MAME uses software to simulate arcade hardware using general purpose computing hardware.
FPGA's are a different beast altogether. An FPGA is similar to a ROM in that patterns can be burned into it but they are far more versatile. FPGA's can have logic burned into them which makes them malleble hardware. An FPGA with a 6502 processor and supporting logic burned into it is NOT emulating say Pac-Man. It is a re-implementation of Pac-Man.
This guy is implementing an Atari 2600 on an FPGA:
http://www.mindspring.com/~2600onachip/
If he pulls it off, one could do neat things like make a super battery efficient handheld. The FPGA is for all intents and purposes a 2600 so there isn't the overhead a StrongARM chip running at 200Mhz would have.
The problem is that the mere act of viewing a webpage involves "downloading copyrighted materials". The browser cache even retains copies of some of it for awhile.
Such a policy taken literally means that the organization in question will have to pull their net connection entirely.
A better policy would forbidding downloading of copyrighted material without permission. Normal websites give implicit permission to do the downloading necessary to view them. This also has the advantage of forbidding illegal mp3 downloads while still permitted downloads of legal mp3s and software.
I'll also point out that mentioning communism has the same intelligent conversation ending abilities.
dmaxwell's Parallel to Godwin's law goes something like this:
In an online discussion involving software development or licencing, the probability that someone will make a comparison to communists, Soviet Russia, or communism approaches one.
It wouldn't change a thing. It would mean that the super-rich can be above the law as long as they do high profile good works. Also, it is not a given that a scofflaw super-rich will engage in charity. If we let Gate's company get away with crimes on the basis that the side effects are good then it will just give a green light to whoever occupies Gate's niche in the future.
That niche WILL be occupied too. We didn't learn a thing from Ma Bell, Standard Oil OR Microsoft. History will likely repeat itself sometime down the road with bio or nano tech. Their philantropy shouldn't be a saving excuse either.
A more sane way to manage source packages on production boxes is to have a machine similar to the production boxes but with the developer toolchain installed.
The production boxes will still use debs or rpms but the compilation boxes can easily use something like checkinstall to make packages. This won't work in a potpurri environment but it would be fine if there's lots of identical machines. You mentioned that you wanted only particular software on your machines. With source compilation, you can even specify that the software only have certain options compiled in.
Since the dev toolchains are confined to a few boxes, maintaining those shouldn't be onerous either.
That they perform charities has absolutely nothing to with their illegal practices. It most certainly is not an excuse or a reason for leiniency.
"But if we let them get away with it then there'll be even MORE money for charity....." Nope. Don't think so."
Good works in non-related areas don't somehow make up for evil works in others. They should live or die in the courts according to the merits of the case.
Gates would still have billions even if the worst possible penalties were leveled against MS. If the charity is genuine then he can continue to give...regardless of whether or not it helps MS in the court of public opinion.
GPLing the whole thing is only one possible remedy. To a judge, a GPL violation is only going to appear to be a copyright violation. Any number of things could happen like an injunction not to distribute the work at all. Basically, the violator has to make things right with the plaintiff. If he can't then the judge will impose a remedy of some sort. That remedy will largely be at the judge's discretion and need not be GPLing the source. A more likely outcome would be negigible compensatory damages since GPL code can be distributed for free but the PUNITIVE damages could be considerable. They would probably have to rip out the GPLed code and replace it with their own stuff. They would probably have to not support the tainted code and require non tainted upgrades for their customers.
The outcome could be very ugly for a GPL violator but it is not mandated that GPLing a violating work is the remedy. Sometimes that remedy is arrived at after arbitration of some sort but isn't necessarily the only remedy. Another remedy may be an alternative license from the copyright holder(s).
I doubt MS would care one way or another about a high profile enforcement of the GPL. Believe it or not, MS' Services for Unix contains some GPLed utilities and they offer the source for them the way they're supposed to. Don't get me wrong, I think MS is run by a lot of scum sucking weasels but they aren't STUPID weasels.
Microsoft has an army of developers and pool of cash that let's them buy pretty much any commercial codebase they want. They have nothing to gain by ripping off GPLed code and quite a lot to lose.
MS whinges about GPLed code not because they can't steal it but because their Embrace Extend Extinguish and FUD tactics are largely useless when it is a competitor.
I was running Woody on my desktops when it was testing and when testing became Sarge it really didn't matter as far as my machines were concerned. Like many people who use Debian on desktops, my machines are always somewhere between Sid and Testing with the odd non-official package here and there. For the most part it is the scenario you have in mind. I think you're right in that it would be nice for a paid support model as well.
I imagine its a similar experience on ports based BSD systems and Gentoo.
I can think of some things that would make a lot of the people here bitch though. Such a distribution would have to hang back 6 months or so from OSS/Free's bleeding edge. If say, an engine for vector graphics on the desktop comes out for XFree86, the distro won't be able to include it until it's solid. Contrast that with the people here who will spend 3 hours compiling tarballs so they'll be the first kid on the block to have it. Those same 'leet kiddies will whinge "Incremental distro will never succeed unless it's more current!" Solidity and up-to-last-week currentness are mutually exclusive.
There's also the question of how to handle major infrastructure transitions. I'm thinking of things like from XFree 3.x to 4.x, libc5 to libc6, KDE2.x to 3.x, kernel 2.4.x to 2.6.x, and last but not least GCC 2.9x to 3.x. Not to mention major changes in server daemons like Apache and Samba. The major libc and GCC increments are thankfully infrequent but they're also the worst. They both have severe consequences for backwards compatibility with older binaries and source trees. My point is that such transitions will force "Incremental Distro" to draw hard lines from time to time on what they'll support and what they won't. Shoot! Some people are still running heavily patched 2.0 kernels.
This brings up the other group of people Incremental Distro can't always make happy : The Ultraconservative Sysadmin. Sooner or later, support for say Apache 1.x will only be handled by boutique consultants. Most everyone but the Ultraconservative Sysadmin will have moved on. I think what will happen is that the distro will have to define brackets in time that start with those major transitions. During the bracket period (two years say) they'll have to maintain a branch of pre-transition compatible packages. The other thing they could do is be cold blooded about Ultraconservatives and just bump everybody up when these changes happen. Ultraconservative Admin is probably clued enough to manage his own upgrade schedule from patched source.
The REAL problem is that OSS/Free is developed and maintained on Internet Time. I suppose another outcome would be a spectrum of (differently organized) incremental distros with more and less aggressive attitudes toward upgrading.
Most of the BSA's power comes from EULAs which basically legalize compulsory amputation for copyright violation. In the case of an organization running all OSS software, no EULA's have been signed. No contracts have been signed and no agreements with BSA members have been made.
Such a company has vastly more legal options than a typical company that is Microsoft/Adobe all the way. Let's see:
1. Libel/Slander law. After all they have to have probable cause from someone to make that raid.
2. Laws regarding harassment. Their standard techniques are certainly harrassment if you are contractually obligated to permit it.
3. Their audits basically shut you down while they take your machines apart. There's a basis for a monstrous civil suit.
4. A really good lawyer could probably think up some criminal charges to go along with the civil charges you're going to nail them with. Get the local DA involved as well. Some people have mentioned RICO. A case could be made for it by an all OSS company.
With some perserverence their typical scenario could be turned back around on them. Settle with US and we won't break you over a wheel. The non-legal threat of extremely bad publicity for the member company they're acting on behalf of should generate some nice pressure as well. Actually, all of the above also applies to the members who think they're going to get a win for.
Some sound cards have multiple dsps. I believe the Soundblaster Live! has one super dsp that can be multiplexed into a number of devices. Perhaps a Live user can elaborate. I use an ES1373 based card in my home rig that has a full record and playback dsp0 and a playback only dsp1. I usually have artsd running on dsp1 (I don't use any apps that record through arts) and I leave dsp0 open for whatever. The ES1373 helped with my hassles in that regard quite a bit.
Work is underway to give OpenOffice first a Quartz interface then a full Aqua interface. The current OpenOffice for the Mac depends on X11 and is clearly labeled as a "Developer's Pre-release".
OpenOffice on OS X only exists in it's current form so that the backend code (common to all ports - filters and so forth) can be debugged insofar as the non-GUI parts don't like Darwin. Once the core is solid and clean on Darwin then it will get an interface that is more pleasing. If they had to make a native interface for it before doing anything else, it would take much longer for a solid OS X port. The roadmap is here.
You're larger point may be valid but the OSX port of OpenOffice (as it currently stands) is not a valid example.
Cable companies send their techs driving around in vans with sniffer gear. Don't put on your tin foil hat just yet. It isn't as obnoxious as it sounds. Each cable connection in your house leaks a predictable amount of RF. Your cable modem would be one such connection. As you connect more things to your cable, that slight leakage increases. What the tech in the van sees is little more than a RF signal strength indication. What they do would work just fine with nothing more than an analog gauge and a little experience.
They don't just look for people outright stealing cable either. They're also interested in people using more hookups than what they paid for. More hookups - more deflection of that RF gauge.
Since all they are doing is monitoring the amount of signal leaking out their own network, they are legally in the clear on this. The fact that they didn't bother with a trap doesn't mean you're entitled to the video portion of the signal. If you hook up a TV, they'll find it if they're inclined to look for it.
Oh, they often use Time Domain Reflectometers to find faults in their cable. Those you who ride the network cabling range know about these too. A TDR will give highly precise readings on just what is sitting on a line. Those will work even if some smartass turns his house into a faraday cage.
The clickety click apps are all twiddling registry keys. Even a console-only variant of NT would need a registry. ini files are a holdover from the 2.x-3.x days. Even services (read daemons) on NT family operating systems are configured from the registry. It would only be necessary to have alternative utilities to do the registry twiddling (well there would be the small matter of intercepting GUI calls.....). For that matter, tools like REG.EXE already exist but aren't commonly employed. Something like curses could be used to make console-style wizards that use things like REG.EXE to install services.
I don't think administration from a console is insurmountable. The registry is just a database and those can be manipulated just fine without a GUI. The fact that most Windows programs expect a GUI to be running is more problimatical. Those calls have to be intercepted and presented in a way appropriate for a console. The complexity of that is comparable to the various WINE forks. It may be simpler to just provide a GUI.
Public key crypto is a little harder to laugh off. If the DRM is ignored in a correctly designed scheme then the video won't play at all. It will just be white noise. They aren't going to rely on simple flags anymore to enfore rules.
Once DRM is pushed to it's logical conclusion, it will require hardware hacks to defeat rather than software. I think DeCSS history will repeat itself. Some Taiwanese vendor will be in a hurry for a profit and will slip up with an easily tapped system board. Even without that help, the most evil Fritzed crypto scheme will be cracked sooner or later but don't kid yourself that DRM isn't possible with Open Source.
Windows we may well have to live with. All forms of DRM will be cracked whether Hollywood likes it or not. For that matter, they will be cracked whether Fritz likes it or not.
Doing whatever the hell you want in your own house is the American Way too. Whether you like it or not.
Cluster node:no Media Player:yes
on
Xbox Linux Cluster
·
· Score: 2, Insightful
I think most of us could have told him those conclusions before he ran his tests. An X-Box is a PITA for a node or as a PC. That doesn't mean there is no practical use for Linux on one. With MPLayer and a few other pieces of software, an X-Box would make a nice media device for the living room. It wouldn't look completely out of place in a stereo rack and just basically has a better form factor than minitower PC. It's already equipped with TV out, an IR reciever and proper audio jacks.
As a bonus, throw MAME and few other games on it. True using an X-Box as Linux gaming machine won't satisfy hardcore gamers. It's just fine for simple games like Pac-Man and Madbomber. Kinda like an Atari 2600 with good graphics. That isn't always a bad thing.
Re:Tabbed browsing?
on
KDE 3.1 Released
·
· Score: 2, Informative
Tabbed browsing is a UI feature. Safari is only using the KHTML rendering engine. The rendering engine shouldn't care whether it is drawing into a tab or window. This means that features added to Konqueror don't necessary become part of Safari or vice versa. Both browers have lots of code apart from the rendering engine. Improvements to the rendering engine will come to both sooner or later. Their interfaces are independent though.
If you actually read the GPL you'll find that it has nothing to say about using the software. Not that RMS would like it but using GPL firmware to control a baby threshing machine doesn't break the license as Theo so eloquently put it. Sure it breaks other laws but the point is the GPL has nothing to say about any use or private modifications made to the software. It even goes out of it's way to point out that agreement with the license is unecessary to use the software.
The default in copyright law is that no one other than the copyright holder can distribute a work. "Fair Use" does permit limited excerpting for commentary, criticism, or parody. The only thing that lets any copyrighted work be distributed is a license. All a would be corporate plagiarist would accomplish in legally attacking the GPL is to get the license ruled null and void in his case. The violator would not suddenly gain full ownership of the work and the right to do anything with it he wants...mu! ha! ha! ha! With the license invalidated, the software doesn't become fully public domain. It reverts back to the copyright law default...which says? Right! The copyright holder has the sole right distribute or license distribution.
How is the copyright holder going to feel after being manhandled by corporate shysters? Pretty pissed I'd imagine. The best possible outcome for a plagiarist in a GPL dispute would be to somehow prove he did not violate the license...this won't accomplish what we all assume GPL violators want to do: close source a widely distributed GPL derived work. Note, the originator of a GPLed work who owns the copyright to all parts of it can relicense subsequent distributions any way he wants.
Invalidating the license would be a pyrrhic victory. He would then have to satisfy the copyright holder and would only have two choices to do so. He can remove the GPLed bits entirely and reimplement them. The other alternative would be an alternative license from the copyright holder who's going to be thoroughly torqued by this point and inclined to insist on the most expensive terms possible.
An attack on the GPL is also an attack on the legal structure that currently allows draconian copyrights. It is not coincidental that we don't see high profile legal attacks on the license itself.
On Debian machines at least, it will only accept connections from localhost until the admin opens it up. I would assume something like this is done on OpenBSD as well. What are you installing MySQL on that it's listening publically by default when installed?
I know you're not serious, but exactly what part of the system being circumvented is digital?
Well, a pin can be either up or down. That's probably digital enough for a shyster to work with. What's more the set of all possible heights for a pin is probably not analog either. A lock system can be perfectly modeled by any type of digital system which pretty much means the locks under discusssion are a special case of a digital system. They just happen to be implemented mechanically rather than electronically.
All it would take is a sharp lawyer to walk a receptive judge through it and there's a precendent for applying the DMCA to non-computer security issues.
Such guerillas also favor weapons like the Kalashnikov. Sure it might be a little less accurate and a lot less whizbangy but that is what makes it so favorable. A Kalashnikov will still fire even if it's been thrown down in the mud. The parent poster is correct. It is much easier to disrupt sophisticated weaponry than it is to design it.
That doesn't always hold though. The Afghan rebels didn't have any trouble finding the best way to use Stinger missiles. Still that's pretty much a fire and forget weapon that doesn't have to be maintained once fired. Guerillas won't go in for anything that requires lots of care and feeding. They will gleefully monkeywrench anything that does though.
A microwave oven is tuned to the resonant frequency of water molecules. This government weapon will be tuned to the frequency of silicon atoms. It'll burn people up close I suppose but be mostly harmless to flesh further out while still frying transistors.
There is a one megaton threshold of "usefulness" for nuclear weapons. In a surface detonation, the blast from a thermonuclear is powerful enough to send some of the atmosphere above ground zero into space. Once 1MT is crossed it seems that most of the excess force powers this effect leaving less energy for the blast wave and radiation to destroy things OUTWARD of the blast. It isn't as counterintuitive as it seems. Most of the energy output of a bomb is X-rays and gamma radiation. A bomb going off in space is little more than a bright flash and a little puff of gas from vaporized bomb components (It still sucks to be anywhere in the vicinity.).
The devasting effects we associate with nukes comes from the effects this radiation flux has on the atmosphere. It's like a vastly oversized thunderclap. The radiation instantly heats up a large amount of atmosphere and this is what creates the blast wave and starts a lot of the fires. Of course, there's lots of radiation left over to flash fry things further out. Heat a quantity of atmosphere up enough and it's going straight up in a hurry.
That isn't to say that there are NO noticable effects of making the bomb bigger but from a military point of view the law of diminishing returns kicks in with a vengence. There is another threshold around a gigaton or so that makes a bomb a planetary threat with some different effects involved (similar to a large asteroid collision) but who wants to set a Backyard Bomb off? It's called a Backyard Bomb because it doesn't NEED a delivery system. You set it off in your backyard and it fries your enemies anyway.
The 50MT Soviet bomb was the biggest public relations stunt in history. Khruschev literally told Sakarov to make something to "scare the ^$#@ out of the Americans" in time for a conference. It also came from a touch of Texas in the Russian mentality. The worlds biggest church bell sits on the ground somewhere in Russia because no one wanted to build the matching bell tower. It is Tsar Bell (the King of Bells). It is an impressive gesture that is practically useless. Tsar Bomba is same thing: a militarily useless ridiculously oversized weapon intended only as a gesture.
Slashdot Mirror
MAME uses software to simulate arcade hardware using general purpose computing hardware.
FPGA's are a different beast altogether. An FPGA is similar to a ROM in that patterns can be burned into it but they are far more versatile. FPGA's can have logic burned into them which makes them malleble hardware. An FPGA with a 6502 processor and supporting logic burned into it is NOT emulating say Pac-Man. It is a re-implementation of Pac-Man.
This guy is implementing an Atari 2600 on an FPGA:
http://www.mindspring.com/~2600onachip/
If he pulls it off, one could do neat things like make a super battery efficient handheld. The FPGA is for all intents and purposes a 2600 so there isn't the overhead a StrongARM chip running at 200Mhz would have.
The problem is that the mere act of viewing a webpage involves "downloading copyrighted materials". The browser cache even retains copies of some of it for awhile.
Such a policy taken literally means that the organization in question will have to pull their net connection entirely.
A better policy would forbidding downloading of copyrighted material without permission. Normal websites give implicit permission to do the downloading necessary to view them. This also has the advantage of forbidding illegal mp3 downloads while still permitted downloads of legal mp3s and software.
I'll also point out that mentioning communism has the same intelligent conversation ending abilities.
dmaxwell's Parallel to Godwin's law goes something like this:
In an online discussion involving software development or licencing, the probability that someone will make a comparison to communists, Soviet Russia, or communism approaches one.
It wouldn't change a thing. It would mean that the super-rich can be above the law as long as they do high profile good works. Also, it is not a given that a scofflaw super-rich will engage in charity. If we let Gate's company get away with crimes on the basis that the side effects are good then it will just give a green light to whoever occupies Gate's niche in the future.
That niche WILL be occupied too. We didn't learn a thing from Ma Bell, Standard Oil OR Microsoft. History will likely repeat itself sometime down the road with bio or nano tech. Their philantropy shouldn't be a saving excuse either.
A more sane way to manage source packages on production boxes is to have a machine similar to the production boxes but with the developer toolchain installed.
The production boxes will still use debs or rpms but the compilation boxes can easily use something like checkinstall to make packages. This won't work in a potpurri environment but it would be fine if there's lots of identical machines. You mentioned that you wanted only particular software on your machines. With source compilation, you can even specify that the software only have certain options compiled in.
Since the dev toolchains are confined to a few boxes, maintaining those shouldn't be onerous either.
That they perform charities has absolutely nothing to with their illegal practices. It most certainly is not an excuse or a reason for leiniency.
"But if we let them get away with it then there'll be even MORE money for charity....." Nope. Don't think so."
Good works in non-related areas don't somehow make up for evil works in others. They should live or die in the courts according to the merits of the case.
Gates would still have billions even if the worst possible penalties were leveled against MS. If the charity is genuine then he can continue to give...regardless of whether or not it helps MS in the court of public opinion.
GPLing the whole thing is only one possible remedy. To a judge, a GPL violation is only going to appear to be a copyright violation. Any number of things could happen like an injunction not to distribute the work at all. Basically, the violator has to make things right with the plaintiff. If he can't then the judge will impose a remedy of some sort. That remedy will largely be at the judge's discretion and need not be GPLing the source. A more likely outcome would be negigible compensatory damages since GPL code can be distributed for free but the PUNITIVE damages could be considerable. They would probably have to rip out the GPLed code and replace it with their own stuff. They would probably have to not support the tainted code and require non tainted upgrades for their customers.
The outcome could be very ugly for a GPL violator but it is not mandated that GPLing a violating work is the remedy. Sometimes that remedy is arrived at after arbitration of some sort but isn't necessarily the only remedy. Another remedy may be an alternative license from the copyright holder(s).
I doubt MS would care one way or another about a high profile enforcement of the GPL. Believe it or not, MS' Services for Unix contains some GPLed utilities and they offer the source for them the way they're supposed to. Don't get me wrong, I think MS is run by a lot of scum sucking weasels but they aren't STUPID weasels.
Microsoft has an army of developers and pool of cash that let's them buy pretty much any commercial codebase they want. They have nothing to gain by ripping off GPLed code and quite a lot to lose.
MS whinges about GPLed code not because they can't steal it but because their Embrace Extend Extinguish and FUD tactics are largely useless when it is a competitor.
I was running Woody on my desktops when it was testing and when testing became Sarge it really didn't matter as far as my machines were concerned. Like many people who use Debian on desktops, my machines are always somewhere between Sid and Testing with the odd non-official package here and there. For the most part it is the scenario you have in mind. I think you're right in that it would be nice for a paid support model as well.
I imagine its a similar experience on ports based BSD systems and Gentoo.
I can think of some things that would make a lot of the people here bitch though. Such a distribution would have to hang back 6 months or so from OSS/Free's bleeding edge. If say, an engine for vector graphics on the desktop comes out for XFree86, the distro won't be able to include it until it's solid. Contrast that with the people here who will spend 3 hours compiling tarballs so they'll be the first kid on the block to have it. Those same 'leet kiddies will whinge "Incremental distro will never succeed unless it's more current!" Solidity and up-to-last-week currentness are mutually exclusive.
There's also the question of how to handle major infrastructure transitions. I'm thinking of things like from XFree 3.x to 4.x, libc5 to libc6, KDE2.x to 3.x, kernel 2.4.x to 2.6.x, and last but not least GCC 2.9x to 3.x. Not to mention major changes in server daemons like Apache and Samba. The major libc and GCC increments are thankfully infrequent but they're also the worst. They both have severe consequences for backwards compatibility with older binaries and source trees. My point is that such transitions will force "Incremental Distro" to draw hard lines from time to time on what they'll support and what they won't. Shoot! Some people are still running heavily patched 2.0 kernels.
This brings up the other group of people Incremental Distro can't always make happy : The Ultraconservative Sysadmin. Sooner or later, support for say Apache 1.x will only be handled by boutique consultants. Most everyone but the Ultraconservative Sysadmin will have moved on. I think what will happen is that the distro will have to define brackets in time that start with those major transitions. During the bracket period (two years say) they'll have to maintain a branch of pre-transition compatible packages. The other thing they could do is be cold blooded about Ultraconservatives and just bump everybody up when these changes happen. Ultraconservative Admin is probably clued enough to manage his own upgrade schedule from patched source.
The REAL problem is that OSS/Free is developed and maintained on Internet Time. I suppose another outcome would be a spectrum of (differently organized) incremental distros with more and less aggressive attitudes toward upgrading.
Most of the BSA's power comes from EULAs which basically legalize compulsory amputation for copyright violation. In the case of an organization running all OSS software, no EULA's have been signed. No contracts have been signed and no agreements with BSA members have been made.
Such a company has vastly more legal options than a typical company that is Microsoft/Adobe all the way. Let's see:
1. Libel/Slander law. After all they have to have probable cause from someone to make that raid.
2. Laws regarding harassment. Their standard techniques are certainly harrassment if you are contractually obligated to permit it.
3. Their audits basically shut you down while they take your machines apart. There's a basis for a monstrous civil suit.
4. A really good lawyer could probably think up some criminal charges to go along with the civil charges you're going to nail them with. Get the local DA involved as well. Some people have mentioned RICO. A case could be made for it by an all OSS company.
With some perserverence their typical scenario could be turned back around on them. Settle with US and we won't break you over a wheel. The non-legal threat of extremely bad publicity for the member company they're acting on behalf of should generate some nice pressure as well. Actually, all of the above also applies to the members who think they're going to get a win for.
I say go for it.
Some sound cards have multiple dsps. I believe the Soundblaster Live! has one super dsp that can be multiplexed into a number of devices. Perhaps a Live user can elaborate. I use an ES1373 based card in my home rig that has a full record and playback dsp0 and a playback only dsp1. I usually have artsd running on dsp1 (I don't use any apps that record through arts) and I leave dsp0 open for whatever. The ES1373 helped with my hassles in that regard quite a bit.
Work is underway to give OpenOffice first a Quartz interface then a full Aqua interface. The current OpenOffice for the Mac depends on X11 and is clearly labeled as a "Developer's Pre-release".
OpenOffice on OS X only exists in it's current form so that the backend code (common to all ports - filters and so forth) can be debugged insofar as the non-GUI parts don't like Darwin. Once the core is solid and clean on Darwin then it will get an interface that is more pleasing. If they had to make a native interface for it before doing anything else, it would take much longer for a solid OS X port. The roadmap is here.
You're larger point may be valid but the OSX port of OpenOffice (as it currently stands) is not a valid example.
Cable companies send their techs driving around in vans with sniffer gear. Don't put on your tin foil hat just yet. It isn't as obnoxious as it sounds. Each cable connection in your house leaks a predictable amount of RF. Your cable modem would be one such connection. As you connect more things to your cable, that slight leakage increases. What the tech in the van sees is little more than a RF signal strength indication. What they do would work just fine with nothing more than an analog gauge and a little experience.
They don't just look for people outright stealing cable either. They're also interested in people using more hookups than what they paid for. More hookups - more deflection of that RF gauge.
Since all they are doing is monitoring the amount of signal leaking out their own network, they are legally in the clear on this. The fact that they didn't bother with a trap doesn't mean you're entitled to the video portion of the signal. If you hook up a TV, they'll find it if they're inclined to look for it.
Oh, they often use Time Domain Reflectometers to find faults in their cable. Those you who ride the network cabling range know about these too. A TDR will give highly precise readings on just what is sitting on a line. Those will work even if some smartass turns his house into a faraday cage.
The clickety click apps are all twiddling registry keys. Even a console-only variant of NT would need a registry. ini files are a holdover from the 2.x-3.x days. Even services (read daemons) on NT family operating systems are configured from the registry. It would only be necessary to have alternative utilities to do the registry twiddling (well there would be the small matter of intercepting GUI calls.....). For that matter, tools like REG.EXE already exist but aren't commonly employed. Something like curses could be used to make console-style wizards that use things like REG.EXE to install services.
I don't think administration from a console is insurmountable. The registry is just a database and those can be manipulated just fine without a GUI. The fact that most Windows programs expect a GUI to be running is more problimatical. Those calls have to be intercepted and presented in a way appropriate for a console. The complexity of that is comparable to the various WINE forks. It may be simpler to just provide a GUI.
Public key crypto is a little harder to laugh off. If the DRM is ignored in a correctly designed scheme then the video won't play at all. It will just be white noise. They aren't going to rely on simple flags anymore to enfore rules.
Once DRM is pushed to it's logical conclusion, it will require hardware hacks to defeat rather than software. I think DeCSS history will repeat itself. Some Taiwanese vendor will be in a hurry for a profit and will slip up with an easily tapped system board. Even without that help, the most evil Fritzed crypto scheme will be cracked sooner or later but don't kid yourself that DRM isn't possible with Open Source.
Windows we may well have to live with. All forms of DRM will be cracked whether Hollywood likes it or not. For that matter, they will be cracked whether Fritz likes it or not.
Doing whatever the hell you want in your own house is the American Way too. Whether you like it or not.
I think most of us could have told him those conclusions before he ran his tests. An X-Box is a PITA for a node or as a PC. That doesn't mean there is no practical use for Linux on one. With MPLayer and a few other pieces of software, an X-Box would make a nice media device for the living room. It wouldn't look completely out of place in a stereo rack and just basically has a better form factor than minitower PC. It's already equipped with TV out, an IR reciever and proper audio jacks.
As a bonus, throw MAME and few other games on it. True using an X-Box as Linux gaming machine won't satisfy hardcore gamers. It's just fine for simple games like Pac-Man and Madbomber. Kinda like an Atari 2600 with good graphics. That isn't always a bad thing.
Tabbed browsing is a UI feature. Safari is only using the KHTML rendering engine. The rendering engine shouldn't care whether it is drawing into a tab or window. This means that features added to Konqueror don't necessary become part of Safari or vice versa. Both browers have lots of code apart from the rendering engine. Improvements to the rendering engine will come to both sooner or later. Their interfaces are independent though.
If you actually read the GPL you'll find that it has nothing to say about using the software. Not that RMS would like it but using GPL firmware to control a baby threshing machine doesn't break the license as Theo so eloquently put it. Sure it breaks other laws but the point is the GPL has nothing to say about any use or private modifications made to the software. It even goes out of it's way to point out that agreement with the license is unecessary to use the software.
The default in copyright law is that no one other than the copyright holder can distribute a work. "Fair Use" does permit limited excerpting for commentary, criticism, or parody. The only thing that lets any copyrighted work be distributed is a license. All a would be corporate plagiarist would accomplish in legally attacking the GPL is to get the license ruled null and void in his case. The violator would not suddenly gain full ownership of the work and the right to do anything with it he wants...mu! ha! ha! ha! With the license invalidated, the software doesn't become fully public domain. It reverts back to the copyright law default...which says? Right! The copyright holder has the sole right distribute or license distribution.
How is the copyright holder going to feel after being manhandled by corporate shysters? Pretty pissed I'd imagine. The best possible outcome for a plagiarist in a GPL dispute would be to somehow prove he did not violate the license...this won't accomplish what we all assume GPL violators want to do: close source a widely distributed GPL derived work. Note, the originator of a GPLed work who owns the copyright to all parts of it can relicense subsequent distributions any way he wants.
Invalidating the license would be a pyrrhic victory. He would then have to satisfy the copyright holder and would only have two choices to do so. He can remove the GPLed bits entirely and reimplement them. The other alternative would be an alternative license from the copyright holder who's going to be thoroughly torqued by this point and inclined to insist on the most expensive terms possible.
An attack on the GPL is also an attack on the legal structure that currently allows draconian copyrights. It is not coincidental that we don't see high profile legal attacks on the license itself.
On Debian machines at least, it will only accept connections from localhost until the admin opens it up. I would assume something like this is done on OpenBSD as well. What are you installing MySQL on that it's listening publically by default when installed?
Forget Q*bert, whatever happened to dig dug??
It's quite sad really. You can read all about it here.
I know you're not serious, but exactly what part of the system being circumvented is digital?
Well, a pin can be either up or down. That's probably digital enough for a shyster to work with. What's more the set of all possible heights for a pin is probably not analog either. A lock system can be perfectly modeled by any type of digital system which pretty much means the locks under discusssion are a special case of a digital system. They just happen to be implemented mechanically rather than electronically.
All it would take is a sharp lawyer to walk a receptive judge through it and there's a precendent for applying the DMCA to non-computer security issues.
Such guerillas also favor weapons like the Kalashnikov. Sure it might be a little less accurate and a lot less whizbangy but that is what makes it so favorable. A Kalashnikov will still fire even if it's been thrown down in the mud. The parent poster is correct. It is much easier to disrupt sophisticated weaponry than it is to design it.
That doesn't always hold though. The Afghan rebels didn't have any trouble finding the best way to use Stinger missiles. Still that's pretty much a fire and forget weapon that doesn't have to be maintained once fired. Guerillas won't go in for anything that requires lots of care and feeding. They will gleefully monkeywrench anything that does though.
A microwave oven is tuned to the resonant frequency of water molecules. This government weapon will be tuned to the frequency of silicon atoms. It'll burn people up close I suppose but be mostly harmless to flesh further out while still frying transistors.
There is a one megaton threshold of "usefulness" for nuclear weapons. In a surface detonation, the blast from a thermonuclear is powerful enough to send some of the atmosphere above ground zero into space. Once 1MT is crossed it seems that most of the excess force powers this effect leaving less energy for the blast wave and radiation to destroy things OUTWARD of the blast. It isn't as counterintuitive as it seems. Most of the energy output of a bomb is X-rays and gamma radiation. A bomb going off in space is little more than a bright flash and a little puff of gas from vaporized bomb components (It still sucks to be anywhere in the vicinity.).
The devasting effects we associate with nukes comes from the effects this radiation flux has on the atmosphere. It's like a vastly oversized thunderclap. The radiation instantly heats up a large amount of atmosphere and this is what creates the blast wave and starts a lot of the fires. Of course, there's lots of radiation left over to flash fry things further out. Heat a quantity of atmosphere up enough and it's going straight up in a hurry.
That isn't to say that there are NO noticable effects of making the bomb bigger but from a military point of view the law of diminishing returns kicks in with a vengence. There is another threshold around a gigaton or so that makes a bomb a planetary threat with some different effects involved (similar to a large asteroid collision) but who wants to set a Backyard Bomb off? It's called a Backyard Bomb because it doesn't NEED a delivery system. You set it off in your backyard and it fries your enemies anyway.
The 50MT Soviet bomb was the biggest public relations stunt in history. Khruschev literally told Sakarov to make something to "scare the ^$#@ out of the Americans" in time for a conference. It also came from a touch of Texas in the Russian mentality. The worlds biggest church bell sits on the ground somewhere in Russia because no one wanted to build the matching bell tower. It is Tsar Bell (the King of Bells). It is an impressive gesture that is practically useless. Tsar Bomba is same thing: a militarily useless ridiculously oversized weapon intended only as a gesture.