Catch up on stories from the past week (and beyond) at the Slashdot story archive

 


Forgot your password?
Close
typodupeerror
×
Linux Business

The Economics of Open Source 115

Posted by CmdrTaco from the now-for-something-completely-different dept.
Jason Kau writes " is a working paper on the economics of open source software from the Nation Bureau of Economic Research entitled "The Simple Economics of Open Source". Focuses primarily on Apache, Perl, and Sendmail but mentions Linux, Debian, VA Linux, etc. It's a 40 page PDF document. Some background in Economics would probably be helpful."
This discussion has been archived. No new comments can be posted.

The Economics of Open Source More

The Economics of Open Source

Comments Filter:
  • You'll find that ftp://ftp.sendmail.org/pub/sendmail/LICENSE contains the Sendmail license. It ain't the GPL. It's a BSD-ish license.

  • Oops, you're right. I should have picked a GPL example.
  • He starts out interestingly on this paper, but around page 7 he begins to get a few things wrong.

    The GPL doesn't preclude your USE of software. I can use Sendmail without having to give source to anyone. But if I modify and plan to distribute that modification, then the GPL comes into play.

    While I'm pleased to see him mention Debian's Social Contract [debian.org], I'm don't see where the GPL ever disallowed the bundling of proprietary code with GPL code at an application level. Isn't TiVo running off a Linux based system? No one has demanded their source code, just their mods to various bits of Linux and GPL'ed software.

    I must admit though, the author has certainly nailed down a number of the motivations for participating in open-source programming. My favorite quote would have to be "the programmer's performance depends on her supervisor's interference"
  • Can be zero. Check to see if you have an email address from your local library. Most libraries, including mine, already have subscriptions to these databases and upon providing NBER with your library address, will give you the paper for free.
  • >"Total aside: I seriously doubt that Windows 2000 is the most complex
    >software project ever undertaken..."
    >Why, because you are an Anti-MS Zealot?

    No because NASA has done software projects that are *FAR* more complex than the crap Microsoft has attempted with Windows 2000. Let's see Microsoft come up with code that's able to run something like the Voyager and Pioneer probes for the lenght of time they've been running in that kind of hostile enviroment.
  • Well, in some ways, there really isn't a free lunch. In open source you're limited by the base of people who will contribute their time for free, and you're also limited by the personal itches that they wish to scratch. I find that open source focuses on those who write the software, not those who use it.

    Supply & demand seems to break down in this system because there is no price system to allocate resources to get a feature done... i.e. I want feature X, but, I'm not a developer. It's a crummy job, so no one wants to develop it. So either I pay a developer, or find a way to create "increased hype" surrounding the feature.

    The problem with this scenario is return on investment.. is a feature really worth paying $50,000 for? [Assuming 1 expert developer working for 4 months.] You're effectively re-releasing it to the community if the original software is under GPL so you can't recoup the cost.

    Hmm. Many things to ponder.
  • Grow up, troll.
  • The best economic argument for open source: it's frictionless. Just download and go. No having to come up with the cash, or beg the purchasing department, justify it to the Powers That Be, etc. While Open Source writers generally don't get paid cash for their direct efforts, they aren't generating taxes either, nor are taxes (massive friction) generated by the exchange with other OSS folks. Aftertax dollars are precious.

    Of course, you'll make good money implimenting the OSS software (web admin, in-house developer, etc), at which point Big Brother will put a gun to your head and confiscate half your labor for having the audacity to flaunt having a brain, but what can y'do. (And most of that money will go towards buying the votes of the morons who clobbered you in school...)

    (I was too lazy to read the .pdf doc, so I don't know if econ boy covered this, but I suspect not..)
  • The Document gos it the history of basicly three open source projects perl, apache, and sendmail.
    It just retelling the story of these applications and people made money off them.


    http://theotherside.com/dvd/ [theotherside.com]
  • Economics cannot judge whether something is moral or not, that is left to other professions. But it can say whether some policy creates losses in value to individuals or society.

    And how can it assign value without reference to a mechanism for assigning value?

    A mandatory open sourcing of software (whether by government decree or societal pressure) will create net economic losses.

    How are these losses measured? What value system are you using to measure them? If the society as a whole feels that the existence of open source is a value in and of itself, then this statement clearly wrong in the context of those values.
  • begs for xhtml.

    You're an optimist. Why do you think people who write sucy, non-validating HTML will start writing well-formed, validating XHTML? The real problem is a history of lenient browsers and maNufacturerS more obsessed with adding features than actually creating a decent product.

  • If you're going to troll Slashdot and repeat a statement 4.38e23 times, at LEAST check your spelling on fundimental words!

  • Comment removed based on user account deletion
  • Comment removed based on user account deletion
  • An economist will make no distinction between the value of money and the value of community. They are both equally valuable. Economically speaking, someone who forgoes a $90,000 job for the pleasure of working with the community as a volunteer receives just as much value and contributes just as much to society. But before you start condemning the capitalists, the converse is equally true. It is one's individual choices that matter.

    I think we're arguing pretty much the same thing: that the intangible benefits provided to open source software developers are both real and significant.

    I'm more interested in criticizing the sanctimonious segment of the open source community that would have you believe that there is some magic good will that causes people to develop open source software, that it's not about money. I don't know where this left wing fantasy comes from, but the self-appointed spokesmen of open source, the religion, will never tell us about the intangible benefits.
  • That's already happening, in a way. "The community" (if Slashdot readers are a microcosm of that) is already pretty harsh on anyone posting anything either critical of open source software, its methods, or being supportive of closed-source software (especially Microsoft).

    I would imagine that it would be difficult to have a high standing in "the community" if you advocated or even admitted to using closed-source software.

    To borrow a quote from 60's radicals, you're either part of the problem or your're part of the solution.
  • Most of them get it wrong, assuming that there has to be some form of tangible gain from contributing code. The very science of economics tells us that basically there is no free lunch, and that people will not give up resources (in this case free time) when they do not recieve something of value in return. The recent rash of IPO wealth has sort of supported this belief, despite the fact that it was unexpected (unless this whole thing has been a clever ploy of Linus and ESR's to get rich slow)

    I'd wager that much of the "benefit" from contributing to the "open source" community is in the form of enhanced standing within said community. It's not a tangible gain, but the value systems of those involved may not desire immediate tangible gains. An increased community standing may allow access to the means to get tangible gains -- better paying jobs, free computer goodies, or the holy grail, a paying job working on open source software.

    If working on open source software was simply doing N units of programming and getting nothing out of it at all except the finished program, it's unlikely that there would be much in the way of open source software. Without the social support networks to encourage this behavior and the rewards they bring, it literally becomes the "work for nothing" that open source advocates claim it is.

    I think the situation is similar to members of a religious order -- doing works to further the religious goals doesn't result in tangible, material gains. But it does increase ones standing among the members of the order, and in the eyes of the deit(y|ies). This reward is often sufficient.

    The challenge for open source is to maintain the community structure that rewards open source contributions. As the open source software world slides further into the capitalist world, where immediate material gain is the pinnacle of sucess, it will find money vs. community gain to be a compelling battle.
  • How about a counterpaper discussing the economic burden of millions of slashdotters attempting to download a 40 page PDF file?
  • I love their anti-Slashdot mechanism...

    NBER Connection Full

    NBER's connection to the outside world has become saturated. This is a temporary situation and we hope to have it fixed soon.
    Please wait 15 Minutes and then hit Reload or Refresh on your browser.

    (This is a load and time based function...hitting the reload button immediately will not help.)

    Thank you for your patience. If the problem persists for longer than hour please contact the system administrators through the link below.

    Gerv
  • He was just making an anolog statment.
    Just to extend the setement I have seen errors made in corprate websites that NEVER get fixed.

    Websites are by nature "source available".. not allways open source but the benifit discribed is a benifit to any code with source available. Open source simply protects this with a liccens like GPL where as "liccensed source" restricts access to source code. Both however have this benifit.

    Yes his rant containned a bit of closed source bashing* but this sort of thing is hard to avoid when your talking about the benifits of open source. Just as the top of the thread is open source bashing.

    *If it is Microsoft bashing to refer to software defects then surely it is a well earned bash.
  • Slashdot Green is made of people! PEOPLE!!!

    -----------

    "You can't shake the Devil's hand and say you're only kidding."

  • I am admittedly naive about economic matters, so feel free to mod down. Um, how do economists cope with things that are innately enjoyable? ESR aside, I suspect that the main reason people code, just judging from what appear to me to be my motivations to write the little things I do, is because after a long coding session I feel like I've just played a particularly satisfying game of chess, or maybe solved a knotty logic problem. It's disturbingly addictive.
    --
    "HORSE."
  • The fixed text of the original post.. (no, I didn't cut anything off at the end).. Taco screwed up something more than html here! I wonder if this post was really intended to make it to the page or not...

    Jason Kau [mailto] writes " is a working paper on the economics of open source software [nber.org] from the Nation Bureau of Economic Research entitled "The Simple Economics of Open Source". Focuses primarily on Apache, Perl, and Sendmail but mentions Linux, Debian, VA Linux, etc. It's a 40 page PDF document. Some background in Economics would probably be helpful. " something you should

  • Following the links on the page, they do an email verify to download a pdf of their paper. WTF is this all about? Anyone have a mirror or a URL that is valid?

  • The author of the paper writes:

    As part of the General Public License (GPL, also known as "copylefting"), the user had to also agree not to impose licensing restrictions on others. Furthermore, all enhancements to the code -- and even code that intermingled cooperatively developed software with that developed separately -- had to be licensed on the same terms. It is these contractual terms that distinguish open source software from shareware....

    The author makes three key mistakes in the above. First, he states that the "copyleft" terms of the GPL are a defining characteristic of open source software -- which they are not. The BSD and MIT X licenses do not encumber programs with the onerous requirement to give away one's work when one uses the code, and both are open source.

    In fact, a strong argument can be made that because the GPL discriminates against a field of endeavor -- the creation of commercial software -- by denying the use of code to authors of closed source programs, it does not qualify as an "open source" license as defined by the "Open Source Definition" posted at http://www.opensource.org. So, the BSD license, the MIT X license, and the Artistic License -- the licenses used for Sendmail, Apache, BIND, etc. -- are open source licenses, but the GPL is not. The author fails to note this.

    Second, it is not true that the GPL requires authors of derivative works "not to impose licensing restrictions on others." In fact, it requires that licensing restrictions be imposed -- the very onerous restrictions which are part of the GPL itself. The author of a derivative work is even required to attach the "preamble" of the GPL -- a political manifesto -- to his own work.

    Finally, the author mistakenly states that the GPL is not "viral." In fact, it is viral, in that a single line of GPLed code can "contaminate" a much larger work and force it to be licensed under the GPL. Apparently, the author has bought the rhetoric of the FSF uncritically and failed to note this.

    In short, the GPL not only fails the test for an "open source" license, but also places onerous restrictions upon the code. This is intentional. It is Stallman's explicit intent, as described on the FSF Web site, that GPLed code transform open source into a weapon against the interests of commercial software developers. This intent is not noted in the text and should be.

    --Brett Glass

  • 17 USC 117(a):

    Making of Additional Copy or Adaptation by Owner of Copy. - Notwithstanding the provisions of section 106, it is not an infringement for the owner of a copy of a computer program to make or authorize the making of another copy or adaptation of that computer program provided: (1) that such a new copy or adaptation is created as an essential step in the utilization of the computer program in conjunction with a machine and that it is used in no other manner

  • these cases, the Courts concluded that possession of a copy does not, by itself, constitute ownership of that copy.
    But for GPL software, it does.
  • After doing some more research I must reluctantly conclude that you are correct with regard to Section 117.

    I seems that Section 117 has been vitiated by some rather questionable reasoning in the courts. It is quite remarkable that Congress has not corrected this rather drastic rewriting of the copyright law in the courts, but it has not.

    I would encourage all those with misconceptions similar to my own to consider the following quote from MAI:

    This conclusion is consistent with its finding, in granting the preliminary injunction, that: "the loading of copyrighted computer software from a storage medium (hard disk, floppy disk, or read only memory) into the memory of a central processing unit causes a copy to be made.
    In the absence of ownership of the copyright or express permission by license such acts constitute copyright infringement., " We find that this conclusion is supported by the record and by the law. (emphesis added)
    Thank you for the education.
  • >how do economists cope with things that are innately enjoyable?

    Short answer: They don't.

    Still a short answer (you don't expect a treatise on economics, right?): The underlying assumption in economic theory is that subjects (that is you and me) receive some sort of utility (i.e. pleasure, money, whatever). Subjects are assumed to maximise their utility.

    Everyone is free to mix in, say monetary compensation, "joy", "anger", the perception of "freedom" into a utility function - utility is an abstract concept that is, if combined with anything than money, inherently unmeasurable.

    >because after a long coding session I feel like I've just played a particularly satisfying game of chess

    Economic theory is unable to capture this satisfaction satisfactorily in a model setting. After all, it is only a model.

    Scanning the text for keywords, I would say that this is either a vastly simplified version of some ongoing research or plainly a bad, rather incomplete paper.

    +3 (educational), IMHO and after just skimming / scanning, is too much - +1 (nice try) appears to be more appropriate.

    FWIW, I have about 6 MB of texts on OSS and economics, all obtained from the 'net - some of that has fewer buzzwords, but definitely more content. Perhaps the audience (supposedly economists) simply wouldn't be able to appreciate a more formal model due to lack of familiarity with the economics (sic!) of OSS.

    There are quite a few other ways to tackle this topic other than "career concerns". I personally do not think that this is the right approach.
  • Well, on page 8, he talks about open source as if it were a license; not a license categorization. He then points to the "License Must Not Contaminate Other Software" clause and infers that open source is not viral, unlike the GPL. He completely missed the fact that GPL is talking about linking, not shared distribution. I'm not sure if he has done all of his homework yet... I guess this is a work in progress?
  • Now we all now what happens when somebody forgets a " in the code... :-)

    -pf

  • Actually the 'common wisdom' seems to be correct here. You have to remember that GPL operates under *copyright* law. Thus if you are not copying ... there is no restriction at all.

    If you modify a GPL and keep it to yourself no problem. If you distribute, you fall under copyright and GPL applies. This has *nothing* to do with licensing in the more common (EULA) sense.

    Copyright has nothing to say about what you do with your copies....

    S.
  • The link in the parent post leads to a listing. To get the paper directly click here [hbs.edu] (pdf document)
  • Footnote 13 on page 18 of the paper says this:

    "An argument often heard in the open source community is that people participate in open source projects because ... they want to be 'part of a team.' While this argument may contain a grain of truth, it is puzzling as it stands; for, it is not clear why programmers who are part of a commercial team could not enjoy the same intellectual challenges and the same team interaction as those engaged in open source development."

    Through my experience, I've noticed that a significant percentage of commercial developers are barely qualified for their job. This has become more so int he current developer shortage. These are people who have no real love for programming, but rather view it as a "job". In the open source community, you have people who are programming because they love to program. This is not a job, it's a hobby. As a wise man once said. "People are defined by what they do in their spare time, not by what they do in their job." There are no nine to fivers in the open source community.

    Interaction in a team environment with the nine to fivers causes more frustration than satisfaction.

    -bk
  • Shut up. Posting something 8000 times doesn't help the actual content of the post. Just calm down a little and stop smoking so much crack. Seriously, it will help.
  • I can't stand PDF, couldn't they have put in html or something. After all, this IS the internet. I know this is off topic, but I really dislike Adobe and a lot of their tactics. David
  • Um, yes, I do know what Open Source means.

    I was trying to make the point that calling a flubbed HTML code "evidence" that Open Source has poor quality control is a bit disingenuous. The fact of the matter is, this page (a) is generated on the fly; (b) not Earth-shattering, even to the people at slashdot; and (c) not bankrolled by more money than God.

    People seemed to think it was reprehensible that Cmd Taco publish something with one glitch. Personally, if that one glitch is evidence that Open Source won't work, then 63,000 glitches is surely evidence that corporate management fails. Would I make that syllogism? No. But then, I didn't say that a glitch in one page is enough to condemn the Open Source movement.

    Since the fix is a measure of responsiveness to user feedback, I think the speed is relevant. I visit many sites with similiar dysfunctional HTML -- many such sites being commercial sites -- and I usually do whip off an email to the webmaster. Often, the sites stays broken for a day or more. So again, if this page is somehow to serve as the effigy and avatar of the Open Source movement, then I think it actually speaks well of the movement.

    Total aside: I seriously doubt that Windows 2000 is the most complex software project ever undertaken, but I'd be fascinated if someone could point the way toward a meaningful assessment of such a question.

  • I'm pleased to see him mention Debian's Social Contract and the Debian Free Software Guidelines as well. It's a shame though, that he compeletly misunderstands these documents, and depicts them as free software licenses. The DFSG is not a license, it is a way to categorize licenses.

    --
  • This is an oft-recited mantra, but it makes little sense. Either you have a license to load the program and use it or you don't.

    So either we are all pirates, or there is at least an implied license to use. That license is found in the right to make copies in Section 1.

    Actually, Section 0 states:
    Activities other than copying, distribution and modification are not covered by this License; they are outside its scope. The act of running the Program is not restricted...

    PS. I've been made aware of my error in using Sendmail as a GPL use example. Sendmail is under a BSD-style license.
  • So Open Source is connected with the repression of free speech and the mass murder of millions. Idiot.

    He didn't in any way infer such a connection. Read Marx ( start with the communist manifesto ), then talk. In particular, tell us where it says anything about "repression of free speech" and "mass murder of millions".

    Oh, by the way, none of this alters the fact that he is completely wrong.

  • So Open Source is connected with the repression of free speech and the mass murder of millions. Idiot.

    He didn't in any way infer such a connection. Read Marx ( start with the communist manifesto ), then talk. In particular, tell us where it says anything about "repression of free speech" and "mass murder of millions".

    Oh, by the way, none of this alters the fact that he is completely wrong. The Open Source philosophy works just fine for capitalists, democrats, republicans, communists, and socialists.

  • Comment removed based on user account deletion
  • Comment removed based on user account deletion
  • If Open Source remains purely voluntary, I will be unabashedly in favor of it. But I dread the day when the religion of free software succeeds in removing the choice of free software and making it mandatory.
  • As the open source software world slides further into the capitalist world, where immediate material gain is the pinnacle of sucess, it will find money vs. community gain to be a compelling battle.

    An economist will make no distinction between the value of money and the value of community. They are both equally valuable. Economically speaking, someone who forgoes a $90,000 job for the pleasure of working with the community as a volunteer receives just as much value and contributes just as much to society. But before you start condemning the capitalists, the converse is equally true. It is one's individual choices that matter.

    Economists seem to focus more on money for two reasons. First, money is easier to measure than good will or community spirit. Second, most people confuse financial analysts who call themselves economists with actual economists, and thus think economics is all about financing and interest rates and stuff.
  • you're confusing reproduction with distribution. The GPL only comes into power when you distribute.
    And that's outside the organization, so RAM doesn't count.

    --
    ba-bu-ba-ba-baaa, da-da-dum. Re-boot the ser-ver.
    ba-bu-ba-ba-baaa, da-da-dum. Re-boot the ser-ver.
  • Unexpected? Yeah, I'll buy that. But don't confuse "unexpected" with "unintentional". Lots of people worked their asses off getting those IPOs together and they didn't do it for the community.

    -----------

    "You can't shake the Devil's hand and say you're only kidding."

  • you're confusing reproduction with distribution. The GPL only comes into power when you distribute.
    And that's outside the organization, so RAM doesn't count.

    Your confusing my remarks entirely. To summarize:

    (i) use is reproduction (that's the legal word for copying) under the Act; (ii) if you aren't licensed to use it under GPL, then in the absence of some other license, the use is unconsented, and we are all pirates.

    Thus, the authority to make copies and derivative works for ourselves under Section 1 is a right to use. If it isn't, then shut down your computer, for your use is otherwise unlicensed.
  • The act of running the Program is not restricted

    And therefore licensed.
  • Congress *did* change the law

    Paraphrasing from "The Princess Bride," these words don't mean what you think they mean.

    No, all these cases followed adoption of Section 117. Interesting enough, Section 117 was recently changed to change the result in MAI, but solely in the context of running an operating system for the purpose of repairing a machine. Having made the change in this limited context, Congress ultimately ratified the 9th Circuit's holdings outside that context.

    EVERY ONE OF THE CASES CITED earlier distinguished Section 117(a), primarily on grounds of the definition of "essential step" and on grounds of the definition of "owner."

    I'm pleased to engage anyone on this subject in detail offline if you like. But this is a well-settled area of law, and you need to know a lot more before you start quoting excerpts out of context. In these cases, the Courts concluded that possession of a copy does not, by itself, constitute ownership of that copy. Section 117 raises issues of course, but on balance, it doesn't change the analysis substantially.
  • Feel free to continue believing this is the case. Call a lawyer before relying on it.
  • The GPL doesn't preclude your USE of software. I can use Sendmail without having to give source to anyone. But if I modify and plan to distribute that modification, then the GPL comes into play.

    This is an oft-recited mantra, but it makes little sense. Either you have a license to load the program and use it or you don't. Under U.S. Copyright law, at least, unauthorized loading and execution of a computer program constitutes a "reproduction" under Section 106 of the Copyright Act.

    So either we are all pirates, or there is at least an implied license to use. That license is found in the right to make copies in Section 1. (If it weren't there, you COULDN'T use the code.)

    I understand that "common wisdom" is that the GPL does not "restrict" use. But mantras don't make law or legal relationships, licenses do. GPL either grants a right to use, or it does not. If it does not, its time to put away your copy of Linux until you get written permission to use it.
  • If your position is correct (that is not clear without talking to someone qualified) then you could be right about trouble with GPL in the us. I sure hope the free world doesn't take it's lead from the collective incompetence that is IP law in the states....

    Actually, I wasn't arguing that we are in trouble. The GPL expressly grants a right to make verbatim copies. No problem.

    We'd only be in trouble to the extent that the ideological lockstep legal analysis proffered in the root remarks of this thread were correct. Since they are not, we are not in trouble.
  • After doing some more research I must reluctantly conclude that you are correct with regard to Section 117.

    I actually wish it weren't so. It doesn't make much difference for the GPL (only in the way some people like to think about it, since GPL *DOES* grant a right to copy), but it makes a great deal of difference in most commercial software cases.

    I seems that Section 117 has been vitiated by some rather questionable reasoning in the courts. It is quite remarkable that Congress has not corrected this rather drastic rewriting of the copyright law in the courts, but it has not.

    Significantly, the Congress took up precisely this question along with the DMCA, and opted not to change "owner" to "owner or licensee," but instead limited the change to use of software in connection with the repair of computer hardware. Thus, the holding of MAI on its facts was reversed without changing the impact of MAI for the rest of us.

    Thanks for your courtesies. Best, A
  • Another poster hinted at this, but I feel like the authors walked into a slaughterhouse and can't smell the stink. Besides stature in the community, there is another benefit to Open Source programming - it's fun. It is widely beleived that most "new" economic theories will come from the crossover of economics to psyhchology, so the authors would be wise to consult (the now-popularized) psych. concept of "flow" and how this relates to Open Source programming.
  • It's a cross-platform and solid-formating, small file size, rich document model, portable and self contained, direct-to-printer format.

    I agree that HTML should be used where and when ever posible, but PDF fills a need nothing else can.
    _________________________

  • *sigh*
    Do you even know what Open Source means?
    Here's a definition [opensource.org] to help you on your way.
    Now to comment on your post first of all you just compared a mis-typed character in an HTML page to what is probably the biggest software engineering project ever embarked upon. What kind of comparison can you make with this that doesn't make you sound like an illogical, fanatical, anti-Microsoft, Open Source apologist?
    Secondly Cmdr Taco viewing the bad HTML page in his browser, opening a text editor and changing it in the time it took you to reload your page has NOTHING to do with Open Source. After all I've never such bad HTML on any corporate website, does this suddenly mean that corporate software development practices are somehow better than Open Source ones?
    Please think before you post next time, posts like this are why lots of people refuse to take Open Source and linux in particular seriously when people like you project yourselves as our advocates
  • I used to ask Malda to implement a java servel/java 1.2 client codebase

    Sounds like he had the common sense to flatly refuse wasting his time on moronic and inefficient approaches to web publishing.

    Without a doubt, every advocate of servelets I have met works on a site that gets fewer than 100k hits a day. News flash folks - you can serve 100k hits with smoke signals...which is about the capacity of servelets in any case.

  • Oh, yes. I'm glad that the commercial houses never release a product that has a glitch. All that time and money spent on management sure pays off ... wait? What's that about a release of a major commercial OS with an estimated 63,000 known bugs in it?

    The issue is not bug-avoidance -- which is essentially an impossible dream -- but bug-correction. In the Open Source world that happens much more quickly than in Corporate Drone Land. Point in fact -- this messed-up page was corrected in the time it took me to hit "reload". That would never happen in the land of corporate drones.

  • Last year, I put together a model upon a more focused question which is "What is the impact that mature open-source products on an existing commercial market"?

    This is a more technical question of market dynamics (where I think that I can contribute something) rather than an explanation of the motivating factors associated with the open-source development process.

    Although this particular question sidesteps many of the interesting issues (of why folks would want to do this), it makes a mathematical model of the dynamics plausible.

    Since the model is framed in the context of autonomous agents, I presented it at an AI conference workshop (GECCO-99). However, I wouldn't mind getting some thoughtful feedback from folks who are more interested in the open-source economics side of things, rather than in the autonomous agents.

    A web-page on the paper is located here [mit.edu], and the paper itself can be found here [mit.edu] (PDF format).

  • Comment removed (Score:3)

    by account_deleted ( 4530225 ) on Saturday March 18, 2000 @09:14AM (#1193735)
    Comment removed based on user account deletion

  • Open Source and Economics (Score:3)

    by Arandir ( 19206 ) on Saturday March 18, 2000 @03:53PM (#1193736) Homepage Journal
    Because of the current economic disinformation campaign (also known as the US Presidential Primaries), I decided to hit some of my old econ texts. I was quickly reminded that all of economics can be distilled down into a few simple premises (with extremely complex interactions).

    One of the basic premises is that a voluntary transaction will not occur unless both sides benefit. Apache itself is not sold but can be obtained for gratuis because it is worth more for the Apache developers not to charge for it. If it were worth more to them to sell it for $50 a copy, they would do so. Since they don't, it isn't.

    The reason that it is more valuable for the Apache developers to release Apache at zero monetary cost is because they receive other values for it instead. Another economic principle is that costs and benefits are not limited to money. The various benefits that ESR lists for open sourcing a project, ego stroking, good will, sense of community, etc., are just as economically valuable as monetary payments.

    The last point is very important, and one that some people completely over look. If money is less valuable than "good will" or any other ephemeral payment, then neither are these ephemeral payments any more valuable than money. We cannot economically judge one to be more important than another. Thus, to deny a corporation the ability to sell its software products, which is what it wants to do, is to create an economic loss for society. Note that when I say "sell software", I mean in the copyright sense of selling rights to it or selling undistributable copies.

    Some folks in this community hold that it is morally wrong to sell software (see above note). Economics cannot judge whether something is moral or not, that is left to other professions. But it can say whether some policy creates losses in value to individuals or society. A mandatory open sourcing of software (whether by government decree or societal pressure) will create net economic losses. Economically, it will become more valuable to a developer to forgo opportunities to create new software in favor of waiting on tables instead.

  • Use of Software, Legally Speaking, *IS* Copying (Score:4)

    by werdna ( 39029 ) on Saturday March 18, 2000 @10:40AM (#1193737) Journal
    Actually the 'common wisdom' seems to be correct here. You have to remember that GPL operates under *copyright* law. Thus if you are not copying ... there is no restriction at all

    That's the thing about common wisdom, it *ALWAYS* seems correct. It this case, however, it is not even a close question:

    If you use computer software, you are copying. Three Circuit Courts cases have held unequivocally (MAI, Southeastern and Apple) that the loading of a computer program from any media into RAM, and the subsequent execution of that program from constitutes reproduction under 35 U.S.C. s. 106. There exists no cases holding to the contrary.

    Until the Congress changes the law, or the Supreme Court opines otherwise, unlicensed use of software constitutes Copyright infringement.

    Copyright has nothing to say about what you do with your copies

    You can't imagine how badly mistaken is this view. Copyright law provides specific exclusive rights in Section 106. Unless you are granted consent, or can find an exception in sections 107 through 120, you are infringing. This is true even if you are the owner of a copy. Ownership of a copy (which is distinct from possession of a copy) does grant certain rights set forth in Section 109 and in the case of software section 117 of the Act. Neither provides a general right to reproduce, and hence, to use, the software.

    The common wisdom, as you have stated it, is clearly in error. See a lawyer before you rely on it.

  • URL for legitimate free download (Score:5)

    by bugger ( 101595 ) on Saturday March 18, 2000 @09:37AM (#1193738)

    Try

    http://www.people.hbs.edu/jlerner/publications.h tml

    for the free download of the working paper.

Slashdot Top Deals

Perfection is acheived only on the point of collapse. - C. N. Parkinson

Close