Where was I when Java 5 came out? I'd been waiting three years for it and complaining loudly that a fully open process would have produced it much earlier. There isn't much in Java 5 that isn't derivative of other work. That's not a criticism, it just follows along my point that Java is now synthetic rather than innovative- that's what Sun and the other consitituencies WANT from Java. Generics are templates. The enum implementation permits a degree of clarity and conciseness that shouldn't have been lacking in the first place. Hell, even the fancy wait-free concurrency that everyone raves about is nothing more than pretty words over a coding style (minimal sync-sets) that has been around for a dozen years.
Because it doesn't need to. Java is now so big and so corporate that stability is far more valuable to its constituents than innovation. It will be the development system of choice for corporate development for another decade, providing jobs for all the new and soon-to-be CS graduates who aren't taught anything else. Yet Java is already completely ossified.
This started out as practically a greenfield project within MS with a brand new team at least five years ago. It's being done by incredibly competent people, and they have done a huge amount of work on interoperability issues- that whole raft of WS standards represents solutions to a whole range of issues that no else is really confronting. And I'm not saying they've solved them in an ideal way but as long as no one else puts anything out there that is less proprietary, they will jump out in front here. Remember how they eventually "got" the Internet? Those of us who love F/OSS and Linux need to be less dismissive and more frightened. MS is fragmented and balkanized internally but there are pockets of real capability. Web Services have not achieved anywhere near the level of adoption they could/should have by now (to all you trolls: the few dozen desultory SOAP projects at your company prove my point, not disprove it). And that's because of lack of "security," which boils down to lack of widely supported standards. We gotta be more proactive about this, and not make the same mistakes we are making in regard to Avalon.
Daimaou signs a (reasonably) standard IP-assignment contract, so everything he wrote while working for the company belongs to the company. That's completely straightforward. Our hero is out of the picture at this point. So now the company owns some code which is an extension of a GPL-ed code base. They're free to do anything they want with it except distribute it to others in any form. If they do that, then they must comply with GPL's source distribution requirements. That's also straightforward. Now the company wants to patent something that this code embodies. (The article doesn't make clear whether the patentable content is part and parcel of the GPL-ed codebase, but I'll assume it is.) If it's an American company, they're free to try to do this, and given how incompetent the USPTO is, they stand a pretty good chance of getting the patent. If the patent examiners are awake, then they'll notice the existence of substantial prior art as embodied in the GPL code base, but they're usually not awake. (Disclosure: I am a patent-holding inventor of certain firewall technology.)
Skype doesn't play nice in the sandbox with the rest of the world. I hope you're not thinking Google would buy Skype and try to suppress SIP... horrors!
What does Google do better than anyone else in the world? They can efficiently make observations and draw inferences across all of the world's electronic information. It's by far the biggest in-cache working set ever.
So how can you apply their unique knowledge to produce a VoIP service that is qualitatively better than can be offered by anyone else in the world? If there is no concrete answer to that question, then they shouldn't offer the service- it would be a bad business decision. Do any of you PhDs out there in slashdot-land know the answer?
Shiva the destroyer of worlds reminded me of another point: Try reading your code out loud, as silly as that sounds. Not only will that immediately cure you of identifiers-without-vowels and the evil curse of hungarian notation, but you'll almost instantly know whether it makes logical sense. Just try it before you flame, especially if you're a young guy who already knows everything;-). My C/C++ and Java almost always compiles the first time. You wipe out a whole class of stupid errors with this technique. Reading code out loud is the next best thing to reading it to someone else.
Shiva: were you in the Manhattan Project or something?:-)
A gentleman of my acquaintance is a phenomenally successful retired CEO of a Fortune 500 company. He's now a visiting professor at a very prestigious B-school, a household name (although I won't name it). It's not too much to say that he was shocked at the lack of ethical sense among today's B-school students. It's probably no shock to slashdotters that PHBs-in-training are only in it for the money, but then we shouldn't be surprised when garbage like this happens either.
To think of code as literature is actually a non-obvious and striking idea. The essence of literature is communication. I've been writing C and C++ for years, constantly fine-tuning and refining code and documentation, line by line, paragraph by paragraph. And how apropos to realize that the point is to express the thought so clearly and concisely that any reader (including yourself three weeks from now) will grasp it as close to immediately as possible. That means you have to think clearly in order to code clearly. This matters more than almost anything else. Too much documentation is just as bad as too little, since it obscures the intent of the code. How much is just enough? It takes years of thoughtful practice to know that. Everyone is posting their favorite rules about coding standards. Most young programmers apply standards as rigid as they are incomplete. I have only two rules: 1) Call everything by its right name. Functions are verbs, variables are nouns. The name you give a object should express what it is or does, no more and no less, and be readable in English (or whatever your spoken language is). If you add something to a function that doesn't fit the function's name, it either belongs in a different function or else you misnamed the function in the first place. This takes constant review of your code, from a literary point of view. 2) Rules are made to be broken. All great writers unhesitatingly break the rules in order to communicate more clearly.
Languages are for communicating, and computer languages are no different. Keep that in mind at all times, and you'll become a wizard.
1) It doesn't work well enough. >>>Meaning what? The combination of cost, features, robustness and stability are not superior to the analogous aggregate for Linux, in low-end server applications. In short, there's not enough reason to abandon Linux for Sun 10. The calculation goes the other way in Sun's core market, which is high-end enterprise application servers.
2) It takes away from Sun's core business, which is and always been selling Sparc-based hardware >>>Sun's core business has always been selling networked Unix systems and the services to go along with them. I think that your statement represents Sun's core marketing spin rather than their core business;-). But even if they believed it themselves (which they may), Sun has become a seriously risk-averse culture in recent years and I don't think they would lightly de-emphasize the decades of brand strength associated with Sparc hardware.
3) It doesn't provide a compelling alternative to Linux in the application space which is targeted by inexpensive x86-based hardware. >>>It doesn't have the device support that Linux does, but it runs on a huge amount of x86 hardware. I think the performance, stability, and features like SMF, Zones, and dtrace make it a compelling alternative. Remains to be seen. dtrace is interesting in low-end applications, the other two less so.
4)I've been trying to do Solaris on x86 for 10 years now, ever since Solaris version 2.3. It just isn't stable. You're making my point with your statement about Sparc's better hardware. Plus, Solaris is seriously optimized for the Sparc architecture, and always has been.
You make a number of very interesting points. The key insight is that hard, physical resources need to be provisioned for servers and bandwidth. The only large-scale counterexamples I can think of are BitTorrent, Gnutella and SETI. Can this kind of community provisioning work for a free/open search system? I'd sure like to know, what does everyone think?
Your objections about the business model (personal ads) are less germane, I think. Those issues go away if you don't need to run the service to make money. Why do we use Yahoo or Google? For the search service. Why does Yahoo or Google provide the search service? For the purpose of selling ads. When I really want to see ads, they're not hard to find:-)
Let's keep sight of the original point, which is that Yahoo is the latest company looking to capture value from the labor of the community. That means from you and from me. That's the part that I think should be resisted. I don't know if an open Yahoo alternative makes any sense. But I do know that smart programmers willing to give free labor to Yahoo need to get a life.
I have an interesting technical problem to solve... would you like to work on it? It doesn't pay anything, but I'll make a lot of money selling your work. Interested?:-)
Running Sun 10 on an Epia: that's a very interesting science project. But remember that Sun is a multi-billion dollar business. They would need to evaluate the market for Sun on mini-ITX platforms and conclude that it can add at least a few hundred million per year to their top line before it gets interesting. Also, their core market is enterprise IT, and it's far from clear that this market has a great interest in mini-ITX platforms, unless you're proposing to run Sun 10 on desktop-replacement laptops.
Regarding Solaris on x86: 1) It doesn't work well enough. Certainly not well enough to displace Linux in the core market for x86-based servers. 2) It takes away from Sun's core business, which is and always been selling Sparc-based hardware and peripherals to enterprise IT. 3) It doesn't provide a compelling alternative to Linux in the application space which is targeted by inexpensive x86-based hardware. If you want to, you could try making the case that you could run Sun on all your machines from top to bottom and avoid the need to hire Linux admins, but I'd need to see the numbers. 4) I've been trying to do Solaris on x86 for 10 years now, ever since Solaris version 2.3. It just isn't stable.
I have no quarrel with the process or the license. (And I didn't say anything at all about GPL, so don't put words in my mouth.) Sun can release all the source code they want. My point is that it's not going to bring them the business benefits they are hoping for, because the underlying economics are broken.
Why on earth would any intelligent, well-motivated and talented hacker want to work for Yahoo/Google/IBM/Sun/whoever WITHOUT getting a salary from them? All of these companies that are talking about tapping the capabilities and intelligence of the "community" must think we have no intelligence at all! It's the same thing with open-sourcing Solaris. Anybody who is talented and enthusiastic enough to make serious contributions to a major search engine or operating system should be doing it to benefit the whole community, not just to make some major corporation even richer. We already know about the open alternatives to Solaris. Where is the open and free alternative to Yahoo? I'll contribute time and money to it!
Solaris is a very very fine piece of technology, and Sun's many enterprise customers will be installing upgrades and consuming services for many years to come. However, Sun is not going to win any new business with this release- all the growth is above them and below them. They know they have to do something, which is why they "open-sourced" 10, but this is nothing more than a lame attempt to get some of the benefits of community development. How many of you Linux/BSD hackers out there are all fired up to start contributing features and bug fixes to Scott and Jon? I thought not. And in regard to Solaris on x86, it's completely senseless, for more than one reason.
Sun's total market cap would fit more than three times into Microsoft's bank account, and that has been true for over three years now. The problem is, no one has a need to acquire Sun, either as an accretive play (no growth) or to kill them. They're going to twist in the wind for at least another decade, but they stopped being interesting long ago.
We offered to fund a portion of the OpenBSD project's budget as a corporate sponsorship. We were thinking about putting in perhaps a million dollars a year. They responded cautiously until we asked how much they might need. The emails we got back are not printable on a family website:-). This was back in 2003, several months before DARPA pulled much of their grant money. When we started hearing from fallen-away project members over the next few months, we realized we were lucky not to have gotten involved. Anyway, no knock against Theo or his achievements. He deserves his recognition, as I said in my original post. He's running his project like a sandbox, however, and that makes it hard to depend on.
Theo deserves a lot of recognition for his technical achievements and his commitment to freedom. Getting this award proves that you can blow off everyone in the world except your personal fanboys and still be a success.
My company based a commercial product on O-BSD, then converted to Linux when it became clear that Theo doesn't know how to anchor a diverse community. We even tried to fund his project but never got past being personally abused.
Not 100% sure but I think the MQ core is written in straight C. It still costs a fortune and is absolutely vicious to to keep running. On Wall St, they employ armies of people to do just that. An open-source alternative is apropos.
I'm still laughing at the Stereo Review cartoon ca. 1975 with a radio announcer introducing a performance of a Stockhausen piece, performed on the original transistors, resistors and capacitors.
TIB's pub/sub model isn't appropriate for many applications (which is why they bought and killed Talarian- remember them?). People use MQ for the robust, guaranteed delivery and asynchrony. Plus, TIB is wickedly expensive, and a bear to admin. An open source alternative is a fabulous idea.
My company built and sold a message-oriented messaging product that merged the pub/sub and transactional models several years back, and it included guaranteed delivery. And yes, there was a lot of discussion about open sourcing the code base. The system is still in daily production in a bunch of Wall St. firms. What these firms all wanted as a key requirement was clean integration with C/C++, and extreme performance. We had to handle thousands of messages (ranging into megabytes each) per second, with full guaranteed delivery, over IP networks and through firewalls. Oh, and none of the flakiness and admin problems of MQSeries. None of the available JMS products could handle all these requirements.
It's a well-considered and interesting article, but misses the point. Are the four freedoms enough to make people use alternatives to proprietary software? In themselves, no. The freedoms are one of the key quality dimensions that determine user choice, but not the only ones. Completeness, stability, suitability for the purpose, and ease of use matter as much or more. Oh yeah, a minimal money-cost to acquire the right to use a given piece of software is also one of these dimensions. The choice of which software to use is a strictly "economic" one, in which all of these dimensions will be added together by the decision-maker, and an optimal choice will be made. So what does it mean to be a "true believer" in F/OSS (as I am)? It means you believe that over time, the quality, stability, and suitability of fully-free software will rise to match and surpass the commercial alternatives.
Fair enough. Wiki will die a quick and unlamented death if it becomes constrained by expert committees. But the basic problem is still: how can we assess the quality of the information? Having observed/. for a long time, I'm skeptical that community-generated mod points will result in anything but the validation of the prevailing (and evanescent) point of view. It's a market after all, and markets exhibit the same behavior- it's a flaw in the "perfect knowledge" theory. What I guess I'm reaching for is some way to validate the quality of the information without losing the community aspect. Someone gave the example of the Linux kernel: it is community-developed but I think we would all agree that those permitted to contribute are the cream of the crop. Would you consider the body of Linux kernel contributors an "expert committee"?
Is it possible to post the affiliations and credentials of Wiki contributors and have these somehow audited? The basic Wiki concept is absolutely right, after all what is knowledge if not the sum-total of everyone's insights? But it's far too easy to abuse this system with the result that there is no way to assess the quality. We don't have this problem with open-source software, because the good stuff bubbles up to the top. Can we possibly set up an informal editorial board? No, I'm not suggesting we pay people to do that. But wouldn't you suppose that the foremost experts will want to have the expanded presence and notability that would come from their presence on a better-audited Wiki?
Slashdot Mirror
Where was I when Java 5 came out? I'd been waiting three years for it and complaining loudly that a fully open process would have produced it much earlier. There isn't much in Java 5 that isn't derivative of other work. That's not a criticism, it just follows along my point that Java is now synthetic rather than innovative- that's what Sun and the other consitituencies WANT from Java. Generics are templates. The enum implementation permits a degree of clarity and conciseness that shouldn't have been lacking in the first place. Hell, even the fancy wait-free concurrency that everyone raves about is nothing more than pretty words over a coding style (minimal sync-sets) that has been around for a dozen years.
Because it doesn't need to. Java is now so big and so corporate that stability is far more valuable to its constituents than innovation.
It will be the development system of choice for corporate development for another decade, providing jobs for all the new and soon-to-be CS graduates who aren't taught anything else. Yet Java is already completely ossified.
This started out as practically a greenfield project within MS with a brand new team at least five years ago. It's being done by incredibly competent people, and they have done a huge amount of work on interoperability issues- that whole raft of WS standards represents solutions to a whole range of issues that no else is really confronting. And I'm not saying they've solved them in an ideal way but as long as no one else puts anything out there that is less proprietary, they will jump out in front here. Remember how they eventually "got" the Internet?
Those of us who love F/OSS and Linux need to be less dismissive and more frightened.
MS is fragmented and balkanized internally but there are pockets of real capability. Web Services have not achieved anywhere near the level of adoption they could/should have by now (to all you trolls: the few dozen desultory SOAP projects at your company prove my point, not disprove it). And that's because of lack of "security," which boils down to lack of widely supported standards. We gotta be more proactive about this, and not make the same mistakes we are making in regard to Avalon.
Daimaou signs a (reasonably) standard IP-assignment contract, so everything he wrote while working for the company belongs to the company. That's completely straightforward. Our hero is out of the picture at this point.
So now the company owns some code which is an extension of a GPL-ed code base. They're free to do anything they want with it except distribute it to others in any form. If they do that, then they must comply with GPL's source distribution requirements. That's also straightforward.
Now the company wants to patent something that this code embodies. (The article doesn't make clear whether the patentable content is part and parcel of the GPL-ed codebase, but I'll assume it is.) If it's an American company, they're free to try to do this, and given how incompetent the USPTO is, they stand a pretty good chance of getting the patent. If the patent examiners are awake, then they'll notice the existence of substantial prior art as embodied in the GPL code base, but they're usually not awake. (Disclosure: I am a patent-holding inventor of certain firewall technology.)
Skype doesn't play nice in the sandbox with the rest of the world. I hope you're not thinking Google would buy Skype and try to suppress SIP... horrors!
What does Google do better than anyone else in the world? They can efficiently make observations and draw inferences across all of the world's electronic information. It's by far the biggest in-cache working set ever.
So how can you apply their unique knowledge to produce a VoIP service that is qualitatively better than can be offered by anyone else in the world? If there is no concrete answer to that question, then they shouldn't offer the service- it would be a bad business decision. Do any of you PhDs out there in slashdot-land know the answer?
Shiva the destroyer of worlds reminded me of another point: ;-). My C/C++ and Java almost always compiles the first time. You wipe out a whole class of stupid errors with this technique.
:-)
Try reading your code out loud, as silly as that sounds. Not only will that immediately cure you of identifiers-without-vowels and the evil curse of hungarian notation, but you'll almost instantly know whether it makes logical sense. Just try it before you flame, especially if you're a young guy who already knows everything
Reading code out loud is the next best thing to reading it to someone else.
Shiva: were you in the Manhattan Project or something?
A gentleman of my acquaintance is a phenomenally successful retired CEO of a Fortune 500 company. He's now a visiting professor at a very prestigious B-school, a household name (although I won't name it). It's not too much to say that he was shocked at the lack of ethical sense among today's B-school students. It's probably no shock to slashdotters that PHBs-in-training are only in it for the money, but then we shouldn't be surprised when garbage like this happens either.
To think of code as literature is actually a non-obvious and striking idea.
The essence of literature is communication. I've been writing C and C++ for years, constantly fine-tuning and refining code and documentation, line by line, paragraph by paragraph. And how apropos to realize that the point is to express the thought so clearly and concisely that any reader (including yourself three weeks from now) will grasp it as close to immediately as possible.
That means you have to think clearly in order to code clearly. This matters more than almost anything else.
Too much documentation is just as bad as too little, since it obscures the intent of the code. How much is just enough? It takes years of thoughtful practice to know that.
Everyone is posting their favorite rules about coding standards. Most young programmers apply standards as rigid as they are incomplete. I have only two rules:
1) Call everything by its right name. Functions are verbs, variables are nouns. The name you give a object should express what it is or does, no more and no less, and be readable in English (or whatever your spoken language is). If you add something to a function that doesn't fit the function's name, it either belongs in a different function or else you misnamed the function in the first place. This takes constant review of your code, from a literary point of view.
2) Rules are made to be broken. All great writers unhesitatingly break the rules in order to communicate more clearly.
Languages are for communicating, and computer languages are no different. Keep that in mind at all times, and you'll become a wizard.
1) It doesn't work well enough.
;-). But even if they believed it themselves (which they may), Sun has become a seriously risk-averse culture in recent years and I don't think they would lightly de-emphasize the decades of brand strength associated with Sparc hardware.
>>>Meaning what?
The combination of cost, features, robustness and stability are not superior to the analogous aggregate for Linux, in low-end server applications. In short, there's not enough reason to abandon Linux for Sun 10. The calculation goes the other way in Sun's core market, which is high-end enterprise application servers.
2) It takes away from Sun's core business, which is and always been selling Sparc-based hardware
>>>Sun's core business has always been selling networked Unix systems and the services to go along with them.
I think that your statement represents Sun's core marketing spin rather than their core business
3) It doesn't provide a compelling alternative to Linux in the application space which is targeted by inexpensive x86-based hardware.
>>>It doesn't have the device support that Linux does, but it runs on a huge amount of x86 hardware. I think the performance, stability, and features like SMF, Zones, and dtrace make it a compelling alternative.
Remains to be seen. dtrace is interesting in low-end applications, the other two less so.
4)I've been trying to do Solaris on x86 for 10 years now, ever since Solaris version 2.3. It just isn't stable.
You're making my point with your statement about Sparc's better hardware. Plus, Solaris is seriously optimized for the Sparc architecture, and always has been.
You make a number of very interesting points. The key insight is that hard, physical resources need to be provisioned for servers and bandwidth. The only large-scale counterexamples I can think of are BitTorrent, Gnutella and SETI. Can this kind of community provisioning work for a free/open search system? I'd sure like to know, what does everyone think?
:-)
Your objections about the business model (personal ads) are less germane, I think. Those issues go away if you don't need to run the service to make money. Why do we use Yahoo or Google? For the search service. Why does Yahoo or Google provide the search service? For the purpose of selling ads. When I really want to see ads, they're not hard to find
Let's keep sight of the original point, which is that Yahoo is the latest company looking to capture value from the labor of the community. That means from you and from me. That's the part that I think should be resisted. I don't know if an open Yahoo alternative makes any sense. But I do know that smart programmers willing to give free labor to Yahoo need to get a life.
I have an interesting technical problem to solve... would you like to work on it? It doesn't pay anything, but I'll make a lot of money selling your work. Interested? :-)
Running Sun 10 on an Epia: that's a very interesting science project. But remember that Sun is a multi-billion dollar business. They would need to evaluate the market for Sun on mini-ITX platforms and conclude that it can add at least a few hundred million per year to their top line before it gets interesting. Also, their core market is enterprise IT, and it's far from clear that this market has a great interest in mini-ITX platforms, unless you're proposing to run Sun 10 on desktop-replacement laptops.
Regarding Solaris on x86:
1) It doesn't work well enough. Certainly not well enough to displace Linux in the core market for x86-based servers.
2) It takes away from Sun's core business, which is and always been selling Sparc-based hardware and peripherals to enterprise IT.
3) It doesn't provide a compelling alternative to Linux in the application space which is targeted by inexpensive x86-based hardware. If you want to, you could try making the case that you could run Sun on all your machines from top to bottom and avoid the need to hire Linux admins, but I'd need to see the numbers.
4) I've been trying to do Solaris on x86 for 10 years now, ever since Solaris version 2.3. It just isn't stable.
I have no quarrel with the process or the license. (And I didn't say anything at all about GPL, so don't put words in my mouth.)
Sun can release all the source code they want. My point is that it's not going to bring them the business benefits they are hoping for, because the underlying economics are broken.
Why on earth would any intelligent, well-motivated and talented hacker want to work for Yahoo/Google/IBM/Sun/whoever WITHOUT getting a salary from them? All of these companies that are talking about tapping the capabilities and intelligence of the "community" must think we have no intelligence at all!
It's the same thing with open-sourcing Solaris. Anybody who is talented and enthusiastic enough to make serious contributions to a major search engine or operating system should be doing it to benefit the whole community, not just to make some major corporation even richer.
We already know about the open alternatives to Solaris. Where is the open and free alternative to Yahoo? I'll contribute time and money to it!
Solaris is a very very fine piece of technology, and Sun's many enterprise customers will be installing upgrades and consuming services for many years to come.
However, Sun is not going to win any new business with this release- all the growth is above them and below them. They know they have to do something, which is why they "open-sourced" 10, but this is nothing more than a lame attempt to get some of the benefits of community development. How many of you Linux/BSD hackers out there are all fired up to start contributing features and bug fixes to Scott and Jon? I thought not. And in regard to Solaris on x86, it's completely senseless, for more than one reason.
Sun's total market cap would fit more than three times into Microsoft's bank account, and that has been true for over three years now. The problem is, no one has a need to acquire Sun, either as an accretive play (no growth) or to kill them. They're going to twist in the wind for at least another decade, but they stopped being interesting long ago.
We offered to fund a portion of the OpenBSD project's budget as a corporate sponsorship. We were thinking about putting in perhaps a million dollars a year. They responded cautiously until we asked how much they might need. The emails we got back are not printable on a family website :-). This was back in 2003, several months before DARPA pulled much of their grant money. When we started hearing from fallen-away project members over the next few months, we realized we were lucky not to have gotten involved.
Anyway, no knock against Theo or his achievements. He deserves his recognition, as I said in my original post. He's running his project like a sandbox, however, and that makes it hard to depend on.
Theo deserves a lot of recognition for his technical achievements and his commitment to freedom. Getting this award proves that you can blow off everyone in the world except your personal fanboys and still be a success.
My company based a commercial product on O-BSD, then converted to Linux when it became clear that Theo doesn't know how to anchor a diverse community. We even tried to fund his project but never got past being personally abused.
Not 100% sure but I think the MQ core is written in straight C. It still costs a fortune and is absolutely vicious to to keep running. On Wall St, they employ armies of people to do just that. An open-source alternative is apropos.
I'm still laughing at the Stereo Review cartoon ca. 1975 with a radio announcer introducing a performance of a Stockhausen piece, performed on the original transistors, resistors and capacitors.
TIB's pub/sub model isn't appropriate for many applications (which is why they bought and killed Talarian- remember them?). People use MQ for the robust, guaranteed delivery and asynchrony. Plus, TIB is wickedly expensive, and a bear to admin. An open source alternative is a fabulous idea.
My company built and sold a message-oriented messaging product that merged the pub/sub and transactional models several years back, and it included guaranteed delivery. And yes, there was a lot of discussion about open sourcing the code base. The system is still in daily production in a bunch of Wall St. firms.
What these firms all wanted as a key requirement was clean integration with C/C++, and extreme performance. We had to handle thousands of messages (ranging into megabytes each) per second, with full guaranteed delivery, over IP networks and through firewalls. Oh, and none of the flakiness and admin problems of MQSeries.
None of the available JMS products could handle all these requirements.
It's a well-considered and interesting article, but misses the point.
Are the four freedoms enough to make people use alternatives to proprietary software? In themselves, no.
The freedoms are one of the key quality dimensions that determine user choice, but not the only ones. Completeness, stability, suitability for the purpose, and ease of use matter as much or more. Oh yeah, a minimal money-cost to acquire the right to use a given piece of software is also one of these dimensions.
The choice of which software to use is a strictly "economic" one, in which all of these dimensions will be added together by the decision-maker, and an optimal choice will be made.
So what does it mean to be a "true believer" in F/OSS (as I am)? It means you believe that over time, the quality, stability, and suitability of fully-free software will rise to match and surpass the commercial alternatives.
Fair enough. Wiki will die a quick and unlamented death if it becomes constrained by expert committees. But the basic problem is still: how can we assess the quality of the information? Having observed /. for a long time, I'm skeptical that community-generated mod points will result in anything but the validation of the prevailing (and evanescent) point of view. It's a market after all, and markets exhibit the same behavior- it's a flaw in the "perfect knowledge" theory.
What I guess I'm reaching for is some way to validate the quality of the information without losing the community aspect.
Someone gave the example of the Linux kernel: it is community-developed but I think we would all agree that those permitted to contribute are the cream of the crop. Would you consider the body of Linux kernel contributors an "expert committee"?
Is it possible to post the affiliations and credentials of Wiki contributors and have these somehow audited? The basic Wiki concept is absolutely right, after all what is knowledge if not the sum-total of everyone's insights? But it's far too easy to abuse this system with the result that there is no way to assess the quality.
We don't have this problem with open-source software, because the good stuff bubbles up to the top. Can we possibly set up an informal editorial board? No, I'm not suggesting we pay people to do that. But wouldn't you suppose that the foremost experts will want to have the expanded presence and notability that would come from their presence on a better-audited Wiki?