[This question is about Corporate ownership
of the government and the impact of global
Corporations on society. I've tried to phrase
it to see if the candidate has an inkling of
what the problem is, rather than get a vague
response like 'I support campaign finance
reform.']
What is your reaction to the protests that have
taken place at the WTO meeting in Seattle, and
at the Republican and Democratic national
conventions. What do you think the protests
are against? How would you address the concerns
of the protestors? How would your solution to
these problems have any effect?
Actually I used JFS for a couple of years (around '95) and quite liked it. The filesystem
is dynamically expandable so you can change your
partitioning and add space to a JFS filesystem
without having to boot single user, backup,
delete, create, restore. Of course you have to
have a dynamic partitioning system to do this,
but that is why LVM is so useful.
And really I can't imagine any filesystem that
could be absolute proof against anecdotal failures like the one you experienced; there are after all
head crashes and the like.
That said, I do think that Linux is much easier
to administrate than AIX (for SunOS junkies
Linux is really easy to pick up, and the automation utilities don't mess with the configuration files the same way that OSF/1
decided to do.)
But even just for personal
use I think that dynamically expanding the
filesystem would be worth the installation
effort, and for people living in the San
Francisco bay area where we've been having
rolling blackouts over the summer, avoiding
fsck's every hot day is worth a minor hit
on directory writes.
The article is so thick with politically
charged words that it is almost impossible
to tell what Sun actually did so that we can
decide for ourselves whether what they did was
reasonable or not.
Who would know how to file a friend of the
court brief to correct some of this BS? Would
anyone else be interested in signing on to an
open letter to the court? (And would that even
be meaningful?)
Patent controls - the algorithms used by Vorbis are not patented and were therefor never researched commercially.
CPU requirements - the Vorbis algorithms were
considered too computationally expensive at the time MP3 was created for companies to use.
Either of these arguments could have equally well have been applied to Elliptical Curve Cryptography when compared with the RSA algorithm, and ECC is only now being 'discovered' again long
after its initial description.
More importantly McDonalds walked into court and
said approximately: 'We don't care about the little old lady. If we didn't sell our coffee hotter than any competitor, we wouldn't sell as much coffee, so the lady can get stuffed.'
So the jury fined them one day's take for coffee
sales.
Which was later reduced in appeal to just medical bills plus lawyers fees.
ECC may not be as simple as RSA, but it source
code is available. We used ECC for protecting
E-speak (which is GPL'd software -- fully open
including the crypto) and published it under the
new rules for public domain, and public domain-
like software.
I spent a couple of hours mulling over the ideas in the article, but after all I'm not so sure the author was right.
His responses to cheating were almost entirely technological, but the problem he describes is sociological. At best the solutions written about are patches -- go steal some other car; this one has a car alarm.
Social responsibility comes from identity and accountability. To make a truly enjoyable game I think there is little choice but to develop a sense of community among the game players. Establish identity at the outset, make it expensive to create new identities, and track notoriety, as reported by other players who have reviewed the game log after it is complete.
I don't think it should be commented on by slashdot either!
The stock market is a lot more important to the future of open source software than seems obvious at first glance.
For the past four or five years small e-form companies have had a huge advantage over large corporations like IBM or HP in that the (potentially) very high stock market valuation of these companies has allowed infrastructure to spring up from nothing.
All of the midsized companies signing on to use Linux couldn't care less about free software or ideals of how information should be handled. They are interested in Linux because there is suddenly a huge amount of support for it. Hundreds of e-businesses like Linuxcare, or Redhat to name the largest, have filled in the market with software solutions customized to fit their needs.
But the viability of that market mechanism is predicated on the e-form funding model. Without that these hundreds of companies couldn't have been funded, the products wouldn't exist, the mid-level companies wouldn't see sufficient support to justify choosing Linux.
So if the market gets bad enough, then yes, I believe that everyone in the Open Source community will care. Linus and many other developers would happily continue writing their software to satisfy themselves, but there is a strong distinction between hobby-ists toolset and product.
It isn't a question of what the GPL requires, but of what the law requires. Unsigned contracts have never been found to be binding on either party.
(Well there are a couple of odd exceptions; such as using software that was licensed only for non-server use on a server, when a licensed version for servers was also available. Sorry I don't have the citation handy.)
The tail end of the article states: Eugene Volokh, a law professor at UCLA, said that Mattel might be able to argue that the GPL is invalid because users don't pay for the free software. "Nonexclusive licenses given for free are generally revocable, even if they purport to be irrevocable," Volokh said. "Even if the GPL license in cphack is treated as signed and is covered by 205(e), it might still be revocable by Mattel as the new owners of the cphack copyright." "It is unfortunately not quite as solid a case for the good guys as the GNU license theory would have at first led us to believe," he said.
This isn't strictly speaking true. Nothing in the GPL prevents the sale for money of the licensed sources, and in fact the products so licensed are often sold.
Hopefully the FSF can show a legitimate financial claim on software that has been provided to them under signed license.
Yeah, there is the CP/M derived chain and the UNIX derived chain. I suspect that the wider compatibility of any competing OS to one of these two comes from the benefit of getting a part of your installed base for free.
I would argue though that NT isn't really derivative of CP/M; it more represents its own chain. Sub-operating systems like Java might eventually grow into their own systems.
There are also other OS's out side of the x86 arena, such as the myriad DOSes created for personal computers like Mac Finder, AmigaDOS, the Apple Basic, TRSDOS, and so forth, but since these were all single source OS's championed not by an industry but by a single company, they were pretty much doomed.
Perhaps the failure of VMS, MVS, DGOS, and DSS to gain a broad acceptance was for the same reasons. -kls -- a broad acceptance is
Presumably because the offer was reasonable. You shouldn't expect everyone to be willing to martyr themselves for the cause of free-speech.
Assuming that the settlement resulted in no admission of guilt, and no lawyers fees; just handing over the code (with a affidavit of reassigned copyrights), I think that this would be a pretty attractive exit plan.
Particularly if you were say a minor, and your father's house might be taken away if you lost, or if you didn't think that you could afford attorney's fees to compete with a wealthy multi-national corporation.
The justice system isn't always about winning -- sometimes it is just reasonably prudent to step out before the discussion gets too heated. -kls
It is always difficult to tell whom to believe in cases like this. The article makes coke.ch sound perfectly reasonable, but reading through the arbitration logs at ICANN you get to see the other side of the story as well.
--
David Brin: The Transparent Society
on
Database Nation
·
· Score: 1
David Brin also has a very good book out on loss of privacy and what it means for the future.
Brin as always is very optimistic, and makes a good argument that privacy, such as it is, is a relatively recent phenomenon: that historically everyone knew everyone else's business anyway.
Recommended if you would like a more positive view of the effects of the electronic age on privacy, and some interesting insights into the downsides of trying to preserve privacy in the face of technology.
A couple of months ago I suggested a poll on the Linux Game Tome asking about the buying habits of linux gamers. Although the results are admittedly non-scientific, I thought the results were quite interesting.
Of the respondents almost two thirds had never purchased a commercial game for Linux; and their reasons for this broke down as follows:
None, Software should be free (27%)
None, the games available don't interest me (15%)
None, the games available are too expensive (12%)
None, the games available come out too late (8%)
What do you think that Loki can/has to do to remove these barriers? And if these challenges are unaddressed, what would you predict is the future of Linux gaming?
That isn't a joke. At microsoft, interoperability *does* mean between MS products. Kind of like at Sun where open means open to anyone who isn't competing with Sun.
Seriously. I've got several friends who used to work at or with microsoft engineers, and that *is* how they define interoperability.
I think that the proposal you make creates a fundamental conflict of interest; and isn't necessary anyway. At Hewlett-Packard we recently had a similar problem come up when we explained to our management why we wanted to open source e-speak, a piece of infrastructure software for creating free (as in speech) services markets.
The conflict of interest is between asking people to accept that you are using the GPL to show the world that the software you are releasing will create a level playing field while at the same time reserving a priveleged place for yourself. At any point you can argue that your software isn't making enough money because you didn't reserve enough of a special position for yourself, leading you to close the source to more of the product, and alienating the open community that is trying to grow around it.
The first step IMHO is to consider the development costs of the software a writeoff -- and start considering instead how you can make money from the service surrounding the software instead of the software itself.
The key argument that I make to my management is that even if the source is open, that doesn't make it a product. There are some people who are willing to download and work with software without any support and packaging, but those are precious few and their attention is divided among hundreds of different pieces of software that they could possibly be working on.
But by combining that source code with a good test suite (to minimize package bugs, and control backward compatibility), a product release stream (to make it easily installable, include GUI management front-ends, precompiled with a top quality compiler, etc.), a call-in support line, and company backed performance guarantees, you have added something that the source by itself doesn't have.
It is an ongoing area of controversy, but I really think that after releasing a set of code under GPL you should never look back, but push forward to really gather in the community you are trying to create and avoid sending mixed messages like: you can use this for free, but please don't compete with me.
That having been said, if you really want to do this I know that there has been at least one successful counter-example to my opinion; take a look at the ghostscript site and see if you can cope with that model.
Hmmm... E-Speak was designed just for this purpose. I work on the product, so I'm not exactly unbiased, but it is:
Free software
At its core designed as an XML middleman
With lots of features (security, manageability, dynamic location, distribution, intermediation (for translation)) that XML needs and I would argue that a web server only inadequately satisfies
I would contend that e-speak doesn't rely on any new programming paradigm, but does allow one. The mental shift is from applications to services, and that is what the author of the tutorial was trying to explain.
The actual technologies involved are ones that everyone would recognize. You can currently write your services using distributed network objects, and document exchange models of programming.
The network objects support uses an API similar to Java/RMI with support for the Java standard RMI in work. Such changes as were made are the minimum necessary to create the platform we were trying to expose; so that is what is available now. Mapping those into Java/RMI will allow Jini and Java Beans to interoperate pretty seamlessly in a services environment.
The document exchange model support is through standard HTTP methods (POST & GET) and their replies; and currently uses cookies to represent conversation sessions. The documents exchanged are standard XML documents, which comply with an e-speak dtd when requesting services of the platform, but can be any dtd at all when sending messages from a miscellaneous client to a miscellaneous service.
My desire with e-speak is not to create a new competing standard for communications, but to unify all of the competing standards into an infrastructure where they can all communicate with one another. It is a great advantage of being open source; we don't have to own the API.
Fundamentally HP in releasing e-speak is trying to foster the creation of an open marketplace for electronic services. The web is also an open market, but it was really designed to be a market for mostly free information.
In the e-speak platform you will find the features that we believe will make providing a service on the internet feasible without spending two years and a couple of million dollars rebuilding the necessary infrastructure from scratch. In other words we are trying to remove the barriers to entry for services that the web removed for publishing.
The basic features of the e-speak platform are distribution, language independence (but the most of the current code is in Java), security, manageability, dynamic resource location and intermediation (which allows services to be dynamically configured into composite services).
If that seems a bit too opaque, I'm happy to discuss what each of these means and why they are important to a services infrastructure on the e-speak mailing lists.
Slashdot Mirror
[This question is about Corporate ownership of the government and the impact of global Corporations on society. I've tried to phrase it to see if the candidate has an inkling of what the problem is, rather than get a vague response like 'I support campaign finance reform.']
What is your reaction to the protests that have taken place at the WTO meeting in Seattle, and at the Republican and Democratic national conventions. What do you think the protests are against? How would you address the concerns of the protestors? How would your solution to these problems have any effect?
Check the web site FAQ for more info.
Actually I used JFS for a couple of years (around '95) and quite liked it. The filesystem is dynamically expandable so you can change your partitioning and add space to a JFS filesystem without having to boot single user, backup, delete, create, restore. Of course you have to have a dynamic partitioning system to do this, but that is why LVM is so useful.
And really I can't imagine any filesystem that could be absolute proof against anecdotal failures like the one you experienced; there are after all head crashes and the like.
That said, I do think that Linux is much easier to administrate than AIX (for SunOS junkies Linux is really easy to pick up, and the automation utilities don't mess with the configuration files the same way that OSF/1 decided to do.)
But even just for personal use I think that dynamically expanding the filesystem would be worth the installation effort, and for people living in the San Francisco bay area where we've been having rolling blackouts over the summer, avoiding fsck's every hot day is worth a minor hit on directory writes.
Wow!
The article is so thick with politically charged words that it is almost impossible to tell what Sun actually did so that we can decide for ourselves whether what they did was reasonable or not.
So much text. So little content.
Who would know how to file a friend of the court brief to correct some of this BS? Would anyone else be interested in signing on to an open letter to the court? (And would that even be meaningful?)
The two main reasons that come to mind are:
Either of these arguments could have equally well have been applied to Elliptical Curve Cryptography when compared with the RSA algorithm, and ECC is only now being 'discovered' again long after its initial description.
It is kind of auspicious listing this article right after the one about universities being
bought out for research.
More importantly McDonalds walked into court and said approximately: 'We don't care about the little old lady. If we didn't sell our coffee hotter than any competitor, we wouldn't sell as much coffee, so the lady can get stuffed.'
So the jury fined them one day's take for coffee sales.
Which was later reduced in appeal to just medical bills plus lawyers fees.
Personally I agree with the jury.
-kls
ECC may not be as simple as RSA, but it source
code is available. We used ECC for protecting
E-speak (which is GPL'd software -- fully open
including the crypto) and published it under the
new rules for public domain, and public domain-
like software.
-kls
I spent a couple of hours mulling over the ideas in the article, but after all I'm not so sure the author was right.
His responses to cheating were almost entirely technological, but the problem he describes is sociological. At best the solutions written about are patches -- go steal some other car; this one has a car alarm.
Social responsibility comes from identity and accountability. To make a truly enjoyable game I think there is little choice but to develop a sense of community among the game players. Establish identity at the outset, make it expensive to create new identities, and track notoriety, as reported by other players who have reviewed the game log after it is complete.
The stock market is a lot more important to the future of open source software than seems obvious at first glance.
For the past four or five years small e-form companies have had a huge advantage over large corporations like IBM or HP in that the (potentially) very high stock market valuation of these companies has allowed infrastructure to spring up from nothing.
All of the midsized companies signing on to use Linux couldn't care less about free software or ideals of how information should be handled. They are interested in Linux because there is suddenly a huge amount of support for it. Hundreds of e-businesses like Linuxcare, or Redhat to name the largest, have filled in the market with software solutions customized to fit their needs.
But the viability of that market mechanism is predicated on the e-form funding model. Without that these hundreds of companies couldn't have been funded, the products wouldn't exist, the mid-level companies wouldn't see sufficient support to justify choosing Linux.
So if the market gets bad enough, then yes, I believe that everyone in the Open Source community will care. Linus and many other developers would happily continue writing their software to satisfy themselves, but there is a strong distinction between hobby-ists toolset and product.
-kls
It isn't a question of what the GPL requires, but of what the law requires. Unsigned contracts have never been found to be binding on either party.
(Well there are a couple of odd exceptions; such as using software that was licensed only for non-server use on a server, when a licensed version for servers was also available. Sorry I don't have the citation handy.)
-kls
--
This isn't strictly speaking true. Nothing in the GPL prevents the sale for money of the licensed sources, and in fact the products so licensed are often sold.
Hopefully the FSF can show a legitimate financial claim on software that has been provided to them under signed license.
-kls
--
I would argue though that NT isn't really derivative of CP/M; it more represents its own chain. Sub-operating systems like Java might eventually grow into their own systems.
There are also other OS's out side of the x86 arena, such as the myriad DOSes created for personal computers like Mac Finder, AmigaDOS, the Apple Basic, TRSDOS, and so forth, but since these were all single source OS's championed not by an industry but by a single company, they were pretty much doomed.
Perhaps the failure of VMS, MVS, DGOS, and DSS to gain a broad acceptance was for the same reasons.
-kls
-- a broad acceptance is
Presumably because the offer was reasonable. You shouldn't expect everyone to be willing to martyr themselves for the cause of free-speech.
Assuming that the settlement resulted in no admission of guilt, and no lawyers fees; just handing over the code (with a affidavit of reassigned copyrights), I think that this would be a pretty attractive exit plan.
Particularly if you were say a minor, and your father's house might be taken away if you lost, or if you didn't think that you could afford attorney's fees to compete with a wealthy multi-national corporation.
The justice system isn't always about winning -- sometimes it is just reasonably prudent to step out before the discussion gets too heated.
-kls
Not to put too fine a point on it, but I think you are just being argumentative. kls --
It is always difficult to tell whom to believe in cases like this. The article makes coke.ch sound perfectly reasonable, but reading through the arbitration logs at ICANN you get to see the other side of the story as well.
--Brin as always is very optimistic, and makes a good argument that privacy, such as it is, is a relatively recent phenomenon: that historically everyone knew everyone else's business anyway.
Recommended if you would like a more positive view of the effects of the electronic age on privacy, and some interesting insights into the downsides of trying to preserve privacy in the face of technology.
-kls
--
A couple of months ago I suggested a poll on the Linux Game Tome asking about the buying habits of linux gamers. Although the results are admittedly non-scientific, I thought the results were quite interesting.
Of the respondents almost two thirds had never purchased a commercial game for Linux; and their reasons for this broke down as follows:
What do you think that Loki can/has to do to remove these barriers? And if these challenges are unaddressed, what would you predict is the future of Linux gaming?
--That isn't a joke. At microsoft, interoperability *does* mean between MS products. Kind of like at Sun where open means open to anyone who isn't competing with Sun.
Seriously. I've got several friends who used to work at or with microsoft engineers, and that *is* how they define interoperability.
I think that the proposal you make creates a fundamental conflict of interest; and isn't necessary anyway. At Hewlett-Packard we recently had a similar problem come up when we explained to our management why we wanted to open source e-speak, a piece of infrastructure software for creating free (as in speech) services markets.
The conflict of interest is between asking people to accept that you are using the GPL to show the world that the software you are releasing will create a level playing field while at the same time reserving a priveleged place for yourself. At any point you can argue that your software isn't making enough money because you didn't reserve enough of a special position for yourself, leading you to close the source to more of the product, and alienating the open community that is trying to grow around it.
The first step IMHO is to consider the development costs of the software a writeoff -- and start considering instead how you can make money from the service surrounding the software instead of the software itself.
The key argument that I make to my management is that even if the source is open, that doesn't make it a product. There are some people who are willing to download and work with software without any support and packaging, but those are precious few and their attention is divided among hundreds of different pieces of software that they could possibly be working on.
But by combining that source code with a good test suite (to minimize package bugs, and control backward compatibility), a product release stream (to make it easily installable, include GUI management front-ends, precompiled with a top quality compiler, etc.), a call-in support line, and company backed performance guarantees, you have added something that the source by itself doesn't have.
It is an ongoing area of controversy, but I really think that after releasing a set of code under GPL you should never look back, but push forward to really gather in the community you are trying to create and avoid sending mixed messages like: you can use this for free, but please don't compete with me.
That having been said, if you really want to do this I know that there has been at least one successful counter-example to my opinion; take a look at the ghostscript site and see if you can cope with that model.
Hmmm... E-Speak was designed just for this purpose. I work on the product, so I'm not exactly unbiased, but it is:
Hope to see you there.
-kls
Distributed network objects and remote procedure calls isn't the only programming model that e-speak supports.
-kls
The actual technologies involved are ones that everyone would recognize. You can currently write your services using distributed network objects, and document exchange models of programming.
The network objects support uses an API similar to Java/RMI with support for the Java standard RMI in work. Such changes as were made are the minimum necessary to create the platform we were trying to expose; so that is what is available now. Mapping those into Java/RMI will allow Jini and Java Beans to interoperate pretty seamlessly in a services environment.
The document exchange model support is through standard HTTP methods (POST & GET) and their replies; and currently uses cookies to represent conversation sessions. The documents exchanged are standard XML documents, which comply with an e-speak dtd when requesting services of the platform, but can be any dtd at all when sending messages from a miscellaneous client to a miscellaneous service.
My desire with e-speak is not to create a new competing standard for communications, but to unify all of the competing standards into an infrastructure where they can all communicate with one another. It is a great advantage of being open source; we don't have to own the API.
-kls
not speaking for HP
Fundamentally HP in releasing e-speak is trying to foster the creation of an open marketplace for electronic services. The web is also an open market, but it was really designed to be a market for mostly free information.
In the e-speak platform you will find the features that we believe will make providing a service on the internet feasible without spending two years and a couple of million dollars rebuilding the necessary infrastructure from scratch. In other words we are trying to remove the barriers to entry for services that the web removed for publishing.
The basic features of the e-speak platform are distribution, language independence (but the most of the current code is in Java), security, manageability, dynamic resource location and intermediation (which allows services to be dynamically configured into composite services).
If that seems a bit too opaque, I'm happy to discuss what each of these means and why they are important to a services infrastructure on the e-speak mailing lists.
-kls (who doesn't officially speak for HP)