Other NForce2 boards are definitely having hard lockup problems though, including the MSI K7N2, and i've seen posts about similar problems with other NForce2 boards from different manufacturers.
Why NVidia doesnt just come clean and release the specs for the chips i'll never know, since its only going to get them bad press, I for one have spent hours of frustration on these problems.
Actually, Linux support for NForce2 is not very good - there are bugs in the chipset, or workarounds in the Windows drivers that the kernel developers are still working on.
SATA support is also pretty poor - several popular controllers either dont work, work at about half the speed in linux as they do under Windows, or won't work with software RAID-1 etc.
Have a look at recent postings to the Linux Kernel mailing list to see the nightmare that an NForce2-based board, or a SATA controller will give you under Linux.
I have both, and while I have got them to work, I had frequent hard lockups before i disabled all the ACPI/APIC stuff, my SATA controller doesn't work with software RAID-1 and 2.4 kernels gave me disk corruption and hard lockups under load.
However, The kernel developers are working on these issues, and with their help I was able to get my system up and running. I am confident that this stuff will be fully supported and stable under Linux, but unfortunately this is not the case today.
That they are going to make more money off Linux than they ever possibly could off Solaris, do a complete about face, and proclaim 'Linux is the best choice for the server as well as the desktop, and Solaris is `legacy` technology.'
SCO has a self-confessed monopoly on the ability to sell fake Linux licenses.
As such, surely it is bound by the various anti-trust laws that prevent it from doing exactly this. ('Oh no you can't buy one, only our Fortune 1000 buddies can')
If it is going to sell it's product, it has to make it available to everyone under the same terms (volume discounts might apply to Fortune 1000 companies, but thats about it).
Someone better call the US DOJ and get them to crack down on these abusers of monopoly power.
This guy, obviously with more than a few clues about security, is able to examine the products, right down to the source level, analyse the security provided, freely publish his findings and suggest improvements (even if all he suggests is 'scrap it', and something about skull-fucking with sound-waves.)
This is great information, and while it might not reinforce the 'open source uber alles' message, it is very useful to anyone who might be considering working on these or similar projects, as well as anyone that uses them.
Even if Mr. Gutmann says these products can't be completely fixed, at least the authors can improve them now based on his comments, and just because this guy says it can't be done, doesn't mean it is gospel.
I say a big thank you to Peter Gutmann (a fellow kiwi, alright!) for taking the time to write this and help to improve the state of open source security products.
If Microsoft loses, sure they can change their browser, but if you believe SCO etc. everyone who has such a piece of software in their possession should be liable for license charges levied by Eolas.
Multiple ports are not the problem - if nothing is using those ports, there would be no traffic on them.
Blocking ports will only cripple legitimate users of those services while the malicious attackers will find other vectors for attack.
You can keep blocking ports until everything is tunnelled over port 80 and content only flows 'one way', but we already have that - its called TV/Radio broadcasting.
If anything, ISPs should filter the users logging onto their systems - e.g. if the system logging on fails security tests, or exhibits virus-carrying behaviour, then outbound access is curtailed or disabled entirely.
Crippling the internet because Microsoft can't get their shit together is the dumbest thing i've heard this week.
What, NEVER EVER Trust anyone except for Microsoft, who you grant rights to access your computer systems and do whatever they like, whenever they like, by virtue of their EULA?
You aren't actually allowed not to trust Microsoft any more, if you run their software.
Microsoft offers no indemnification for Internet Explorer, as far as I am aware, so should companies using IE hire a team of lawyers and pregrammers to carefully vet every single technology used in these products to avoid liability in this kind of lawsuit?
Red Hat are in business to make money - they do this by providing paid-for distributions with full support, custom-tweaked kernels and applications, and provide a validated platform on which to run commercial apps like Oracle.
Want to put that together yourself? Go for it, nobody at Redhat is stopping you. All the stuff they integrate in their product is free, just go do it yourself.
But don't complain because you can't do it yourself and don't see why you should pay Redhat to provide you with a quality product.
Its not like you don't have a choice of vendors, or that your apps only run on a single vendor's platform.
Linux is never free - you either pay for it with money, or you pay for it with your commitment to the GPL and/or the time you invest into making it work for you.
We need people like you in the Linux community i.e. 'waah waah linux is too expensive, even when i can download it for free' like we need a frickin hole in the head.
So what does AfterStep do that WindowMaker doesn't? Its not very clear from the website.
WindowMaker CVS has antialiased fonts, and with a little bit of work on icons and tile backgrounds, my desktop looks great.
The menu editor/prefs utility is easy to use, the clip's desktop switching is perfect for what I want, theres nothing obvious about AfterStep that would make me want to switch back.
WindowMaker + GNOME apps/libs + ROX-Filer make up my ideal GUI environment.
As far as I know, theyre both written in C, as opposed to Objective-C for GNUStep, they both attempt to mimic some or all aspects of NeXTSTEP, and WindowMaker was created because the Afterstep crew was too anal about incorporating features that did not appear in the original NeXT system.
So now theyre adding features to Afterstep - dumping the whole idea behind what AfterStep was - 'Its strictly a clone of NeXTSTEP', and somehow this is a step (STEP?) forward?
So why switch from WindowMaker, when AfterStep is clearly just playing catch-up now?
That being said, there are a few rough edges in Window Maker, so perhaps competition from AfterStep will smooth them out.
If IBM were to make AIX open source - that is, make the AIX kernel, which included and depended on SysV code to function, open source, then they would indeed be in breach of their UNIX license.
AIX, as an OS kernel which includes Sys V code, is indeed a derivative work.
However, to claim that any SysV-derived UNIX cannot and has not received IP contributions from existing projects, which themselves are not licensed under the terms of the AT&T/Novell license, and as such distributing these works independently is illegal, is ludicrous.
Like, stunningly, incredibly ludicrous. In fact, I would say such a licensing agreement violates anti-trust laws, and would be ruled illegal and unenforceable, possibly rendering the entire contract void.
Free/Open/NetBSD cannot legally exist with this interpretation of 'derivative work', because they include code that was once linked with SysV code.
According to SCO, it doesn't matter if the BSD people own the copyright to every line of code, because it was once linked with SysV code, and hence is a derivative work.As such, they may not disclose it
This is quite clearly not the situation.
To claim that any IP placed in any SysV-derived UNIX strips the copyright-holder of that work's right to redistribute the original work under any license they see fit is absurd.
Are they really trying to say that because the text of the BSD license is included in SysV that nobody may distribute the text of the BSD license because it is covered by AT&T licensing agreements, and must be kept secret?
If this release was not authorized by AOL, and it isn't entirely clear that Nullsoft did not have the right to do this, then I would say that you would be forbidden from distributing the code, regardless of the fact that it carries a GPL license.
You can't take, for example, Microsoft sourcecode from the WinCE SDK, slap a GPL license on it and claim it is now covered by the GPL.
If, however, the WinCE division at MS, who presumably has full responsibility and authority to handle code releases put the WinCE code under the GPL and released it, in good faith, then that would be binding.
i.e. if the city of New York sells the Brooklyn Bridge for $1 in a legal transaction, then the new owner owns that bridge, regardless of the seemingly low price.
If it was cool to buy hundreds of thousands of acres of land off the indigenous peoples of america for some muskets and smallpox-infected blankets, this is cool too.
However, the code has been released, so while you could not distribute the code as-is, there is nothing to prevent anyone studying the code and and releasing a compatible implementation, unless it infringes copyright (contains cut n pasted sections) or patents (not sure whether P2P patents exist or who owns them). You are not doing anything illegal if you use an MP3 (regardless of its origin) of a pop song to figure out how to make a pop song of your own. The person who sold/gave you the MP3 might have a problem however, and the owner of the copyright that covers the MP3 could demand you destroy/return it upon discovery of it's improper distribution to you.
This would have to be ordered by a court to be legally enforceable, but you may be guilty of a crime by delaying the destruction/return if you do this in bad faith. i.e you know the copyright they hold is valid, yet you ignore their reasonable and legal request for its destruction/return.
AOL could, assuming their claims that Nullsoft were not authorised to release the code under GPL are true, sue anyone they can prove is distributing the code for copyright infringement.
However, if such a lawsuit was pressed, you could request that AOL prove that Nullsoft were not authorised to release software that carried AOL-owned copyright, or that they prove that Nullsoft were acting in bad faith - that is they knew the licensing terms of the software in question would violate the law or go against AOL's wishes.
If they cannot prove this, then I would guess the GPL stands, and tough cookies AOL.
But certainly the mere presence of a GPL notice does not convey legitimacy to the GPL license terms.
So, what it really comes down to is 'do you trust AOL to tell you the truth that this code was released improperly'.
If you can convince a court that you were honestly unable to determine the legitimacy of their demands (not sure you could use this as a defense against AOL, it would be watertight against SCO), then you are also OK for keeping and distributing this code despite requests from them to remove it.
After all, they can lie to you about this and not run any significant risk because of the size of their bank account, yet you have no way to verify the authenticity of their claim without a court order for them to unseal the terms of their contracts with Nullsoft, or their sworn statement in court of law.
How can we know that AOL is not lying about the fact they did legitimately GPL this software, and since there is no law against making false claims outside the realm of contract or consumer law, it seems a pretty murky area. Its not like AOL has any disincentive for lying about this.
I can stand up and say 'I am the Pope of Hudson County! Bow before my lily white ass', but failing to bow is not a crime, much to my chagrin.
It is an interesting position, and bears remarkable similarity to the whole SCO debacle.
I would have thought you could avoid the need for the HDD etc. if you could boot a miniaml linux distro off an 8MB memory card, and use the CPU power of the EE along with NAS for data storage.
Is this possible?
But wouldn't this apply to the GPL too?
on
Today's SCO News
·
· Score: 1
Vendor A (disposable puppet company supported by Vendor B) takes Linux kernel, and illegally licenses kernel under BSD license to Vendor B.
Vendor A gets sued by FSF and goes out of business.
By this logic, Vendor B still retains the BSD-licensed Linux kernel, and can build a closed-source proprietary OS around it?
Running the DB from RAM is nice, but as far as I can see this won;t require any changes to the software itself, you could just mount your DB on a RAMdisk and be done with it. Whats the big deal?
What MySQL and PostgreSQL really lack is the ability to replicate on-the-fly and to support running on clusters for *real* failover and fault tolerance.
For Postgres, this means multiple 'postmaster' processes being able to access the same database concurrently, and probably something similar for MySQL.
Being able to run a database on an OpenMOSIX cluster, for example, would make it massively scalable, and being able to run multiple independent machines with an existing HA (High Availability) monitoring system would provide a truly fault-tolerant database.
There are of course major technical difficulties involved in making databases work this way, but an Open Source DB that can compete with Oracle's 'Unbreakable' claims would be a huge shot in the arm for OSS in the business world.
One thing that nobody has ever been able to satisfactorily explain to me is how the global end-to-end routing that IPV6 offers will operate in practice.
IPV4 can't be routed in this manner, because of router resource limits and the unreliability of route information - e.g. keeping route information for every individual/24 would overload most routers currently in use, and that John J. Average can advertise routes for microsoft.com and traffic goes down a black hole.
CIDR (Classless Internet Domain Routing) alleviates both these problems to some degree by removing the ability to effectively route any address block smaller than a/16. Since the only people who can realistically maintain a/16 are ISPs, mega-corps or state institutions, this basically takes the 'dream of a global network in which everyone is connected', breaks its legs and ties it directly to the financial interests of it's largest players.
Some would say (and I would largely agree) that the commercial benefits to infrastructure providers of offering routing to the 'smaller' nodes on the internet are non-existent or negative. So it just won't happen.
CIDR + IpV6 gives us nothing we don't already have. It simply offers more addresses behind each NAT machine - and really, nobody I have ever heard of is NATing more machines than will fill the 192.168.0.0/16 + 10.10.0.0/16 blocks.
I could be wrong, maybe new routing protocols have been developed to handle IPv6 routing out to network's edge, and maybe these protocols have been adopted as industry standards and are present in the firmware of all IPv6-capable hardware from the vendors that matter.(I don't imagine it will be a problem to deploy this stuff to 'software' routers)
I, along with many others, desperately want to be able to take advantage of the internet's fault-tolerance, but are prevented from doing so by CIDR etc. More addresses don't mean anything unless they are visible from more than one upstream point.
The internet has already reached its commercially supportable size limits, in terms of route-branches. It can still grow in terms of users, but becomes more and more vulnerable to destabilisation and control by the manipulation of single entities as it does.
IPv6 is a solution in search of a problem, and it won't be adopted, ever, unless there is either:
1) A wholesale replacement of router infrastructure coupled with a sudden burst of altruism from the corporations that own that infrastructure
2) A revolution on routing algorithms, resulting in the massive reduction of route-table storage and lookup cost - maybe quantum computing combined with a cheap and compatible framework for deploying QC devices
3) A widespread IPv6 tunneled network becomes prevalent - P2P networks are the pioneers of this type of thing, but it is unclear how such a network could perform, given the constraints imposed by the underlying IPv4 system, and given this situation it is quite likely that it will be more efficient to devise a new protocol than to do IP-within-IP.
A 'cruise missile' without an explosive payload is just a model jet with a sophisticated guidance system.
Perhaps the term 'missile' is a term that carries a negative connotation, but semantics should really not affect the fundamental issue that it is OK to experiment with aeronautics and electronics in your back yard because its your back yard and we (well, Bruce does) live in a moe-or-less free society.
Personally, i would think a more interesting goal would be to build something akin to a Predator UAV than a cruise missile, but that is just me.
John Carmack is trying to build a fucking InterContinental Ballistic Missile in his backyard, but everyone seems to love that project.
The problem is that the published.NET standards don't actually provide a specification for a standard set of tools to engineer useful software. And if you can do some things with the standardized portions of.NET, they are just as easily, more portably and more efficiently achieved using existing (non.NET) tools.
I would doubt very much whether it is possible to build.NET apps using MS's tools that do not depend on MS's proprietary implementation and extensions to it's published standards.
This is just 'Plan B to kill Java' because Plan A, trying to deliberately break Sun's proprietary standards failed so badly. So now they try to give the appearance of being 'ultra-standards-compliant' with a new and wholly redundant platform.
Having these pointless 'standards' is just a checklist item so MS product managers can construct more plausible falsehoods about.NETs supposed superiority to Java.
MS shows its commitment to 'standards' with its compatibility-breaking implementation on Kerberos, with its release of specifications for SMB/CIFS that nobody can look at without giving up any rights to work on a free implementation, by providing MFC classes under a license that specifically prohibits their use in engineering a product that competes with MS.
MS is about as interested in standards-compliance and platform neutrality as George Bush is in Solar energy and world peace.
I bought Redhat 8.0 a few weeks ago (it had been out for some time before that, of course) and have been pretty impressed with the completeness of the package and the work they have done on adding some consistency to the configuration apps bundled.
However, I can't really see what Redhat are going to put in this release to justify a +1 version upgrade.
I agree with other posters that frequent version changes will threaten the release of 'industry standard' apps on the RH Linux platform, and as such Linux in general because of the perceived volatility of the environment.
However, strong sales of 8.0 might have given Redhat the impression that consumers look favourably on 'integer' releases, when really I think 8's popularity was almost entirely due to the well-publicised 'out of the box' antialiased fonts and UI work. If it was called 7.4 it would still have been very popular for these reasons.
It would be nice to see Redhat give a clear rationale behind it's numbering scheme and clear up the confusion that obviously reigns in this area.
Slashdot Mirror
Yeah ASUS A78NX deluxe for me.
Other NForce2 boards are definitely having hard lockup problems though, including the MSI K7N2, and i've seen posts about similar problems with other NForce2 boards from different manufacturers.
Why NVidia doesnt just come clean and release the specs for the chips i'll never know, since its only going to get them bad press, I for one have spent hours of frustration on these problems.
Actually, Linux support for NForce2 is not very good - there are bugs in the chipset, or workarounds in the Windows drivers that the kernel developers are still working on.
SATA support is also pretty poor - several popular controllers either dont work, work at about half the speed in linux as they do under Windows, or won't work with software RAID-1 etc.
Have a look at recent postings to the Linux Kernel mailing list to see the nightmare that an NForce2-based board, or a SATA controller will give you under Linux.
I have both, and while I have got them to work, I had frequent hard lockups before i disabled all the ACPI/APIC stuff, my SATA controller doesn't work with software RAID-1 and 2.4 kernels gave me disk corruption and hard lockups under load.
However, The kernel developers are working on these issues, and with their help I was able to get my system up and running. I am confident that this stuff will be fully supported and stable under Linux, but unfortunately this is not the case today.
That they are going to make more money off Linux than they ever possibly could off Solaris, do a complete about face, and proclaim 'Linux is the best choice for the server as well as the desktop, and Solaris is `legacy` technology.'
I give it a till June next year.
by high-spec x86 machines, across the board.
As such, claiming that it is the worlds fastest personal computer is a complete lie.
It would be like honda claiming their new civic is 'The worlds fastest car' because in their tests against other slower cars, it beat them.
Since they have the right to sublicense it, can they put this in NSA Secure Linux as GPLed code?
Hang on a second.
SCO has a self-confessed monopoly on the ability to sell fake Linux licenses.
As such, surely it is bound by the various anti-trust laws that prevent it from doing exactly this. ('Oh no you can't buy one, only our Fortune 1000 buddies can')
If it is going to sell it's product, it has to make it available to everyone under the same terms (volume discounts might apply to Fortune 1000 companies, but thats about it).
Someone better call the US DOJ and get them to crack down on these abusers of monopoly power.
Ah, you forget the 1974 Mazda REPU (rotary engine pick up)
:)
http://www.mazdarepu.com/RoadTrack.html
Hauled bricks in a pretty convincing fashion, according to owners.
I imagine a ~180hp renesis rotary in a modern mazda pickup would make a damn sweet brick-hauler.
my RX-7 hauls heavy loads (my fat ass) and never misses a beat
Obviously, the guys who made the Kentucky Fried Movie were seriously prescient,
This guy, obviously with more than a few clues about security, is able to examine the products, right down to the source level, analyse the security provided, freely publish his findings and suggest improvements (even if all he suggests is 'scrap it', and something about skull-fucking with sound-waves.)
This is great information, and while it might not reinforce the 'open source uber alles' message, it is very useful to anyone who might be considering working on these or similar projects, as well as anyone that uses them.
Even if Mr. Gutmann says these products can't be completely fixed, at least the authors can improve them now based on his comments, and just because this guy says it can't be done, doesn't mean it is gospel.
I say a big thank you to Peter Gutmann (a fellow kiwi, alright!) for taking the time to write this and help to improve the state of open source security products.
If Microsoft loses, sure they can change their browser, but if you believe SCO etc. everyone who has such a piece of software in their possession should be liable for license charges levied by Eolas.
I'm confused.
Multiple ports are not the problem - if nothing is using those ports, there would be no traffic on them.
Blocking ports will only cripple legitimate users of those services while the malicious attackers will find other vectors for attack.
You can keep blocking ports until everything is tunnelled over port 80 and content only flows 'one way', but we already have that - its called TV/Radio broadcasting.
If anything, ISPs should filter the users logging onto their systems - e.g. if the system logging on fails security tests, or exhibits virus-carrying behaviour, then outbound access is curtailed or disabled entirely.
Crippling the internet because Microsoft can't get their shit together is the dumbest thing i've heard this week.
What, NEVER EVER Trust anyone except for Microsoft, who you grant rights to access your computer systems and do whatever they like, whenever they like, by virtue of their EULA?
You aren't actually allowed not to trust Microsoft any more, if you run their software.
Are the end-users guilty in this situation?
Microsoft offers no indemnification for Internet Explorer, as far as I am aware, so should companies using IE hire a team of lawyers and pregrammers to carefully vet every single technology used in these products to avoid liability in this kind of lawsuit?
Red Hat are in business to make money - they do this by providing paid-for distributions with full support, custom-tweaked kernels and applications, and provide a validated platform on which to run commercial apps like Oracle.
Want to put that together yourself? Go for it, nobody at Redhat is stopping you. All the stuff they integrate in their product is free, just go do it yourself.
But don't complain because you can't do it yourself and don't see why you should pay Redhat to provide you with a quality product.
Its not like you don't have a choice of vendors, or that your apps only run on a single vendor's platform.
Linux is never free - you either pay for it with money, or you pay for it with your commitment to the GPL and/or the time you invest into making it work for you.
We need people like you in the Linux community i.e. 'waah waah linux is too expensive, even when i can download it for free' like we need a frickin hole in the head.
So what does AfterStep do that WindowMaker doesn't? Its not very clear from the website.
WindowMaker CVS has antialiased fonts, and with a little bit of work on icons and tile backgrounds, my desktop looks great.
The menu editor/prefs utility is easy to use, the clip's desktop switching is perfect for what I want, theres nothing obvious about AfterStep that would make me want to switch back.
WindowMaker + GNOME apps/libs + ROX-Filer make up my ideal GUI environment.
As far as I know, theyre both written in C, as opposed to Objective-C for GNUStep, they both attempt to mimic some or all aspects of NeXTSTEP, and WindowMaker was created because the Afterstep crew was too anal about incorporating features that did not appear in the original NeXT system.
So now theyre adding features to Afterstep - dumping the whole idea behind what AfterStep was - 'Its strictly a clone of NeXTSTEP', and somehow this is a step (STEP?) forward?
So why switch from WindowMaker, when AfterStep is clearly just playing catch-up now?
That being said, there are a few rough edges in Window Maker, so perhaps competition from AfterStep will smooth them out.
If IBM were to make AIX open source - that is, make the AIX kernel, which included and depended on SysV code to function, open source, then they would indeed be in breach of their UNIX license.
AIX, as an OS kernel which includes Sys V code, is indeed a derivative work.
However, to claim that any SysV-derived UNIX cannot and has not received IP contributions from existing projects, which themselves are not licensed under the terms of the AT&T/Novell license, and as such distributing these works independently is illegal, is ludicrous.
Like, stunningly, incredibly ludicrous. In fact, I would say such a licensing agreement violates anti-trust laws, and would be ruled illegal and unenforceable, possibly rendering the entire contract void.
Free/Open/NetBSD cannot legally exist with this interpretation of 'derivative work', because they include code that was once linked with SysV code.
According to SCO, it doesn't matter if the BSD people own the copyright to every line of code, because it was once linked with SysV code, and hence is a derivative work.As such, they may not disclose it
This is quite clearly not the situation.
To claim that any IP placed in any SysV-derived UNIX strips the copyright-holder of that work's right to redistribute the original work under any license they see fit is absurd.
Are they really trying to say that because the text of the BSD license is included in SysV that nobody may distribute the text of the BSD license because it is covered by AT&T licensing agreements, and must be kept secret?
If this release was not authorized by AOL, and it isn't entirely clear that Nullsoft did not have the right to do this, then I would say that you would be forbidden from distributing the code, regardless of the fact that it carries a GPL license.
You can't take, for example, Microsoft sourcecode from the WinCE SDK, slap a GPL license on it and claim it is now covered by the GPL.
If, however, the WinCE division at MS, who presumably has full responsibility and authority to handle code releases put the WinCE code under the GPL and released it, in good faith, then that would be binding.
i.e. if the city of New York sells the Brooklyn Bridge for $1 in a legal transaction, then the new owner owns that bridge, regardless of the seemingly low price.
If it was cool to buy hundreds of thousands of acres of land off the indigenous peoples of america for some muskets and smallpox-infected blankets, this is cool too.
However, the code has been released, so while you could not distribute the code as-is, there is nothing to prevent anyone studying the code and and releasing a compatible implementation, unless it infringes copyright (contains cut n pasted sections) or patents (not sure whether P2P patents exist or who owns them). You are not doing anything illegal if you use an MP3 (regardless of its origin) of a pop song to figure out how to make a pop song of your own. The person who sold/gave you the MP3 might have a problem however, and the owner of the copyright that covers the MP3 could demand you destroy/return it upon discovery of it's improper distribution to you.
This would have to be ordered by a court to be legally enforceable, but you may be guilty of a crime by delaying the destruction/return if you do this in bad faith. i.e you know the copyright they hold is valid, yet you ignore their reasonable and legal request for its destruction/return.
AOL could, assuming their claims that Nullsoft were not authorised to release the code under GPL are true, sue anyone they can prove is distributing the code for copyright infringement.
However, if such a lawsuit was pressed, you could request that AOL prove that Nullsoft were not authorised to release software that carried AOL-owned copyright, or that they prove that Nullsoft were acting in bad faith - that is they knew the licensing terms of the software in question would violate the law or go against AOL's wishes.
If they cannot prove this, then I would guess the GPL stands, and tough cookies AOL.
But certainly the mere presence of a GPL notice does not convey legitimacy to the GPL license terms.
So, what it really comes down to is 'do you trust AOL to tell you the truth that this code was released improperly'.
If you can convince a court that you were honestly unable to determine the legitimacy of their demands (not sure you could use this as a defense against AOL, it would be watertight against SCO), then you are also OK for keeping and distributing this code despite requests from them to remove it.
After all, they can lie to you about this and not run any significant risk because of the size of their bank account, yet you have no way to verify the authenticity of their claim without a court order for them to unseal the terms of their contracts with Nullsoft, or their sworn statement in court of law.
How can we know that AOL is not lying about the fact they did legitimately GPL this software, and since there is no law against making false claims outside the realm of contract or consumer law, it seems a pretty murky area. Its not like AOL has any disincentive for lying about this.
I can stand up and say 'I am the Pope of Hudson County! Bow before my lily white ass', but failing to bow is not a crime, much to my chagrin.
It is an interesting position, and bears remarkable similarity to the whole SCO debacle.
I would have thought you could avoid the need for the HDD etc. if you could boot a miniaml linux distro off an 8MB memory card, and use the CPU power of the EE along with NAS for data storage.
Is this possible?
Vendor A (disposable puppet company supported by Vendor B) takes Linux kernel, and illegally licenses kernel under BSD license to Vendor B.
Vendor A gets sued by FSF and goes out of business.
By this logic, Vendor B still retains the BSD-licensed Linux kernel, and can build a closed-source proprietary OS around it?
Running the DB from RAM is nice, but as far as I can see this won;t require any changes to the software itself, you could just mount your DB on a RAMdisk and be done with it. Whats the big deal?
What MySQL and PostgreSQL really lack is the ability to replicate on-the-fly and to support running on clusters for *real* failover and fault tolerance.
For Postgres, this means multiple 'postmaster' processes being able to access the same database concurrently, and probably something similar for MySQL.
Being able to run a database on an OpenMOSIX cluster, for example, would make it massively scalable, and being able to run multiple independent machines with an existing HA (High Availability) monitoring system would provide a truly fault-tolerant database.
There are of course major technical difficulties involved in making databases work this way, but an Open Source DB that can compete with Oracle's 'Unbreakable' claims would be a huge shot in the arm for OSS in the business world.
One thing that nobody has ever been able to satisfactorily explain to me is how the global end-to-end routing that IPV6 offers will operate in practice.
/24 would overload most routers currently in use, and that John J. Average can advertise routes for microsoft.com and traffic goes down a black hole.
/16. Since the only people who can realistically maintain a /16 are ISPs, mega-corps or state institutions, this basically takes the 'dream of a global network in which everyone is connected', breaks its legs and ties it directly to the financial interests of it's largest players.
IPV4 can't be routed in this manner, because of router resource limits and the unreliability of route information - e.g. keeping route information for every individual
CIDR (Classless Internet Domain Routing) alleviates both these problems to some degree by removing the ability to effectively route any address block smaller than a
Some would say (and I would largely agree) that the commercial benefits to infrastructure providers of offering routing to the 'smaller' nodes on the internet are non-existent or negative. So it just won't happen.
CIDR + IpV6 gives us nothing we don't already have. It simply offers more addresses behind each NAT machine - and really, nobody I have ever heard of is NATing more machines than will fill the 192.168.0.0/16 + 10.10.0.0/16 blocks.
I could be wrong, maybe new routing protocols have been developed to handle IPv6 routing out to network's edge, and maybe these protocols have been adopted as industry standards and are present in the firmware of all IPv6-capable hardware from the vendors that matter.(I don't imagine it will be a problem to deploy this stuff to 'software' routers)
I, along with many others, desperately want to be able to take advantage of the internet's fault-tolerance, but are prevented from doing so by CIDR etc. More addresses don't mean anything unless they are visible from more than one upstream point.
The internet has already reached its commercially supportable size limits, in terms of route-branches. It can still grow in terms of users, but becomes more and more vulnerable to destabilisation and control by the manipulation of single entities as it does.
IPv6 is a solution in search of a problem, and it won't be adopted, ever, unless there is either:
1) A wholesale replacement of router infrastructure coupled with a sudden burst of altruism from the corporations that own that infrastructure
2) A revolution on routing algorithms, resulting in the massive reduction of route-table storage and lookup cost - maybe quantum computing combined with a cheap and compatible framework for deploying QC devices
3) A widespread IPv6 tunneled network becomes prevalent - P2P networks are the pioneers of this type of thing, but it is unclear how such a network could perform, given the constraints imposed by the underlying IPv4 system, and given this situation it is quite likely that it will be more efficient to devise a new protocol than to do IP-within-IP.
Which do you think will happen first?
A 'cruise missile' without an explosive payload is just a model jet with a sophisticated guidance system.
Perhaps the term 'missile' is a term that carries a negative connotation, but semantics should really not affect the fundamental issue that it is OK to experiment with aeronautics and electronics in your back yard because its your back yard and we (well, Bruce does) live in a moe-or-less free society.
Personally, i would think a more interesting goal would be to build something akin to a Predator UAV than a cruise missile, but that is just me.
John Carmack is trying to build a fucking InterContinental Ballistic Missile in his backyard, but everyone seems to love that project.
http://hammer.prohosting.com/~ikekrull/art/07.gif
The problem is that the published .NET standards don't actually provide a specification for a standard set of tools to engineer useful software. And if you can do some things with the standardized portions of .NET, they are just as easily, more portably and more efficiently achieved using existing (non .NET) tools.
.NET apps using MS's tools that do not depend on MS's proprietary implementation and extensions to it's published standards.
.NETs supposed superiority to Java.
I would doubt very much whether it is possible to build
This is just 'Plan B to kill Java' because Plan A, trying to deliberately break Sun's proprietary standards failed so badly. So now they try to give the appearance of being 'ultra-standards-compliant' with a new and wholly redundant platform.
Having these pointless 'standards' is just a checklist item so MS product managers can construct more plausible falsehoods about
MS shows its commitment to 'standards' with its compatibility-breaking implementation on Kerberos, with its release of specifications for SMB/CIFS that nobody can look at without giving up any rights to work on a free implementation, by providing MFC classes under a license that specifically prohibits their use in engineering a product that competes with MS.
MS is about as interested in standards-compliance and platform neutrality as George Bush is in Solar energy and world peace.
I bought Redhat 8.0 a few weeks ago (it had been out for some time before that, of course) and have been pretty impressed with the completeness of the package and the work they have done on adding some consistency to the configuration apps bundled.
However, I can't really see what Redhat are going to put in this release to justify a +1 version upgrade.
I agree with other posters that frequent version changes will threaten the release of 'industry standard' apps on the RH Linux platform, and as such Linux in general because of the perceived volatility of the environment.
However, strong sales of 8.0 might have given Redhat the impression that consumers look favourably on 'integer' releases, when really I think 8's popularity was almost entirely due to the well-publicised 'out of the box' antialiased fonts and UI work. If it was called 7.4 it would still have been very popular for these reasons.
It would be nice to see Redhat give a clear rationale behind it's numbering scheme and clear up the confusion that obviously reigns in this area.