The claim of the LGPL's effect is bogus anyway
on
Two Helpings of WINE
·
· Score: 5, Informative
The LGPL does not prevent proprietary additions the way the GPL does. They can be static linked, too. So it would not do anything to the Direct3D work. And the "stealing" claim is entirely specious.
If the WINE team wants to avoid leeches, they need some more license consultation.
Bruce
LGPL does not prevent proprietary additions
on
Two Helpings of WINE
·
· Score: 5, Insightful
This is a most mixed-up story! The use of the LGPL does not prevent anyone from making a proprietary addition! It would not "steal" the Direct3D work. Proprietary code may be linked to LGPL code. It may be static-linked, dynamic liked, anything. Libraries may include mixed proprietary and LGPL work as long as the two can be separated for re-linking (which means the.o files have to be available, big deal). The decision to go to the LGPL does not retroactively change the license on old-code, either, only new additions.
You are correct that there is still strategy being worked on. The integration team was sort of clean-room until the merger closed - it would have been a bad idea to contaminate all of HP management with Compaq insider information if the merger for some reason did not close, so a lot of them are being brought in now. I have some future deliverables in this regard.
The purpose of this document is to tell customers what is going away. Linux isn't. You would not believe how much customer action Linux has been getting of late. If I took all of the sales presentation invitations I get, I'd never see my 2-year-old again. HP has to go where the customers are, and they are asking for Linux.
Red Hat offers two embedded systems: Linux and eCos. I was wondering if the problem was more with eCos than Linux. But then again, I think the problem for Red Hat is that embedded is not their specialty, and if someone is going to go to a company at all for embedded Linux (they don't have to - after all) they will go to one that needs its embedded customers.
Actually, most embedded modifications can be kept proprietary, through the modules exception on the kernel and through placing of proprietary code in a user-mode program.
I created the user-mode half of an Open Source embedded system - a program called busybox. It's everywhere. It is hosted on Linux much more often than BSD. People don't seem to need BSD licensing on the kernel that much.
The problem is not a lack of customers. Linux is the standard for embedded systems these days. The problem, if you call it a problem, is that a lot of vendors aren't profitable at offering embedded Linux. It's too easy for the customer to get along without those vendors. And I have no problem with that.
Sure, a number of companies will shake out. Some people blame this on the Open Source nature of the software, but they are really missing the point. Open Source is not a business, it's a cost-center that is shared among a large number of collaborators. A cost-center is less than half of your business plan. Open Source is non-differentiating for your business, because everybody else can offer the same thing, and thus the best plan a business can have for it is to use it to reduce cost so that they can spend more on their differentiators and profit centers.
In making a business plan, you will need to figure out what your differentiators and profit centers will be. It's obviously not going to be software sales - the folks who tried a royalty-based embedded system based on Linux went out of business faster, because they had free competition. In general it will be consulting services, and this is going to be a difficult business in a slow economy as businesses will try to do more with their own engineers. Businesses also have an incentive to use their own engineers for embedded work, as they don't want to be in the situation of losing the recipie for one of their own products. That can happen more easily when an outside vendor does the work.
I created the user-mode half of most Linux embedded systems - a program called busybox. It's everywhere. I used the GPL. Because of that decision, the person who put the most effort into maintaining that program, after me, is still working on it and is able to offer his consulting services on it. Had I not used the GPL, he would have had to give it up when his previous employer was one of those shaken out. I have a lot more sympathy for him than the employer. Also, had I used a license other than the GPL, the program would not have become an open standard for embedded - everyone would have been making their own proprietary additions rather than cooperating. And I didn't care that companies could not lock in a revenue capture on busybox - why should I?
I think Debian has the best "business plan" of any Linux distribution: don't even try to make money. The people who use Debian as a cost center (HP, for example) pay for its development, and they are very clear about what their profit centers are. This is why I think Debian will eventually end up on top.
The reason it's important is a matter of corporate culture. IBM does not have a person like me, and never will, because the corporate culture would not allow it. They can't "get it" to the extent that HP does. Nor would IBM employ the Debian project leader (Bdale Garbee).
I don't think you need to question HP's Linux committment. We have to go where our customers are going, and we get very firm "Linux" signals from them.
You now have Jim Gettys, me, Bdale Garbee, David Mosberger, and Jeremy Allison in the same company, along with another 100 people I really should mention. There's a bigger array of Linux expertise than VA ever assembled, and most of them are working on GPL projects, and are also driving the company significantly. That's got to be good for Free Software.
OK, I can't claim to know much about HP's old lines. But the 3000 and 9000 are very similar - you can actually convert one to the other. The difference is that the 3000 runs MPE, an old mainframe OS, and the 9000 runs HP-UX. HP recently announced a 5-year de-support schedule for MPE, with other vendors coming in to continue support if the customer wishes it.
I have a hard time understanding the MPE world. The CTO of a big MPE consulting firm sat me down at HP World last year and told me why Unix would never make it in business. Not Linux, Unix. So, I decided I would only worry about the Free Software operating systems at HP.
In the case of HP-UX, on the IA-64 it will be binary-compatible with Linux. If you write to that interface, you have an application that runs on two operating systems, one of which gives you the option to self-support indefinitely (or buy consulting) and not worry about when someone decides to de-support it. HP had no choice but to embrace Linux, the company has to go where its customers are going.
Given that both companies had handheld lines, there is a question of which ones will continue to be developed. The Jornada 700 series ran Linux well too (without official support from HP, although their labs were involved in porting) and the iPAQ does (again, without official support although Jim Gettys and his lab were porting). I would prefer to have a Linux palmtop that comes with Linux and doesn't need to get it loaded on after the purchase. That way, we'd get decent peripheral and application support. HP's only effort in that direction so far was directed toward the teenage market, and died when they saw how well the Palm m100 was doing. It would be nice to see more now.
I think the intel IA-64, a direct descendant of PA-RISC, will take over at the high end.
Regarding the other operating systems, I'm fortunate that I only have to worry about the Free Software ones. I do know that HP-UX will be around for a long time.
The stock symbol "HP" belongs to some mining company. This has been a source of confusion for a generation or so.
Bruce
HP is probably the largest Linux company now
on
HP, Compaq Deal Approved
·
· Score: 4, Interesting
This probably makes HP the largest Linux company by shipped product. Did you know that HP offers 24/7 support for Debian?
The merger has not been a comfortable thing from day one, and the press coverage has been very disquieting. It's clear what people like me in the company should do now - our best to make it work, regardless of anything that happened on the way. I said a long time ago that this could be excellent for Linux, and I still think so. It's going to be fun.
TurboTax and Quicken install advertising icons on the user's desktop. A whole bunch of Windows applications do that, often icons for Internet providers, but in the case of TurboTax and Quicken they install icons for banks.
These folks really must think that they own the user once the user buys their product, becuase even a "respectable" company like Intuit doesn't seem to have any problem with monkeying around with the private parts of the user's computer for their own purposes. Certainly those icons are paid placements.
If somebody formed a start-up around it, they thought their chance of success was good. In contrast, say your software has a 10% chance of success. Do you really think you'll get investment? I think that non-profit avenues are really the only way to produce that sort of software.
You might try to summarize what Smith has to say if you have a point to make. Not that Smith's is the last word on capitalism. You might be arguing some sort of trickle-down economics. I can't tell.
Oh good! I'm happy to have the chance to argue with a real economist.
Bruce: In contrast, once you have amortized the cost of creating a piece of software, there is essentially no marginal cost associated with creating another copy.
Daytrip: This is not exactly true. True enough that each physical product associated with software has a marginal cost of zero, however more goes into any specific software product than just the cd's and the packaging. There also significant marketing costs, research costs, and support costs associated with each purchased item.
OK. Let's examine the three factors you pointed out: marketing, support, and research.
We do marketing communications differently. We rely on the software being on
hand for the user to try. It's either on their system or downloadable via the
internet, so that the customer can see if it solves their problem.
This doesn't have a significant cost for us.
I don't think you have addressed strategic marketing rather than marketing communications. We do that differently as well.
Support can take place via the usual pay-for-service model (although there are alternatives). Support is not coupled
to the product purchase in our model.
That leaves us with research. But that's a cost that can be amortized in the
cost of creating the product. Yes, for a business it's an ongoing cost, but
that's not how we pay for it.
Bruce: Can we amortise the creation cost of software without a direct revenue capture per unit sold? The answer seems to be yes for a lot of people.
Daytrip: While I certainly agree with this point, most firms (the ones without an idealogical agenda, but simply those in the business of making money) maximize profit.
But you are only considering businesses that sell software. What about most
businesses, which use software as a means to carry out some other activity?
Many do employ their own programmers, because off-the-shelf often won't do.
Consider Apache in this light. It was created by people who had to serve
web pages for some business that most often wasn't software development.
Daytrip: Moreover, the viral nature of the GPL further prevents any corportation from truly maximizing profits once they use GPL'ed software, even though these corporations (with the taxes they pay) actually supported the development of those products.
Again, you are only considering this from the perspective of a business that sells software. For other sorts of businesses, software would otherwise be make-or-buy, and there may well be savings due to collaboration with other businesses, ease of customization, etc.
Bruce: if you want to consider me as selling out the software development profession, I'm doing it for the customer.
Daytrip: I object to this argument in particular. Naiively, the best model for consumers is for everyone to produce software for free, and provide support for free and give everything away for free. While this, in the short run, would be quite advantageous for consumers, after a while, all corporate profits (and earnings) would run dry, killing the industry.
Again, you aren't considering the role of the customer in developing their own Open Source. You are only considering this from the perspective of a business that produces software for its income. But there are many customers who produce their own software for their own use. These are the people who carry out Open Source development.
The result of my argument if taken farther than it will perhaps ever go would be that proprietary software development might dry up. But it could be possible that nobody would miss it. Business as a whole would not dry up, and efficiency could improve.
Daytrip: Moreover, if all industries were to do this, and consumers were only to pay for the natural resources involved in making a product, this would essentially de-value labor and make fixed resources the only tenable currency
You are postulating that the Open Source model applies to the entire economy, then disproving that. This is of course taking my argument to the point of absurdity. But my argument doesn't apply to the entire economy, as I've made clear. It is a very
specialized exception for commodities that: 1) can have their design cost amortized
some way other than by per-unit-sale revenue capture and 2) have essentially no marginal
cost to duplicate. There are science-fictional scenarios where this might
someday be more than just software, but I don't think they will be true for
a long time.
In practice, software that has a low risk of paying off is only developed as Open Source. People won't take the risk of commercial development without significant confidence of a pay-off. That's why the web server and browser were Open Source creations.
Well, there isn't a way for me to copy chips the way I copy software, until there are fast field-programmable gate-arrays in popular computers. But yes, once that day comes, there could be essentially no marginal cost associated with making a copy. And perhaps that could be true for any hard good if we ever get really good automated fabrication, although there most probably will be a significant energy and material cost.
The argument applies today to any media that can take on digital form. Once you amortize the design cost, you can make copies for free. The question is: can you come up with a scheme to amortize the design cost without a per-unit revenue capture? It happens to be true for many kinds of software, because software enables other sorts of sales. Maybe this doesn't work for music or movies, I don't know. Regarding patents, that's a whole different argument - I think most patents are not justly awarded.
Hm. I think you are saying that a lot of Open Source projects go on because of personal, rather than business, motivation. A microeconomist would argue that personal motivation can be examined in economic terms, and that one can establish economic viability by comparing outlay (personal time, etc.) vs. utility (intangible and tangible benefits derived).
Slashdot Mirror
If the WINE team wants to avoid leeches, they need some more license consultation.
Bruce
Bruce
Bruce
Bruce
Bruce
Bruce
I created the user-mode half of an Open Source embedded system - a program called busybox. It's everywhere. It is hosted on Linux much more often than BSD. People don't seem to need BSD licensing on the kernel that much.
The problem is not a lack of customers. Linux is the standard for embedded systems these days. The problem, if you call it a problem, is that a lot of vendors aren't profitable at offering embedded Linux. It's too easy for the customer to get along without those vendors. And I have no problem with that.
Thanks
Bruce
In making a business plan, you will need to figure out what your differentiators and profit centers will be. It's obviously not going to be software sales - the folks who tried a royalty-based embedded system based on Linux went out of business faster, because they had free competition. In general it will be consulting services, and this is going to be a difficult business in a slow economy as businesses will try to do more with their own engineers. Businesses also have an incentive to use their own engineers for embedded work, as they don't want to be in the situation of losing the recipie for one of their own products. That can happen more easily when an outside vendor does the work.
I created the user-mode half of most Linux embedded systems - a program called busybox. It's everywhere. I used the GPL. Because of that decision, the person who put the most effort into maintaining that program, after me, is still working on it and is able to offer his consulting services on it. Had I not used the GPL, he would have had to give it up when his previous employer was one of those shaken out. I have a lot more sympathy for him than the employer. Also, had I used a license other than the GPL, the program would not have become an open standard for embedded - everyone would have been making their own proprietary additions rather than cooperating. And I didn't care that companies could not lock in a revenue capture on busybox - why should I?
I think Debian has the best "business plan" of any Linux distribution: don't even try to make money. The people who use Debian as a cost center (HP, for example) pay for its development, and they are very clear about what their profit centers are. This is why I think Debian will eventually end up on top.
Bruce
I don't think you need to question HP's Linux committment. We have to go where our customers are going, and we get very firm "Linux" signals from them.
You now have Jim Gettys, me, Bdale Garbee, David Mosberger, and Jeremy Allison in the same company, along with another 100 people I really should mention. There's a bigger array of Linux expertise than VA ever assembled, and most of them are working on GPL projects, and are also driving the company significantly. That's got to be good for Free Software.
Bruce
I have a hard time understanding the MPE world. The CTO of a big MPE consulting firm sat me down at HP World last year and told me why Unix would never make it in business. Not Linux, Unix. So, I decided I would only worry about the Free Software operating systems at HP.
In the case of HP-UX, on the IA-64 it will be binary-compatible with Linux. If you write to that interface, you have an application that runs on two operating systems, one of which gives you the option to self-support indefinitely (or buy consulting) and not worry about when someone decides to de-support it. HP had no choice but to embrace Linux, the company has to go where its customers are going.
Bruce
Given that both companies had handheld lines, there is a question of which ones will continue to be developed. The Jornada 700 series ran Linux well too (without official support from HP, although their labs were involved in porting) and the iPAQ does (again, without official support although Jim Gettys and his lab were porting). I would prefer to have a Linux palmtop that comes with Linux and doesn't need to get it loaded on after the purchase. That way, we'd get decent peripheral and application support. HP's only effort in that direction so far was directed toward the teenage market, and died when they saw how well the Palm m100 was doing. It would be nice to see more now.
Bruce
Bruce
Regarding the other operating systems, I'm fortunate that I only have to worry about the Free Software ones. I do know that HP-UX will be around for a long time.
Bruce
Bruce
The merger has not been a comfortable thing from day one, and the press coverage has been very disquieting. It's clear what people like me in the company should do now - our best to make it work, regardless of anything that happened on the way. I said a long time ago that this could be excellent for Linux, and I still think so. It's going to be fun.
Thanks
Bruce
Enough already!
AOL recently purchased my entire family.
These folks really must think that they own the user once the user buys their product, becuase even a "respectable" company like Intuit doesn't seem to have any problem with monkeying around with the private parts of the user's computer for their own purposes. Certainly those icons are paid placements.
Bruce
Bruce
Seriously, the editor did it. He knows his readers.
Bruce
Bruce
Oh good! I'm happy to have the chance to argue with a real economist.
Bruce: In contrast, once you have amortized the cost of creating a piece of software, there is essentially no marginal cost associated with creating another copy.
Daytrip: This is not exactly true. True enough that each physical product associated with software has a marginal cost of zero, however more goes into any specific software product than just the cd's and the packaging. There also significant marketing costs, research costs, and support costs associated with each purchased item.
OK. Let's examine the three factors you pointed out: marketing, support, and research.
We do marketing communications differently. We rely on the software being on hand for the user to try. It's either on their system or downloadable via the internet, so that the customer can see if it solves their problem. This doesn't have a significant cost for us.
I don't think you have addressed strategic marketing rather than marketing communications. We do that differently as well.
Support can take place via the usual pay-for-service model (although there are alternatives). Support is not coupled to the product purchase in our model.
That leaves us with research. But that's a cost that can be amortized in the cost of creating the product. Yes, for a business it's an ongoing cost, but that's not how we pay for it.
Bruce: Can we amortise the creation cost of software without a direct revenue capture per unit sold? The answer seems to be yes for a lot of people.
Daytrip: While I certainly agree with this point, most firms (the ones without an idealogical agenda, but simply those in the business of making money) maximize profit.
But you are only considering businesses that sell software. What about most businesses, which use software as a means to carry out some other activity? Many do employ their own programmers, because off-the-shelf often won't do. Consider Apache in this light. It was created by people who had to serve web pages for some business that most often wasn't software development.
Daytrip: Moreover, the viral nature of the GPL further prevents any corportation from truly maximizing profits once they use GPL'ed software, even though these corporations (with the taxes they pay) actually supported the development of those products.
Again, you are only considering this from the perspective of a business that sells software. For other sorts of businesses, software would otherwise be make-or-buy, and there may well be savings due to collaboration with other businesses, ease of customization, etc.
Bruce: if you want to consider me as selling out the software development profession, I'm doing it for the customer.
Daytrip: I object to this argument in particular. Naiively, the best model for consumers is for everyone to produce software for free, and provide support for free and give everything away for free. While this, in the short run, would be quite advantageous for consumers, after a while, all corporate profits (and earnings) would run dry, killing the industry.
Again, you aren't considering the role of the customer in developing their own Open Source. You are only considering this from the perspective of a business that produces software for its income. But there are many customers who produce their own software for their own use. These are the people who carry out Open Source development.
The result of my argument if taken farther than it will perhaps ever go would be that proprietary software development might dry up. But it could be possible that nobody would miss it. Business as a whole would not dry up, and efficiency could improve.
Daytrip: Moreover, if all industries were to do this, and consumers were only to pay for the natural resources involved in making a product, this would essentially de-value labor and make fixed resources the only tenable currency
You are postulating that the Open Source model applies to the entire economy, then disproving that. This is of course taking my argument to the point of absurdity. But my argument doesn't apply to the entire economy, as I've made clear. It is a very specialized exception for commodities that: 1) can have their design cost amortized some way other than by per-unit-sale revenue capture and 2) have essentially no marginal cost to duplicate. There are science-fictional scenarios where this might someday be more than just software, but I don't think they will be true for a long time.
Thanks
Bruce
Bruce
The argument applies today to any media that can take on digital form. Once you amortize the design cost, you can make copies for free. The question is: can you come up with a scheme to amortize the design cost without a per-unit revenue capture? It happens to be true for many kinds of software, because software enables other sorts of sales. Maybe this doesn't work for music or movies, I don't know. Regarding patents, that's a whole different argument - I think most patents are not justly awarded.
Bruce
Bruce