Because the USPS issues money orders. They are subject to financial regulations and are required to refuse or report the purchase of money orders over a specific amount (in the name of another war, the war on drugs). It to prevent drug dealers from laundering money since money order can be used as cash.
Change the names of the constants in the headers and perform a mass search-and-replace of the constant names across the source tree. Won't take more than a few hours.
Unless SCO is claiming ownership of the meanings of some NUMBERS, then that effectively gets around the problem. Besides, haven't these constants been around forever and been used and reused throughout UNIX history?
However, the works in question contain the copyrights of dozens or hundreds of indviduals, licensed jointly. So you are receiving a financial because when you agree to the license, you gain access to all of the copyrighted works as well as the right to modify and distribute them on your own. And the license also gives you the ability to make money off that copyrighted work, if you so chose and can convince someone to pay you for it. So there is definitely a "financial gain" angle there and that option to make money on GPL'd works is explicitly stated in the GPL itself.
The problem with kernel modules is as Linus has stated: they fall into a grey area of the GPL.
Under a strict GPL interpretation, modules can be considered "derived works" because they link dynamically into a GPL environment (the kernel). The modules depends on kernel headers to compile and use internal kernel data structures exposed through module interfaces that are technically under the GPL. Thus, if you interpret the GPL as written, then yes, all modules would be required to be released under the GPL.
Clearly such a situation is not acceptable to hardware manufacturers and Linus recognized this. Hence his "binary only is allowed for modules" rule. It is the kernel developers themselves that are changing the module environment to try and enforce the strict constructionist view of the GPL.
RMS recognized this issue and developled the LGPL. The intent behind the LGPL was to allow for the development of GPL'd code that could be incorporated into binary-only products without tainting those products provided the binary product announced was linking to LGPL code and made it available on request (or distributed it with the product). As long as a clear separation exists between the proprietary product and the LGPL'd code (in other words, the proprietary code links to the LGPL'd work. If it copied LGPL'd code into its code base, then it would become tainted), the two models can co-exist.
RMS doesn't like the LGPL. It is a compromise license designed to fit the realities of the world rather than what he wishes it could be.
In my opinion, maybe the simplest solution is to relicense future versions of the Linux kernel under the LGPL. The distribution and modification requirements for the kernel proper would still be in force as with the GPL and make it compatible with binary-only modules explicitly since this very issue of linkage and mixing is exactly what the LGPL was designed to address. For the record, I prefer the LGPL over the GPL for practical, real-world open source development. It provides the best of both worlds when approached correctly.
If the LGPL cannot fit, then clarification is required for the definition of derived works as encompassed by copyright law and how the GPL perceives it. v3 of the license should maybe incorporate a better, clearer definition of dynamic linking when no source is exchanged between two programs.
I agree with Linus. Binary-only modules should be acceptable under the GPL since they do not incorporate GPL'd works directly except through runtime linkage.
My company originally outsourced a lot of software work overseas. In the past 18 months, they've learned a harsh lesson that goes with outsourcing without doing proper research. That lesson was: you need people experienced with the system in-house to support it once its delivered or costs skyrocket.
After watching software projects be continuously late (for many reasons), overbudget and often outright fail, they decided to actively bring all development work in-house. The company decided the only way to control their destiny was to have people in IT that understood the business and had a vested interest in its destiny. Very enlightened philosophy in this climate and it works. They went on an IT hiring spree about a year ago and today 90% of its software development is done by permanent employees. Mind you, it is a very small number of employees (there are less than 10 permanently staffed programmers in the company worldwide) but they are talented and highly motivated. And the company rewards and nutures success.
Outsourcing is a short term cost savings when you are looking one or two quarters ahead. Long term, outsourcing is much more expensive done improperly (and most outsourcing is done improperly).
Coding "low skill"? To the uninitiated, maybe. Mentally gruelling, unappreciated and only moderately well compensated for the level of effort involved to produce good, practical code.
Bottom line, people who write code who believe in the business and take a great degree of professional pride in their work (even if they are the only ones who see it) will produce software 10 times better than any number of "average", low cost programmers. Yes, you can get coders overseas for 1/4 the cost, but the ratio of exceptional to average programmers is about the same. Personally, our IT shop with our five staff programmers has more talent than 100 average programmers and will produce on-time with high quality code every time.
In order to settle, the other side has to want that to happen. IBM does not want that to happen. IBM was the settlement to come from the result of a trial, not by handing over a check or signing an agreement to end it. Nothing short of "guilty as charged" or "Bankruptcy for SCO" will be acceptable outcomes to this case from IBM's perspective.
I love articles like this and clueless ones who continue to come up with half-assed ideas like this.
The problem with spam is that it turns the traditional "bulk mail" model around.
We all understand junk mail. We get crap in our mail boxes all the time: free offers, ads, etc. But you don't see us complaining too much. Why? Because the SENDER is paying out their pocket for the privilege of "spamming" you via the postal service. Even at bulk mail rates, sending a million flyers out isn't a cheap option except for corporations.
When fax machines became the next great thing in communications, junk fax appeared almost immediately. It took an act of legislation to stop that. How was junk fax different from junk mail? Simple: the RECEIVER paid to receive it! Not in terms of a direct payment, but in terms of consumables. It cost the RECEIVER money in terms of paper and ink to reproduce the received junk fax into paper form. In countries that metered fax/data lines, they was also the possible cost of phone charges. Because a junk faxer is essentially stealing from the receiver of the junk, the receiver either needs to be compensated or you do it under penalty of law. It is theft otherwise. As long as the onus is on the sender to incur all costs of advertising, that is acceptable and right to the average person. The general public should NOT be forced to pay to receive advertising. The junk fax laws are harsh enough that you hardly hear about people receiving tons of junk fax nowadays.
When you get right down to it, that is all spam is: junk fax. Instead of stealing paper and ink, the spammers are stealing bandwidth (not theirs but that of the ISP sending the mail) and storage (both on the ISP's mail server and in your personal mailbox). Although its tough to touch the bits flowing across a network, it is still a limited, consumable resource and the person using it should be compensating the person who owns it or they should be barred from using it.
I don't know how to solve the overall problem but I have ideas about where it should go.
We need to turn the tables on the spammers and turn spam into a SENDER PAYS system like junk mail. I'd be more accepting of spam if I knew the spammer was paying the 5 cent bulk spam rate per e-mail to get access to a mail server to blast their messages out there. We could all enjoy the idea that this fuckup just spent $1.5M to send 30 million "bigger breasts/penis" spams and 99.9% got flushed. Suddenly, it isn't so profitable anymore to send spam. Such a system would at least produce less spam but probably much more tightly marketed and targeted.
Mail relays should be run in a closed mode and only exchange mail with hosts they are setup to recognize and trust. Open relays should be banned outright. Simple: if a downstream ISP mail server suddenly receives a flood of spams from an upstream open relay, that server has all its traffic blocked immediately, no questions asked and no quarter given, until its owner rectifies the issue. Trust me, if that relay was the mail server for 100,000 paying customers who suddenly couldn't send e-mail, that owner would be fixing the problem REAL FAST. That's their one warning. Do it again and the server is blocked permanently. Yes, a few examples will need to be made but the message will get across: Run an open mail relay on the Internet under penalty of virtual death.
Apply telecommunications and junk fax laws to spam. I consider spam a "theft of service". These bastards and scumbags are stealing MY PAID FOR bandwidth AND MY storage and processor time to wade through their crap. Make sending unsolicited e-mail the equivalent of telecommunications theft. One count per e-mail. And place the burden of proof on the spammer to prove beyond a shadow of a doubt that they have my consent to send the e-mail to me. Assume the spammer is guilty. Construct this very narrowly but make sure the spammer knows: you need to prove I gave you permission to do this. Make the penalt
Re:Going after HP's customers...
on
SCO News Roundup
·
· Score: 2, Interesting
Just because HP is a 500lb gorilla instead of an 800lb gorilla doesn't make them any less dangerous to mess with. I have a feeling if SCO actually sues an HP Linux user that is covered under their indeminification policy, one of two things will happen:
HP will back off and construe their indeminification policy in such narrow terms that it won't apply to the sued customer let alone anyone else. Questionable whether HP could withstand the backlash that would then ensue
HP digs in tooth-and-nail and fights back with the same tenacity as IBM.
It's bad enough to have IBM after you in a grudge match. Remember, IBM tied up the Federal Gov. in antitrust actions for the better part of a decade. A small fry like SCO stands no chance against legal endurance like that. Multiply it by two and have a pissed IBM and HP after you? I'd sooner round Cape Horn on a liferaft in the middle of a south seas gale than fight those two at the same time.
Darl, remember men named Napoleon and Hitler who started two front wars? Remember which one of them succeeded? Take a hint from history and think about this...oh wait, you need a BRAIN to think!
Idiot troll. If you pointed HST at the Earth, it's sensors would burn out in an instant. They already have spy sats with 96 inch+ mirrors up there looking down. KH-11 and KH-12. Been around since the 70s.
Can't do it. Hubble isn't hardened against deep space radiation beyond low earth orbit. The Van Allen Belt protects a lot of Hubble's circuitry. The HST has problems when it travels through the SAA (South Atlantic Anomaly). FYI, the SAA is a region of high radiation that dips lower than usual into low earth orbit that is normally deflected by the Van Allen belt.
If you park HST at a LaGrange point, it might get there in one piece but it's computers will be toast when it arrives.
Actually, NO ONE wants the UNIX Sys V code and all the contracts and copyrights that go with it! AT&T dumped it because it was becoming a pit they couldn't escape from in terms of contractual arrangements, royalties, rights, etc. So many companies are tied up in one way or the other with the original UNIX licensing agreements that it is quite the Gordian knot to unravel.
It was for sale and Novell stepped up to the plate. Sun didn't want it. IBM didn't want it. HP didn't want it. They knew what a disaster the thicket of licensing on the original UNIX was. Let it be someone else's problem. AT&T knew it and Novell found out shortly thereafter. Then they realized how much of a problem it was and went looking another sucker..er..buyer for the UNIX codebase. And SCO was dumb enough to think it was a good idea.
You can't open source Sys V. Too many confidentiality and licensing agreements are in place and those licensees would have to agree as a whole before that code could get released since all of them use some of that code in one form or another. I am sure that very little original SysV code exists in AIX except for compatibility purposes. SunOS/Solaris is based on BSD Unix so they probably licensed Sys V just to cover their bases.
Methods and concepts for Unix are already more or less in the public domain. The POSIX standard and the Open Group have that covered. Unix and Unix-like clones have been implemented and reimplemented so many times over the years and the ideas are so well known, who would want the Sys V code at all even for historical reasons? I'd sooner want to see the code to the PDP-11 than the Sys V codebase. At least the PDP-11 code has a rich history behind it that might bring back some fond memories for some. Plus, it's dead and gone. Unix is alive and kicking and if I want to see how Unix works, I need look no further than the Linux Kernel Whitepapers and Annotated Code guides.
Offer MD5 fingerprints for the version of the software that is compiled for use in an election along with the exact compilation and hardware settings used to build it. Print that MD5 sum on the screen for a voter along with a printed copy available at the polling station so a technically inclined voter can compare the printed fingerprint and the screen version.
If the two match, such a technically inclined voter can download the source code and using the provided build instructions and produce a binary that should produce the same fingerprint. You don't need data to build the code but the two damn well better match. By making the code available, anyone can configure a like environment and following the same instructions provided by the software vendor they should be able to verify the integrity of the build and settings used to produce. Identical source code and identical build settings should produce identical binaries. If there is a mismatch or no one can produce a matching binary, election officials and the software vendor better have a damn good explanation as to why.
Empower anyone with a technical bent to verify the process. There are enough folks out there who have the knowledge AND care about the democratic process that they would spend the time to perform the check and do it ongoing during an election. I would.
Just finished a quick bit of research and found a couple of interesting things...
1) I can download Caldera OpenLinux ISO images from here.
2) The SCO "register to access downloads" feature is a bunch of crap. Wander on over to Caldera's FTP site and download the source RPMs at will, the legal notice notwithstanding.
So, what happens if I install OpenLinux 3.1.1 workstation on my box at home (it is free and I have a spare machine I'm not using) and SCO closes the Caldera website down? If I go to SCO and say I need the sources, they ask "Are you a Caldera customer, I say, 'No, I am using a free downloaded version of OpenLinux 3.1.1' and they reply "You need to buy a SCO Linux License.", I think I have the grounds at that point to sue them for violation of the GPL's terms of distribution since they cannot deny me the sources on a product they distributed just because I didn't buy it.
The question is: Did SCO ever offer Caldera Linux for free download by anyone much in the same way SuSe or Red Hat offer free downloads of their products? If so, then SCO has an obligation to make those sources available to that community as well. Just because you didn't buy the product does not waive SCO's obligation under the GPL to make those sources available to you on demand for up to three years via electronic or physical media after your download.
If they did allow free downloads of Caldera, then they just committed an open, beyond-a-shadow-of-a-doubt violation of the GPL, not one that can be danced around with intrepretive, legal-sounding rhetoric.
The public does not care what their government runs on. What they do care about, as others have said, is being able to access data that is important to them freely.
Despite being a Linux user, I have mellowed over the past few years and I really don't care what an organization uses. If it gets the job done, so be it. My issue (both at home and in the corporate world) is being able to share data. As long as I can read the document, data file, database, spreadsheet, etc, using the tools of my choice and provide that data to my peers in the formats they need it in, what one uses to make use of the data is largely irrelevant. It isn't a Microsoft vs. Linux debate.
In government, the public has the right be able to access their data freely and fairly. Corporations can impose limits on the tools used to access its data (i.e. everyone must use IE on the Intranet). In the public domain, cost is secondary to interoperability. You cannot dictate to the public that every MUST use IE or they are out of luck in dealing with the government. They must support all, IMHO, at least 95% of the populace. That means supporting IE, Mozilla, Linux, Opera and Mac users, to name a few. And it isn't hard, kids. Just don't use browser specific extensions.
Use commonly available formats. Use HTML, PDF, XML. Offer documents in MS Word, StarOffice/OpenOffice (SXW), ASCII text, RTF and XML formats, to name a few. Offer data in CSV, text and XML formats. Do that and the public can choose their favorite tools, be it Windows, UNIX or Linux. Governments have a civic duty to do this if they want to offer their data electronically to the public-at-large.
Microsoft's true crime is the control of file formats. Break that one monopoly and their Windows desktop monopoly will start to come apart. Education eventually triumphs over ignorance. That one ruling in the antitrust suit could've changed the world. Break their lock on the data and the rest of their business won't be able to compete except on merit.
I use OpenOffice on Windows daily for document production despite the fact I have Office installed. Just personal preference. Only my immediate co-workers know the documents aren't being produced by Office. The rest of the business couldn't care less. As long as the data is transparent and sharable, the world doesn't care how it gets produced and prcessed. That has always been the key in the enterprise and should be priority one for any e-government initiative. Run whatever you want, just make sure anyone and everyone can make use of it with the tools of their choice.
If SCO actually manages to sell one license to a Fortune 1000 company and the name of that company comes out, I would not be surprised to hear about the FSF knocking on that company's doorstep to have their head counsel explain the terms of the GPL to them. The mere act of SCO selling the license violates the GPL immediately for BOTH SCO AND THE FORTUNE 1000 company! SCO cannot limit the rights of the Fortune 1000 company and prevent them from redistribution of the kernel so if the Fortune 1000 company was a Linux reseller/vendor/supplier, etc, then they are f*cked too!
I can see a countersuit coming along real fast with the FSF's backing. I don't necessarily agree with RMS's leanings but I do respect the license and the FSF in general and I can see them going after SCO once they had a blatant violation of the GPL handed them by SCO.
Can you imagine what will happen to SCO (assuming there is anything left of them when this is all said and done) if it comes out in the IBM/Red Hat trials that their Linux Kernel Personality (LKP) code contains GPL code? It's one thing to taunt a shark from the beach with a scrape on your palm, quite another to jump in the water with slit wrists. They'll get ripped apart.
Plus, then there's the fun of all the investor lawsuits when SCO's stock crashes through the floor.
SCO is in the wrong business. You can't pay for this type of entertainment! SCO Improv Amateur Night anyone?:)
What is wrong with Viacom/CBS? I think they and they are conveniently forgetting the fact that broadcast radio and television are a public commons. They exist in their free forms to ensure that the general public will always have fair and free access to general programming.
There is only one problem with forcing a conversion of the broadcast band to digital: by doing so you are eradicating the intent and spirit behind broadcast television. By sending signals, you are deriving folks of programming that they were receiving just fine with the older analog mode who do not wish to upgrade. In effect, you are creating a group of second-class citizens that are being denied access to the commons. It would be like cordoning off the Mall in Washington DC and saying only people who are wearing certain clothing will be free to use it and cross it. It is all fine and well as long as everyone agrees to the rules but causes serious problems when a tourist shows up in unapproved clothing and wants to enjoy the Mall. Can you imagine the firestorm that would ensue? Same thing here.
Media companies can do whatever they want on the for-pay cable networks. CBS wants to stop HDTV broadcasts over the public airwaves? Fine, go ahead. They will be just hurting themselves in the long run. Someone else will happily fill the HDTV void and attract viewers over. But HDTV should be forced upon the general public just because the media industry wants to do it. Unless, of course, they want to buy everyone who doesn't want to spend the money to upgrade an HD set. Then I don't see an issue with it.
How is HD piracy going to be any different really? I can tape a show on a good videotape, encode it and distribute it over the Internet with sufficient quality to be more than acceptable to the masses at large. Until we live in the age where everyone has a 100Mbps broadband connection, sharing 1000 scanline HDTV over the net, even with the best compression technology on earth, isn't going to be happening the near-term or even 5 years from now. This is a play by the media industry to exert an unprecedented amount of control over how you can exercise your fair use rights and the desire to not be forced to view advertising you don't want to see.
I think if the force the flag and the hardware upgrade, it will be a dismal failure. People may not understand or care how the Web works, how their DVD player works or how to program the VCR, but mess with their ability to use and watch their television and watch out! Cripple TV viewing and the backlash against the media conglomorates will be swift, brutal and violent. Just the way any good revolt should be.
The difference between Verisign and Microsoft is the fact that Microsoft implemented their redirection/helper page as an application feature in response to a DNS "unable to resolve" response. Microsoft's feature, whether you agree with it or not, only affects users of IE and doesn't not affect the functioning of the Internet worldwide. If you don't like the feature in IE, use Mozilla, Opera, etc, that don't implement it.
Verisign's redirection is done by changing the fundamental and expected behavior of DNS worldwide without allowing the user's to turn it off or use a different product. Ironically, the use of SiteFinder disables the Microsoft IE search feature since the non-existent web address now resolves!
To Verisign: the DNS is not yours! It is a utility like the power company. The power company just can't change the voltage they deliver to your house just because it might save/make them money and you can't change the DNS protocol interface just because you want to! IT DOES NOT BELONG YOU BUT THE WORLD-AT-LARGE, YOU UNETHICAL AND SHORT-SIGHTED BASTARDS!
Can these folks accept defeat gracefully and learn to play nice with others?
I've been running the Great Internet Mersenne Prime Search (GIMPS) for the past several years non-stop on several PCs. I leave the machines on 24 hours a day and only shut them down for hardware upgrades and an occasional cleaning.
The power costs are negligible on a single machine. Run a farm and it can get expensive when you factor in cooling, which is the primary expense. Air conditioning running 24/7 or close to it in a house is far more costly than the consumption of a typical PC.
The advantage of the heavy CPU usage clients like GIMPs is the fact they are often the first things to detect an impending CPU failure. My GIMPs client running on an Athlon wound up saving the machine. I brought it up to see its progress and it was reporting hardware register mismatches. Turns out the heat sink fan had failed and the CPU was overheating. Fortunately, there was enough supplemental cooling in the case to keep the chip from frying outright but Windows was chugging along fine without any indication of a problem while GIMPs was saying "turn me off or die".
These programs exercise the CPUs to limits that few programs ever use. They make wonderful test and benchmark applications. When Cray tested their supercomputer CPUs, they used to do prime number calculations since any error in the floating point hardware would come out instantly.
Sounds like just the latest generation of processors geared for number crunching. Not a new idea. Floating Point Systems built array processors that were tied to IBM mainframes and fed computations and data for them to chew on years ago. Since they were dedicated number crunchers, they blew away the mainframe by a factor of 10. Oil companies and other consumers of raw numerical power bought these things by the dozens because they were cheap ($100K vs. millions for the mainframe they were attached to) and could be moved from task to task (disconnect it and roll it to the workstation for that geology analysis...).
Heck, when the NSF built the five National Centers for Supercomputing Applications, one of the facilities used a bunch of FPS array processors to give them the number crunching power instead of a traditional IBM, CDC or Cray supercomputer.
Nothing special about these guys and I'd take what they have with a grain of salt. Unless they're offering half a teraflop for $500K out-the-door in real world workloads (fluid dynamics, oil exploration, weather simulations, finite element analysis, etc), forget it. 25GFlops peak? Big deal. It's what it does on real work that matters, not the amount of useless calculations you can perform. Plus, they don't even have pre-production silicon. The problem is the compiling and the programming. If you can get those tasks down and be able to write code that can harness that power efficiently, their nothing but expensive serial computers at best and boat anchors at worst.
Absolutely true. Programming these beasts is the true challenge, not acquiring the hardware. Theoretical peak performance is measured with every CPU maxed at 100% with little or communication latency. Tack in message passing and hungry CPUs waiting for data from another node, plus the latency of the interconnect and your performance plummets. Plus, the application domain has a huge effect. Applications that don't or can't scale to the nodes won't make use of that untapped power.
The Top 500 benchmark is a reasonable test designed to test both theoretical and "real world" workloads assuming application loads that exhibit a high degree of parallelism. Realistically, such a cluster (including the G5 based one) can reasonably expect to achieve 25-60% of its theoretical performance on real world work depending on everything factored in.
Still, a few TFlops is impressive. Sites are shooting for 100TFlops sustained by 2008-2010. Work is on the horizon on a Petaflop machine.
What's amusing is the fact that nowadays a lot of folks have sitting under their desks the equivalent of a Cray 2 supercomputer that cost millions of dollars in the 1980s and could beat it soundly. The Cray 2 was the last of the serial supercomputers and had a peak throughput of 2GFlops clocked at 500Mhz (2ns). A modern Athlon or P4 can hammer that and contains equal or better memory and storage than what that Cray had access to. That was not even 20 years ago! Isn't technology grand.
Check out O'Reilly's High Performance Computing, 2nd Ed for an excellent primer on supercomputing and clusters. Great reading if you are interested in this stuff.
Unfortunately, due to a software bug, the story goes, the robot would sometimes lift a captured piece and crush it into powder.
I don't think that endgame was programmed into the computer.:)
Or maybe it was intended to be human intimidation. Robots crushes chess piece, human holds up taser gun and looks for weak point on robot. I think a robot would back off when threatened with 50,000 volts.
Slashdot Mirror
Because the USPS issues money orders. They are subject to financial regulations and are required to refuse or report the purchase of money orders over a specific amount (in the name of another war, the war on drugs). It to prevent drug dealers from laundering money since money order can be used as cash.
I would mod this up! :) This is too funny.
Unless SCO is claiming ownership of the meanings of some NUMBERS, then that effectively gets around the problem. Besides, haven't these constants been around forever and been used and reused throughout UNIX history?
However, the works in question contain the copyrights of dozens or hundreds of indviduals, licensed jointly. So you are receiving a financial because when you agree to the license, you gain access to all of the copyrighted works as well as the right to modify and distribute them on your own. And the license also gives you the ability to make money off that copyrighted work, if you so chose and can convince someone to pay you for it. So there is definitely a "financial gain" angle there and that option to make money on GPL'd works is explicitly stated in the GPL itself.
Under a strict GPL interpretation, modules can be considered "derived works" because they link dynamically into a GPL environment (the kernel). The modules depends on kernel headers to compile and use internal kernel data structures exposed through module interfaces that are technically under the GPL. Thus, if you interpret the GPL as written, then yes, all modules would be required to be released under the GPL.
Clearly such a situation is not acceptable to hardware manufacturers and Linus recognized this. Hence his "binary only is allowed for modules" rule. It is the kernel developers themselves that are changing the module environment to try and enforce the strict constructionist view of the GPL.
RMS recognized this issue and developled the LGPL. The intent behind the LGPL was to allow for the development of GPL'd code that could be incorporated into binary-only products without tainting those products provided the binary product announced was linking to LGPL code and made it available on request (or distributed it with the product). As long as a clear separation exists between the proprietary product and the LGPL'd code (in other words, the proprietary code links to the LGPL'd work. If it copied LGPL'd code into its code base, then it would become tainted), the two models can co-exist.
RMS doesn't like the LGPL. It is a compromise license designed to fit the realities of the world rather than what he wishes it could be.
In my opinion, maybe the simplest solution is to relicense future versions of the Linux kernel under the LGPL. The distribution and modification requirements for the kernel proper would still be in force as with the GPL and make it compatible with binary-only modules explicitly since this very issue of linkage and mixing is exactly what the LGPL was designed to address. For the record, I prefer the LGPL over the GPL for practical, real-world open source development. It provides the best of both worlds when approached correctly.
If the LGPL cannot fit, then clarification is required for the definition of derived works as encompassed by copyright law and how the GPL perceives it. v3 of the license should maybe incorporate a better, clearer definition of dynamic linking when no source is exchanged between two programs.
I agree with Linus. Binary-only modules should be acceptable under the GPL since they do not incorporate GPL'd works directly except through runtime linkage.
After watching software projects be continuously late (for many reasons), overbudget and often outright fail, they decided to actively bring all development work in-house. The company decided the only way to control their destiny was to have people in IT that understood the business and had a vested interest in its destiny. Very enlightened philosophy in this climate and it works. They went on an IT hiring spree about a year ago and today 90% of its software development is done by permanent employees. Mind you, it is a very small number of employees (there are less than 10 permanently staffed programmers in the company worldwide) but they are talented and highly motivated. And the company rewards and nutures success.
Outsourcing is a short term cost savings when you are looking one or two quarters ahead. Long term, outsourcing is much more expensive done improperly (and most outsourcing is done improperly).
Coding "low skill"? To the uninitiated, maybe. Mentally gruelling, unappreciated and only moderately well compensated for the level of effort involved to produce good, practical code.
Bottom line, people who write code who believe in the business and take a great degree of professional pride in their work (even if they are the only ones who see it) will produce software 10 times better than any number of "average", low cost programmers. Yes, you can get coders overseas for 1/4 the cost, but the ratio of exceptional to average programmers is about the same. Personally, our IT shop with our five staff programmers has more talent than 100 average programmers and will produce on-time with high quality code every time.
In order to settle, the other side has to want that to happen. IBM does not want that to happen. IBM was the settlement to come from the result of a trial, not by handing over a check or signing an agreement to end it. Nothing short of "guilty as charged" or "Bankruptcy for SCO" will be acceptable outcomes to this case from IBM's perspective.
The problem with spam is that it turns the traditional "bulk mail" model around.
We all understand junk mail. We get crap in our mail boxes all the time: free offers, ads, etc. But you don't see us complaining too much. Why? Because the SENDER is paying out their pocket for the privilege of "spamming" you via the postal service. Even at bulk mail rates, sending a million flyers out isn't a cheap option except for corporations.
When fax machines became the next great thing in communications, junk fax appeared almost immediately. It took an act of legislation to stop that. How was junk fax different from junk mail? Simple: the RECEIVER paid to receive it! Not in terms of a direct payment, but in terms of consumables. It cost the RECEIVER money in terms of paper and ink to reproduce the received junk fax into paper form. In countries that metered fax/data lines, they was also the possible cost of phone charges. Because a junk faxer is essentially stealing from the receiver of the junk, the receiver either needs to be compensated or you do it under penalty of law. It is theft otherwise. As long as the onus is on the sender to incur all costs of advertising, that is acceptable and right to the average person. The general public should NOT be forced to pay to receive advertising. The junk fax laws are harsh enough that you hardly hear about people receiving tons of junk fax nowadays.
When you get right down to it, that is all spam is: junk fax. Instead of stealing paper and ink, the spammers are stealing bandwidth (not theirs but that of the ISP sending the mail) and storage (both on the ISP's mail server and in your personal mailbox). Although its tough to touch the bits flowing across a network, it is still a limited, consumable resource and the person using it should be compensating the person who owns it or they should be barred from using it.
I don't know how to solve the overall problem but I have ideas about where it should go.
It's bad enough to have IBM after you in a grudge match. Remember, IBM tied up the Federal Gov. in antitrust actions for the better part of a decade. A small fry like SCO stands no chance against legal endurance like that. Multiply it by two and have a pissed IBM and HP after you? I'd sooner round Cape Horn on a liferaft in the middle of a south seas gale than fight those two at the same time.
Darl, remember men named Napoleon and Hitler who started two front wars? Remember which one of them succeeded? Take a hint from history and think about this...oh wait, you need a BRAIN to think!
Idiot troll. If you pointed HST at the Earth, it's sensors would burn out in an instant. They already have spy sats with 96 inch+ mirrors up there looking down. KH-11 and KH-12. Been around since the 70s.
If you park HST at a LaGrange point, it might get there in one piece but it's computers will be toast when it arrives.
It was for sale and Novell stepped up to the plate. Sun didn't want it. IBM didn't want it. HP didn't want it. They knew what a disaster the thicket of licensing on the original UNIX was. Let it be someone else's problem. AT&T knew it and Novell found out shortly thereafter. Then they realized how much of a problem it was and went looking another sucker..er..buyer for the UNIX codebase. And SCO was dumb enough to think it was a good idea.
You can't open source Sys V. Too many confidentiality and licensing agreements are in place and those licensees would have to agree as a whole before that code could get released since all of them use some of that code in one form or another. I am sure that very little original SysV code exists in AIX except for compatibility purposes. SunOS/Solaris is based on BSD Unix so they probably licensed Sys V just to cover their bases.
Methods and concepts for Unix are already more or less in the public domain. The POSIX standard and the Open Group have that covered. Unix and Unix-like clones have been implemented and reimplemented so many times over the years and the ideas are so well known, who would want the Sys V code at all even for historical reasons? I'd sooner want to see the code to the PDP-11 than the Sys V codebase. At least the PDP-11 code has a rich history behind it that might bring back some fond memories for some. Plus, it's dead and gone. Unix is alive and kicking and if I want to see how Unix works, I need look no further than the Linux Kernel Whitepapers and Annotated Code guides.
If the two match, such a technically inclined voter can download the source code and using the provided build instructions and produce a binary that should produce the same fingerprint. You don't need data to build the code but the two damn well better match. By making the code available, anyone can configure a like environment and following the same instructions provided by the software vendor they should be able to verify the integrity of the build and settings used to produce. Identical source code and identical build settings should produce identical binaries. If there is a mismatch or no one can produce a matching binary, election officials and the software vendor better have a damn good explanation as to why.
Empower anyone with a technical bent to verify the process. There are enough folks out there who have the knowledge AND care about the democratic process that they would spend the time to perform the check and do it ongoing during an election. I would.
Right here. Knock yourself out.
1) I can download Caldera OpenLinux ISO images from here.
2) The SCO "register to access downloads" feature is a bunch of crap. Wander on over to Caldera's FTP site and download the source RPMs at will, the legal notice notwithstanding.
So, what happens if I install OpenLinux 3.1.1 workstation on my box at home (it is free and I have a spare machine I'm not using) and SCO closes the Caldera website down? If I go to SCO and say I need the sources, they ask "Are you a Caldera customer, I say, 'No, I am using a free downloaded version of OpenLinux 3.1.1' and they reply "You need to buy a SCO Linux License.", I think I have the grounds at that point to sue them for violation of the GPL's terms of distribution since they cannot deny me the sources on a product they distributed just because I didn't buy it.
Also, here is a fun page on SCO's own website OpenLinux Supplemental Open Source Software. Hmm, does IBM need some more ammunition provided by SCO themselves?
If they did allow free downloads of Caldera, then they just committed an open, beyond-a-shadow-of-a-doubt violation of the GPL, not one that can be danced around with intrepretive, legal-sounding rhetoric.
Despite being a Linux user, I have mellowed over the past few years and I really don't care what an organization uses. If it gets the job done, so be it. My issue (both at home and in the corporate world) is being able to share data. As long as I can read the document, data file, database, spreadsheet, etc, using the tools of my choice and provide that data to my peers in the formats they need it in, what one uses to make use of the data is largely irrelevant. It isn't a Microsoft vs. Linux debate.
In government, the public has the right be able to access their data freely and fairly. Corporations can impose limits on the tools used to access its data (i.e. everyone must use IE on the Intranet). In the public domain, cost is secondary to interoperability. You cannot dictate to the public that every MUST use IE or they are out of luck in dealing with the government. They must support all, IMHO, at least 95% of the populace. That means supporting IE, Mozilla, Linux, Opera and Mac users, to name a few. And it isn't hard, kids. Just don't use browser specific extensions.
Use commonly available formats. Use HTML, PDF, XML. Offer documents in MS Word, StarOffice/OpenOffice (SXW), ASCII text, RTF and XML formats, to name a few. Offer data in CSV, text and XML formats. Do that and the public can choose their favorite tools, be it Windows, UNIX or Linux. Governments have a civic duty to do this if they want to offer their data electronically to the public-at-large.
Microsoft's true crime is the control of file formats. Break that one monopoly and their Windows desktop monopoly will start to come apart. Education eventually triumphs over ignorance. That one ruling in the antitrust suit could've changed the world. Break their lock on the data and the rest of their business won't be able to compete except on merit.
I use OpenOffice on Windows daily for document production despite the fact I have Office installed. Just personal preference. Only my immediate co-workers know the documents aren't being produced by Office. The rest of the business couldn't care less. As long as the data is transparent and sharable, the world doesn't care how it gets produced and prcessed. That has always been the key in the enterprise and should be priority one for any e-government initiative. Run whatever you want, just make sure anyone and everyone can make use of it with the tools of their choice.
I can see a countersuit coming along real fast with the FSF's backing. I don't necessarily agree with RMS's leanings but I do respect the license and the FSF in general and I can see them going after SCO once they had a blatant violation of the GPL handed them by SCO.
Can you imagine what will happen to SCO (assuming there is anything left of them when this is all said and done) if it comes out in the IBM/Red Hat trials that their Linux Kernel Personality (LKP) code contains GPL code? It's one thing to taunt a shark from the beach with a scrape on your palm, quite another to jump in the water with slit wrists. They'll get ripped apart.
Plus, then there's the fun of all the investor lawsuits when SCO's stock crashes through the floor.
SCO is in the wrong business. You can't pay for this type of entertainment! SCO Improv Amateur Night anyone? :)
There is only one problem with forcing a conversion of the broadcast band to digital: by doing so you are eradicating the intent and spirit behind broadcast television. By sending signals, you are deriving folks of programming that they were receiving just fine with the older analog mode who do not wish to upgrade. In effect, you are creating a group of second-class citizens that are being denied access to the commons. It would be like cordoning off the Mall in Washington DC and saying only people who are wearing certain clothing will be free to use it and cross it. It is all fine and well as long as everyone agrees to the rules but causes serious problems when a tourist shows up in unapproved clothing and wants to enjoy the Mall. Can you imagine the firestorm that would ensue? Same thing here.
Media companies can do whatever they want on the for-pay cable networks. CBS wants to stop HDTV broadcasts over the public airwaves? Fine, go ahead. They will be just hurting themselves in the long run. Someone else will happily fill the HDTV void and attract viewers over. But HDTV should be forced upon the general public just because the media industry wants to do it. Unless, of course, they want to buy everyone who doesn't want to spend the money to upgrade an HD set. Then I don't see an issue with it.
How is HD piracy going to be any different really? I can tape a show on a good videotape, encode it and distribute it over the Internet with sufficient quality to be more than acceptable to the masses at large. Until we live in the age where everyone has a 100Mbps broadband connection, sharing 1000 scanline HDTV over the net, even with the best compression technology on earth, isn't going to be happening the near-term or even 5 years from now. This is a play by the media industry to exert an unprecedented amount of control over how you can exercise your fair use rights and the desire to not be forced to view advertising you don't want to see.
I think if the force the flag and the hardware upgrade, it will be a dismal failure. People may not understand or care how the Web works, how their DVD player works or how to program the VCR, but mess with their ability to use and watch their television and watch out! Cripple TV viewing and the backlash against the media conglomorates will be swift, brutal and violent. Just the way any good revolt should be.
Verisign's redirection is done by changing the fundamental and expected behavior of DNS worldwide without allowing the user's to turn it off or use a different product. Ironically, the use of SiteFinder disables the Microsoft IE search feature since the non-existent web address now resolves!
To Verisign: the DNS is not yours! It is a utility like the power company. The power company just can't change the voltage they deliver to your house just because it might save/make them money and you can't change the DNS protocol interface just because you want to! IT DOES NOT BELONG YOU BUT THE WORLD-AT-LARGE, YOU UNETHICAL AND SHORT-SIGHTED BASTARDS!
Can these folks accept defeat gracefully and learn to play nice with others?
The power costs are negligible on a single machine. Run a farm and it can get expensive when you factor in cooling, which is the primary expense. Air conditioning running 24/7 or close to it in a house is far more costly than the consumption of a typical PC.
The advantage of the heavy CPU usage clients like GIMPs is the fact they are often the first things to detect an impending CPU failure. My GIMPs client running on an Athlon wound up saving the machine. I brought it up to see its progress and it was reporting hardware register mismatches. Turns out the heat sink fan had failed and the CPU was overheating. Fortunately, there was enough supplemental cooling in the case to keep the chip from frying outright but Windows was chugging along fine without any indication of a problem while GIMPs was saying "turn me off or die".
These programs exercise the CPUs to limits that few programs ever use. They make wonderful test and benchmark applications. When Cray tested their supercomputer CPUs, they used to do prime number calculations since any error in the floating point hardware would come out instantly.
Heck, when the NSF built the five National Centers for Supercomputing Applications, one of the facilities used a bunch of FPS array processors to give them the number crunching power instead of a traditional IBM, CDC or Cray supercomputer.
Nothing special about these guys and I'd take what they have with a grain of salt. Unless they're offering half a teraflop for $500K out-the-door in real world workloads (fluid dynamics, oil exploration, weather simulations, finite element analysis, etc), forget it. 25GFlops peak? Big deal. It's what it does on real work that matters, not the amount of useless calculations you can perform. Plus, they don't even have pre-production silicon. The problem is the compiling and the programming. If you can get those tasks down and be able to write code that can harness that power efficiently, their nothing but expensive serial computers at best and boat anchors at worst.
The Top 500 benchmark is a reasonable test designed to test both theoretical and "real world" workloads assuming application loads that exhibit a high degree of parallelism. Realistically, such a cluster (including the G5 based one) can reasonably expect to achieve 25-60% of its theoretical performance on real world work depending on everything factored in.
Still, a few TFlops is impressive. Sites are shooting for 100TFlops sustained by 2008-2010. Work is on the horizon on a Petaflop machine.
What's amusing is the fact that nowadays a lot of folks have sitting under their desks the equivalent of a Cray 2 supercomputer that cost millions of dollars in the 1980s and could beat it soundly. The Cray 2 was the last of the serial supercomputers and had a peak throughput of 2GFlops clocked at 500Mhz (2ns). A modern Athlon or P4 can hammer that and contains equal or better memory and storage than what that Cray had access to. That was not even 20 years ago! Isn't technology grand.
Check out O'Reilly's High Performance Computing, 2nd Ed for an excellent primer on supercomputing and clusters. Great reading if you are interested in this stuff.
I don't think that endgame was programmed into the computer. :)
Or maybe it was intended to be human intimidation. Robots crushes chess piece, human holds up taser gun and looks for weak point on robot. I think a robot would back off when threatened with 50,000 volts.
Yes, it does.