and rebukes the extremists, the trash heap of history is where this is all headed.
Well, I guess it's the trash heap of history then, because no group of people has ever been able to get a handle on this. Why, we can't even get computer columnists making inflammatory remarks on matters they know little about to shut up, and those people are not even necessarily certifiable lunatics.
This doesn't seem to generate much of a performance hit, so I wonder why Microsoft is going for a different approach? Apple's seems to make more sense.
Both designs have been around for quite some time. Updating the index immediately when a file changes may be OK for desktop use, but it has unpredictable consequences for other kinds of uses. The conservative and safe thing is to update on a schedule unless the user explicitly requests otherwise.
Apple's implementation is integrated into the filesystem layer, so it indexes everything as it is written to disk
"Integrating" any of this in the file system layer is unnecessary. All you need is update notifications from the kernel to a user mode index server, a tiny addition to most kernel, and one that Linux has supported for a while now.
Microsoft has a Tabbed Browser called ".NET SDK Documentation". Rather than being "basic", you can drag-n-drop the tabs and Z-Order works. In other words, they're more advanced than Firefox tabs and hopefully that's what they'll use for IE7.
Sorry, that's not particularly original to Microsoft. There are several extensions for Firefox that offer such features, as well as several other browsers based on the Mozilla renderer (Galeon tabs work really nicely out of the box). The concepts themselves are older than that.
Anyone can produce a JVM. There are several good FOSS ones out there now, as well as AOT compilers with GCJ nearly fully complete. There are NO legal impediments to ANYONE producing their own JVM and there is NO Sun-only bits in OOo.
Anyone could implement GIF as well; people found out that there were intellectual property problems with GIF only years later. Anyone could implement FAT or SMB, but Microsoft is now claiming that they own intellectual property in those standards. Anyone could implement a UNIX-like operating system, but SCO now claims that they own it all and has started a billion dollar lawsuit over it. The open source community needs to avoid such problems because they are disruptive and expensive.
In the case of Java, just read the damned licenses on Sun's site. You will see that Sun explicitly states that many Java specifications belong to Sun. And go check on Sun's patents: there are a dozen different patents on important Java technologies.
Quit spreading FUD.
Quit trying to trick people into using proprietary standards by pretending that they are free.
Sun is a failing computer maker with an army of lawyers, they have a widely used proprietary platform, their licenses clearly state that they own the platform and the APIs, and they are "contributing" functionality that depends on their proprietary platform to several open source projects. Add two and two together.
Slashdot Mirror
Neither the police nor SGAE did anything. Why? Because it was legal.
Guess what, the SGAE didn't do anything illegal either, they just picked up the phone and asked his boss to get him to resign. So, both he and the SGAE each did something private and legal. What are you getting so upset about, then (rhetorical question)?
I'm sympathetic to his cause, but he went about this in an ineffective and stupid way. If he wanted to be a martyr to the cause, at least he should have done it in a way that resulted in them doing something actionable (like a wrongful termination).
It doesn't matter who came first. Eclipse is the free, vendor independent system, and it has by far the larger user community. Unlike Netbeans, Eclipse also runs on entirely free platforms, which is one of the reasons it can ship with Linux.
Netbeans may be "coming back with a vengeance", but it is still just an IDE, and Eclipse is still the IDE that plugins are written for.
Sun's competition with Eclipse is wasteful of their own resources and confusing to the market. And it's pointless: no matter how good a job Sun does on Netbeans, they have nothing lined up that is compelling enough to overcome the huge advantage that Eclipse has.
Well, you haven't told us what platform you run on (Windows?). Under X11, Swing sucks horribly, and not just in terms of performace. The worst part of Swing is that it almost looks like a native toolkit, but it behaves wrong in so many ways.
There are decent cross platform toolkits. There are even decent cross platform toolkits that do their own rendering. Swing is not one of them.
NetBeans may be a decent IDE, but is it substantially better than Eclipse? Does it have any specific, compelling advantages? If not, then what's the reason to use it or waste time on its development?
Eclipse not only is fully open source, it's what everybody is developing plug-ins for. And, unlike NetBeans, Eclipse actually runs on open source implementations of Java, which means that it ships with Linux distributions.
Why does Sun keep wasting resources on NetBeans? Don't they have anything better to do?
Sun has megabytes of unfixed bugs, but instead of focusing there, they are trying to compete with a highly successful, well-written free tool. This push for NetBeans ultimately comes from Sun's pathological desire to own and control everything: Sun absolutely hates the fact that Eclipse doesn't require their proprietary toolkit and that Eclipse can compile with open Java tools. Sun wants a desktop based on Sun Java, a server-side platform based on Sun Java, an office suite based on Sun Java, and an IDE based on Sun Java. It's really the same thing Microsoft is doing, only that resource constraints and public opinion constrain them a little more (e.g., they can't start a new GUI project from scratch, they just have to hack Sun Java into Gnome).
Eclipse is the only sensible choice for a Java IDE at this point: NetBeans may be a little better in some areas, and Eclipse in others, but those differences are minor. The deciding factor is that Eclipse has become the de-facto standard platform for plug-ins.
Actually, VNC is horrendously slow on OSX too, but it's very fast on X11 i agree, mainly because X is designed with remote displays in mind.
The fact that VNC is fast with X11 is only indirectly related to X11's network transparency. The real problem with implementing something like VNC or RDP is that you need to hook into the OS. X11 makes that easy because the X11 server is a user-mode process and is open source.
On Windows, there are hooked versions of X11 that are very fast; but they are tricky to install and are OS version dependent because there is no defined API for doing tihs.
On Macintosh, the display server is closed source, it doesn't have hooks either, and nobody has bothered to reverse engineer it. That's why VNC with the native Macintosh window system is slow as well.
Incidentally, exporting the primary screen via x11vnc is fairly slow on X11 as well, for the same reason it's slow on Windows and Macintosh: the hooks just aren't there yet. However, the Xdamage extension will provide the necessary hooks.
Except that X is far more low-level than RDP,
Quite to the contrary: X is actually quite high-level.
and the fact that it is "backwards" from the perspective of accessing a server. To get X working over NAT, you need something like X-forwarding with SSH
That's because X11 is a network transparent window system, while RDP is merely a remote desktop protocol. And, as you observe, there are simple, modular tools that give you the ability to switch the direction of the connection.
Not to mention that you *need* SSH or an equivelent because X doesn't have built-in encryption.
Geez, imagine that: a protocol that actually layers and separates different functionality. That kind of approach to software and protocol design must come as quite a shock to you!
Not to mention that GTK+ and QT both backbuffer their drawing, so you effectively are transmitting large bitmaps accross the link
Yeah, Gtk+ and Qt have crappy X11 bindings; what else is new? But they work well enough for enough people that most people don't really lose much sleep over it.
in some cases without any encryption.
I very much hope that neither Gtk+ nor Qt ever encrypt; that is not their function.
No, RDP isn't "new", but it's far better implemented than X.
X isn't an implementation, it's a protocol, and a damned good one at that. It has been around longer than Windows and Macintosh, it has remained stable, has dozens of independent, high-quality implementations, runs on dozens of operating systems, and has hundreds of toolkits supporting it. X survived even though people often had to uninstall their native window systems and install MIT X11 on their workstations in order to use it; it's survived because users demanded it. X will likely be around long after Microsoft and Apple have to rewrite their window systems again, and nothing Microsoft or Apple have ever done has had that kind of longevity.
RDP is usable over 1xRTT. Try that with X.
X11 runs fine at those speeds. But X11 is actually optimized for Ethernet speeds, something where RDP peforms poorly.
Unfortunately, you exemplify the same problem Microsoft frequently has: you don't even understand the problems. You take a superficial look at X11, you erroneously think you know what it does, and then you conclude that there is a quicker and simpler solution and that RDP does it better. That kind of ignorance is why Microsoft keeps producing such poor software, and why people keep buying it, even though it clearly is not a good solution to their problems (ignorance isn't bliss in this case, it just perpetuates the problem).
He proposed to demo what sort of legal uses one could make of copyrighted works from P2P networks, and informed the Spanish collecting society, the national police and the attorney general to let them know what he was up to.
Well, he was kind of asking for it. If the download was legal, he didn't have to inform the "collecting society" or the police about it. But he put it as a challenge, to both the collecting society and the police, and they rose to the challenge. From their point of view, what he announced he was going to do was arguably illegal, and they stopped it.
I'm not sure what he wanted to achieve, but whatever it was, this doesn't seem like this was the best way of going about it. At least, he could have forced the university to fire him, in which case he might as well have sued for wrongful termination, but he closed even that possibility.
If you don't see why encryption can solve this problem, then you don't have a technical mind.
And the problem is that you do. If there is no separate pin, you don't have to break encryption, a bad guy can simply carry out completely normal purchases with a normal credit card terminal through a normal, legitimate terminal while the card is still in the person's wallet. This has lots of potential for both criminal abuse and just simple problems, like unintended multiple charges. And since many credit cards are linked to debit cards now, you don't even have the usual purchase protections.
The acts of physically swiping a card and of signing a piece of paper both are important security features. Tampering with the process make it work less well.
Don't get me wrong folks, OSS is great, and I would love to see the beast from redmond defeated, but Terminal Services/Remote desktop is a solution done right, give credit where its due.
The credit belongs to the creators of X11; RDP is just your typical "me-too" product from Microsoft, a decade late and proprietary.
VNC is something different from both RDP and X11, but VNC is useful and "done right" in its own way.
Terminal Services/Remote desktop is a nice solution period. Its fast, easily an order of magnitude faster than even TightVNC.
That's an API issue specific to the Windows platform: Microsoft just doesn't make the right hooks available. An X11-based TightVNC server actually performs quite well.
Hell, the US would be happy enough if they were just European socialists (you know, the type with taxes > 50%
Top tax rates in the US and many European countries are actually comparable. Perhaps the biggest difference is that European governments are fiscally more conservative than the US and still manage to provide more services with the money they do have available. How that reality translates into your myth of "European socialists" is a mystery to me.
The US gov't's problem with Cuba is narrowly limited to its practice of oppression.
That's total bullshit--the US has no qualms about doing business with far more oppressive regimes than Cuba. Furthermore, lots of other nations have serious concerns about US treatment of its own citizens, but that doesn't give them the right to launch assassination attempts against US heads of state or to orchestrate economic embargoes against the US.
The reality simply is that Cuba is run by a corrupt and incompetant military dictator
Nobody who manages to stay in power that long and survives numerous assassination attempts by the US secret service is completely incompetent. Furthermore, think of all the nations where the US has interfered militarily or economically; some of them have done better than Cuba, but a lot of them have not.
The complete mismanagement of the economy by his everlasting regime led to scarcity, and the spoils system inherent in any communist regime has led to a disparity whereby most Cubans live in abject poverty, but the priveledged few live in opulant comfort.
Cuba is not like North Korea; we actually have fairly good information about the country from many sources and that doesn't square with any information that's coming out of Cuba.
The most straightforward answer to your claim is the Gini index, a formal measure of income distribution. Cuba's Gini index seems to be comparable to that of egalitarian societies like Sweden and Norway and much lower than that of the US.
I hope Cuba will become democratic and that the Cuban people will get to choose freely what kind of economic system they want (which may well not be US-style capitalism, which is still fairly unique in the world). I also hope that political change in Cuba won't come at the hands of US assassins; the US should get out of the regime change business.
Ballmer should be worried about Microsoft and leave worrying about Google to others.
I don't know whether Google is going to be around longer than Microsoft or not; but I'm pretty sure it's downhill for Microsoft from this point on.
The idea is not new, and there is even an open source application (and as an open source developer, let me say: it's about time that we can touch each other without proprietary protocols).
You're right--that's simpler than I thought. The web page is a bit confusing, because the install-from-CD instructions are tacked onto the description, followed by a big, major section with what turns out to be old install instructions.
People invest so much time in figuring out how to run this sort of thing from a USB stick, and then they fail to make it trivial to install it on a stick.
Please, instead of lots of DOS commands, create a little self-contained application (e.g., in FLTK) that pulls over all the necessary files, finds the memory stick, copies everything over, and makes the thing bootable.
That's particularly important given one of the likely user communities for these kinds of Linux distributions: people who want to start experimenting with Linux without devoting a whole machine to it.
Apple? The company that started look-and-feel lawsuits and wanted to keep everybody in the world from using GUIs? Apple patents a lot, and they try to enforce their patents--they are no better in this regard than Microsoft, just smaller.
If you want to do something against these sorts of patent system abuses, use (and contribute to) open source software on generic hardware.
Patents are written in legalese, not in english. Unless you are a patent lawyer, I don't think you can really judge them.
Patents are supposed to be written by practitioners for practitioners. If only lawyers can understand what a patent means, then the patent is a bad patent no matter what it covers (and it may also be invalid).
This is not a patent for autocomplete. It is much more specific.
Yes, it is more specific: it is autocompletion for Email addresses, which is a straightforward extension of autocompletion for everything else.
We did innovate in this space in MacOE.
No, they did not. The fact that they thought they innovated is just a testament to their ignorance.
David Reed doesn't know what he is talking about. Interference is very real and well defined, independent of the particular receiver, once you have decided on frequency and bandwidth allocations for different users.
Historically, because of technological limitations, receivers have gotten more bandwidth than they needed for their tasks, which allowed them to be more robust to interference, but the interference was still there.
Somehow, I had hoped that the Presidio would be home to something with a bit more class and style than Lucas and his company. It's just such a beautiful and unique spot.
If anything is going to kill Linux and the open-source movement, it's the presence of certifiable lunatics in the ranks representing the users.
Here is a clue: there are certifiable lunatics present in almost any random group of people that has more than maybe a hundred people in it. Human beings deal as best as they can with that reality through things like the police and the courts. If someone made death threats against O'Gara, it's a matter for the police and has nothing to do with Linux.
Unless the community gets a handle on this, grows up, and rebukes the extremists, the trash heap of history is where this is all headed.
Well, I guess it's the trash heap of history then, because no group of people has ever been able to get a handle on this. Why, we can't even get computer columnists making inflammatory remarks on matters they know little about to shut up, and those people are not even necessarily certifiable lunatics.
This doesn't seem to generate much of a performance hit, so I wonder why Microsoft is going for a different approach? Apple's seems to make more sense.
Both designs have been around for quite some time. Updating the index immediately when a file changes may be OK for desktop use, but it has unpredictable consequences for other kinds of uses. The conservative and safe thing is to update on a schedule unless the user explicitly requests otherwise.
Apple's implementation is integrated into the filesystem layer, so it indexes everything as it is written to disk
"Integrating" any of this in the file system layer is unnecessary. All you need is update notifications from the kernel to a user mode index server, a tiny addition to most kernel, and one that Linux has supported for a while now.
Microsoft has a Tabbed Browser called ".NET SDK Documentation". Rather than being "basic", you can drag-n-drop the tabs and Z-Order works. In other words, they're more advanced than Firefox tabs and hopefully that's what they'll use for IE7.
Sorry, that's not particularly original to Microsoft. There are several extensions for Firefox that offer such features, as well as several other browsers based on the Mozilla renderer (Galeon tabs work really nicely out of the box). The concepts themselves are older than that.
Check out Mozilla Extensions
Anyone can produce a JVM. There are several good FOSS ones out there now, as well as AOT compilers with GCJ nearly fully complete. There are NO legal impediments to ANYONE producing their own JVM and there is NO Sun-only bits in OOo.
Anyone could implement GIF as well; people found out that there were intellectual property problems with GIF only years later. Anyone could implement FAT or SMB, but Microsoft is now claiming that they own intellectual property in those standards. Anyone could implement a UNIX-like operating system, but SCO now claims that they own it all and has started a billion dollar lawsuit over it. The open source community needs to avoid such problems because they are disruptive and expensive.
In the case of Java, just read the damned licenses on Sun's site. You will see that Sun explicitly states that many Java specifications belong to Sun. And go check on Sun's patents: there are a dozen different patents on important Java technologies.
Quit spreading FUD.
Quit trying to trick people into using proprietary standards by pretending that they are free.
Sun is a failing computer maker with an army of lawyers, they have a widely used proprietary platform, their licenses clearly state that they own the platform and the APIs, and they are "contributing" functionality that depends on their proprietary platform to several open source projects. Add two and two together.