Re the various comments "with an open source monopoly all the jobs would go away": Note that in our article, we were primarily thinking of and talking about platform software. There will always be a niche for software that performs specialized tasks (i.e. if I want a control program for my homemade veeblewhitzer that uses a propriertary protocol using parts of an EEG machine to monitor the my cat's water bowl via passenger pigeon, I'm going to have to either write it or pay someone because it's just not generally useful enough). I think we were clearer about the platform distinction in the original draft of the article than we ended up being in the final, shorter, draft. My bad.
For things that are useful to a broad spectrum of people, open source just makes sense. For some things it doesn't.
I posted a more detailed response to some of the issues brought up in the earlier conversation on Slashdot and this one here if anyone's interested.
Now *that*'s ironic. I know and have worked with Mike, and agree completely with the article you site. Bear in mind that the article you're reading was written by two [fallible] *people*, not some vast impersonal corporate machine (albeit we had to run a legal gauntlet to publish the article).
I'm Tim Boudreau, one of the authors of the CNET article. Great discussion - I had no idea it would get so widely read! I'd like to respond semi-randomly to a some of the various responses in this thread:
"Open monopoly" is an oxymoron. No argument here. It's also a useful concept. Essentially, the notion of software occupying a monopoly position in the software market (i.e. ubiquity), without it being controlled by a single entity.
No one will correctly predict where things are going. Computers and the devices that run them are too varied and change too quickly.
Agreed, nor are we trying to make predictions - rather to point out what is predictable given our non-omnicient understanding of the rules of the marketplace.
This chaos is not limited to the effect of the
transistor.
Absolutely - and any single company having sole control of a piece of critical infrastructure causes software to advance more slowly.
I'll pour a few hundred hours of blood, sweat and tears into it! Shiny new! Everyone wants it! It's hot! But how do I parlay it into a commercial venture when everyone can get it for free and fix it up as they want? Hmm.
I can't speak for every project, but I can speak for ones I've been involved with. In terms of dealing with platform level software, it's clear that developing platform level software is a cost. In the case of the project I work on, NetBeans, it's a generic IDE platform (actually a generic large desktop application platform) that people are free to write extensions to, create their own applications with, and give away or sell those creations as they wish. E.g. if I come up with a great debugger, odds are I wanted to write a debugger, not a full blown IDE in order to have a usable interface to my debugger. Having standard, open platforms means more software is developed faster, and everybody benefits. The question is one of deciding what pieces you want to give away, what if any pieces you'd prefer to sell, and doing that in a way that allows you to eat.
They don't care about the long term future of the software industry, they care about the need to compete with voracious rivals in THIS economy! I wouldn't generalize that to every company in the software industry - if you want to survive, you have to think long-term, and since everybody's actions have an effect, you have to think in terms of what is going to help create an environment in which you can survive.
...it seems pretty clear to me that only a radical overthrow of the entire system can improve our lot and stop the evils.
Seems to me most of the time throughout history, a radical overthrow of anything resulted in either more of the same or worse conditions. For better or worse, humans are self interested creatures. Noble sentiment, but I expect that evolving an more equitable system starting from the existing less equitable one is more likely to get you the result you want - when you create a brand new system from whole cloth, you don't know what perverse incentives you've built into it until they come to bite you.
...A more restrictive liscence for the people denying access to selfish concerns would be a great boost and a bigger threat to MS than anything.
GPL anyone? Or if you mean more restrictive than GPL, how would it be policed/enforced?
So they're hoping to replace Microsoft with "open source" - though they mean by that an open source operating system that is ruled by industrial comittee:
Huh? Who said anything about industrial committee? Yes, we refer a lot to companies as participating entities, but our entire point is that nobody is blocked from participation, whether corporate or individual.
With companies such as IBM and SUN backing Open Source the question remains as to where do these companies see there future revenue being generated.
I don't forsee all commercial software going away. There are plenty of people/companies in the world who need software that is plain uninteresting or unpleasant to write. My last job as a programmer/contractor before going to NetBeans/Sun was building a large scale emulation of a DOS application in JavaScript. Completely dumb project that should have never been started. I would never have done it if I wasn't being paid to do it, but there are also plenty of things I've worked on that I didn't get paid to work on. So there's always going to be a market for commercial software, but I think there's also always going to be a migration of things that prove generally useful toward open standards and open source.
Unfortunately, the only people who are able to participate effectively in the design and creation of an open source project under existing models are computer programmers.
I don't know about this - with NetBeans, we have UI designers and usability people, not to mention documentation writers who are not programmers and have no problems participating, and a healthy community of folks on the users mailing list who submit bugs and help each other. Coding is not the only form of important participation in an open source project.
Now can you spell a-d-v-o-c-a-c-y ? This puff piece was meant for the suits.
If it's not already an acronym, can I coin IANAS (I am not a suit)? Heck, the article started as a 3AM rant. Did the fact that it had suit appeal help get it out the door? Sure. Mea culpa.
I would not be surprised if one of those three companies effectively had the power to dictate and licence ALL computer use, of any kind, anywhere.
At which point that company would have killed the very market it relies on for survival. No company finds it strategic or has the resources to implement all of the things people want. Open standards and open source empower people to create what they want. Anyone who gets themselves in this position has just committed hari kari.
The author argues that when the MS Monopoly falls the successor (Open Source?) will be the next adopted monopoly. This is a fallacy. It's like calling VHS a monopoly because there is no viable competetion (like betamax.) The VHS standard is the PREFERED technology of the consumer. There is no single vendor (monopoly) of VHS.
This is quibbling over semantics. VHS is a good analogy for what we're talking about, but VHS is not an evolving technology like software, it's a static standard.
...the argument falls apart when you start getting into really popular (widely-used) projects. How many users of Apache (on a percentage basis) have participated in its development/design? StarOffice/OpenOffice? The Linux kernel?
How many needed to? The point isn't that everyone who uses something participate in its creation, but that enough who use it do, and that nobody is barred from such participation.
Open-source is no panacea when it comes to usability and design.
True. Any company building on and contributing to open source software will find it to their benefit to also do usability testing (e.g. videotape what volunteers do and find what usability problems they have, as Sun does with NetBeans) and employ some skilled UI experts.
Slashdot Mirror
For things that are useful to a broad spectrum of people, open source just makes sense. For some things it doesn't.
I posted a more detailed response to some of the issues brought up in the earlier conversation on Slashdot and this one here if anyone's interested.
(wow, slashdotted twice...I feel special :-)
-Tim Boudreau
Now *that*'s ironic. I know and have worked with Mike, and agree completely with the article you site. Bear in mind that the article you're reading was written by two [fallible] *people*, not some vast impersonal corporate machine (albeit we had to run a legal gauntlet to publish the article).
-Tim Boudreau
"Open monopoly" is an oxymoron.
No argument here. It's also a useful concept. Essentially, the notion of software occupying a monopoly position in the software market (i.e. ubiquity), without it being controlled by a single entity.
No one will correctly predict where things are going. Computers and the devices that run them are too varied and change too quickly.
Agreed, nor are we trying to make predictions - rather to point out what is predictable given our non-omnicient understanding of the rules of the marketplace.
This chaos is not limited to the effect of the transistor.
Absolutely - and any single company having sole control of a piece of critical infrastructure causes software to advance more slowly.
I'll pour a few hundred hours of blood, sweat and tears into it! Shiny new! Everyone wants it! It's hot! But how do I parlay it into a commercial venture when everyone can get it for free and fix it up as they want? Hmm.
I can't speak for every project, but I can speak for ones I've been involved with. In terms of dealing with platform level software, it's clear that developing platform level software is a cost. In the case of the project I work on, NetBeans, it's a generic IDE platform (actually a generic large desktop application platform) that people are free to write extensions to, create their own applications with, and give away or sell those creations as they wish. E.g. if I come up with a great debugger, odds are I wanted to write a debugger, not a full blown IDE in order to have a usable interface to my debugger. Having standard, open platforms means more software is developed faster, and everybody benefits. The question is one of deciding what pieces you want to give away, what if any pieces you'd prefer to sell, and doing that in a way that allows you to eat.
They don't care about the long term future of the software industry, they care about the need to compete with voracious rivals in THIS economy!
I wouldn't generalize that to every company in the software industry - if you want to survive, you have to think long-term, and since everybody's actions have an effect, you have to think in terms of what is going to help create an environment in which you can survive.
Seems to me most of the time throughout history, a radical overthrow of anything resulted in either more of the same or worse conditions. For better or worse, humans are self interested creatures. Noble sentiment, but I expect that evolving an more equitable system starting from the existing less equitable one is more likely to get you the result you want - when you create a brand new system from whole cloth, you don't know what perverse incentives you've built into it until they come to bite you.
GPL anyone? Or if you mean more restrictive than GPL, how would it be policed/enforced?
So they're hoping to replace Microsoft with "open source" - though they mean by that an open source operating system that is ruled by industrial comittee:
Huh? Who said anything about industrial committee? Yes, we refer a lot to companies as participating entities, but our entire point is that nobody is blocked from participation, whether corporate or individual.
With companies such as IBM and SUN backing Open Source the question remains as to where do these companies see there future revenue being generated.
I don't forsee all commercial software going away. There are plenty of people/companies in the world who need software that is plain uninteresting or unpleasant to write. My last job as a programmer/contractor before going to NetBeans/Sun was building a large scale emulation of a DOS application in JavaScript. Completely dumb project that should have never been started. I would never have done it if I wasn't being paid to do it, but there are also plenty of things I've worked on that I didn't get paid to work on. So there's always going to be a market for commercial software, but I think there's also always going to be a migration of things that prove generally useful toward open standards and open source.
Unfortunately, the only people who are able to participate effectively in the design and creation of an open source project under existing models are computer programmers.
I don't know about this - with NetBeans, we have UI designers and usability people, not to mention documentation writers who are not programmers and have no problems participating, and a healthy community of folks on the users mailing list who submit bugs and help each other. Coding is not the only form of important participation in an open source project.
Now can you spell a-d-v-o-c-a-c-y ? This puff piece was meant for the suits.
If it's not already an acronym, can I coin IANAS (I am not a suit)? Heck, the article started as a 3AM rant. Did the fact that it had suit appeal help get it out the door? Sure. Mea culpa.
I would not be surprised if one of those three companies effectively had the power to dictate and licence ALL computer use, of any kind, anywhere.
At which point that company would have killed the very market it relies on for survival. No company finds it strategic or has the resources to implement all of the things people want. Open standards and open source empower people to create what they want. Anyone who gets themselves in this position has just committed hari kari.
The author argues that when the MS Monopoly falls the successor (Open Source?) will be the next adopted monopoly. This is a fallacy. It's like calling VHS a monopoly because there is no viable competetion (like betamax.) The VHS standard is the PREFERED technology of the consumer. There is no single vendor (monopoly) of VHS.
This is quibbling over semantics. VHS is a good analogy for what we're talking about, but VHS is not an evolving technology like software, it's a static standard.
How many needed to? The point isn't that everyone who uses something participate in its creation, but that enough who use it do, and that nobody is barred from such participation.
Open-source is no panacea when it comes to usability and design.
True. Any company building on and contributing to open source software will find it to their benefit to also do usability testing (e.g. videotape what volunteers do and find what usability problems they have, as Sun does with NetBeans) and employ some skilled UI experts.
Otay, let the flames begin :-)
-Tim