Home > Economics - General > Money Ruins Everything

Money Ruins Everything

May 17th, 2008

Dan Hunter and I have a paper coming out in the Hastings Communications and Entertainment Law Journal, which economic and technical innovation is increasingly based on developments that don’t rely on economic incentive or public provision. The main examples, obvious enough for readers here, include open source software, blogs and associated technical and social innovations, and wikis. Abstract and links to SSRN over the fold.

Paper is at SSRN (not paywalled, I hope)

Abstract:
In the economy of the 21st century, economic and technical innovation is increasingly based on developments that don’t rely on economic incentive or public provision. Unlike 20th century innovation, the most important developments in innovation have been driven not by research funded by governments or developed by corporations but by the collaborative interactions of individuals. In most cases, this modality of innovation has not been motivated by economic concerns or the prospect of profit. This raises the possibility of a world in which some of the sectors of the economy particularly the ones dealing with innovation and creativity are driven by social interactions of various kinds, rather than by profit-oriented investment. This article examines the development of this amateur modality of creative production, and explains how it came to exist. It then deals with why this modality is different from and potentially inconsistent with the typical modalities of production that are at the heart of modern views of innovation policy. It provides a number of policy prescriptions that should be used by governments to recognize the significance of amateur innovation, and to further the development of amateur productivity.

Categories: Economics - General Tags:
  1. El Mono
    May 17th, 2008 at 15:28 | #1

    Do you think that the rise of amature contributors to local arts and culture may make some of the assitance and protection provided to professionals nnessecary?

  2. Brendan
    May 17th, 2008 at 15:48 | #2

    In think an additional motive for choosing the GPL software license may be lack of trust between parties in a software development project. Better the GPL than possibly loosing all rights to the work in a dispute. The alturistic outcome is the a side effect.

  3. rog
    May 17th, 2008 at 16:24 | #3

    It’s an interesting phenomenon, one in which technology has enabled equality and the rise of peer-2-peer structures which Friedman addresses in “world is flat”.

    Countries or societies with vertical integrated structures, like Japan, may have problems if they dont adapt.

  4. Fmark
    May 17th, 2008 at 18:16 | #4

    It’s important to note in the case of GPL software, much of the work is done by paid employees of corporations. For example, the linux kernel which started as a hobby project but is now one of the worlds largest, most dynamic software projects has between 70% and 85% of its work done by professionals in a paid capacity (although clearly 15-30% is still significant).

    See http://www.linux-foundation.org/publications/linuxkerneldevelopment.php

  5. jquiggin
    May 17th, 2008 at 19:58 | #5

    To be sure, corporations have taken up open source software, blogs, wikis and all sorts of other innovations, just as they took up the Internet in the boom of the late 90s. And they are certainly doing some good work in developing these innovations. But they are following the lead of amateurs in all these areas.

  6. Brendan
    May 17th, 2008 at 22:10 | #6

    As a further thought and given point #5, quite a few of the innovative ‘amateurs’ are experts associated with universities either as academics or students. For these I think it is reasonable to consider them as making an unremunerated vocational (rather than avocational) contibution, but also possible to see it as being a weak type of ‘public provision’.

  7. John Mashey
    May 18th, 2008 at 05:20 | #7

    I’ll work up a serious discussion, but first a quick thought exercise for people:

    1) In computing, how long ago did individuals start writing/designing software/systems, distributing it to others (or sharing in a common pool), without getting paid in $$ for the transactions? [I have 2 particular key items in mind, but maybe someone can suggest an earlier one].

    2) When did universities support such efforts?

    3) When did corporations?

    4) When did governments?

    5) Which of the following are/were “amateurs”?
    (just to pick a few I know)
    Eric Allman
    Vint Cerf
    Charles Forney
    Brian Kernighan
    John Lions [to represent Oz]
    Ken Thompson
    David Wheeler

    (put another way, is “amateur” the right term with regard to software?)

    6) Was Linux started as an innovative operating system? Can you compare it to any previous OS’s, especially one whose name doesn’t appear even once in John’s article?

    7) Without looking them up, what are SHARE? DECUS? STUG? USENIX? ALGOLW? ASSIST? PL/C? SPASM? WATFIV? SNOBOL? HASP? MTS? MAD?

  8. jquiggin
    May 18th, 2008 at 06:49 | #8

    John M, we didn’t cover much of the earlier history of software (most obviously Unix) for a variety of reasons, including the fact that we don’t know that much about it (on your list at #7 I’m confident only of ALGOL and SNOBOL). The other reason is that discussion of these issues has (in my opinion, anyway) been focused too much on open-source software, which is just one example, and can be misleading (see our discussion of Raymond).

    It’s clear that the norms of sharing that are explicit in open source software go back a long way before that, and I’d be interested in your promised longer reply.

  9. TerjeP (say tay-a)
    May 18th, 2008 at 23:22 | #9

    Surely open source software and wikipedia are just specialised cases of civil society in action. And civil society has existed for a very long time. I recall reading about the history of the first hospital in Alice Springs which was neither a government venture nor a for profit business. Lots of similar examples abound in every community. Surf life saving organisations, volunteer bush fire brigades and bush regeneration initiatives are all classic examples. Perhaps none of these institutions are innovators in terms of generating new technology but they certainly demonstrate that profit and government (ie carrot and stick) are not the only mode of operation in which significant and important things can get done.

  10. May 19th, 2008 at 01:01 | #10

    Terje,
    Just one disagreement – I would say that all of those have from time to time generated new technologies. The SLS clubs have innovated ceaselessly to make rescues better and have worked well to improve initial response medical procedures. The Vollies have done significant work in better fighting bushfires and I would guess, so have the bush regeneration people.

  11. Phil
    May 19th, 2008 at 02:01 | #11

    Point 9. seems to hit the nail on the head. This is being discussed as if it is a “new” phenomenon, but we only have to look at the history of sport – an activity that underlines much of Australia’s shared community life – to witness the significant contribution of “amateurs”.

    Terje’s examples of RFS and SLSC are also prime examples, and I’d argue that many have provided new technology, especially in lifesaving and firefighting (mainly from learnt lessons).

    On top of all this we can also witness foodbanks, theatre, 99 percent of the music, fiction writing. There is also the great deal of ‘experimental’ film and music activity that happens below the radar – to the point that an entire genre of music is called Garage, because that’s where the innovation happens.

    IMHE profit is a very poor motive for many people, almost perceived as a sociopathic activity in the community I live in (country town NSW).

    In some ways people with “careers” don’t get it – life is the real stuff we do; sport, hobbies and family; work is just how we pay for it all.

    No one ever wants to admit it but the greatest creativity and R&D funding in this country is the dole and Austudy. It’s the people on that money that are doing the real innovation, everything else is just marketing.

  12. May 19th, 2008 at 06:58 | #12

    John – I look forward to reading this with interest – *great* title!

  13. John Mashey
    May 20th, 2008 at 02:26 | #13

    TERMINOLOGY
    A further discussion is needed of the right words and taxonomies of “amateur”, “professional”, “volunteer”, “paid/unpaid”, “government”, “corporate”, especially as applied to computer software, where there is a very complex history. I wouldn’t yet offer a comment on the extent to which software is alike/unalike blogs, Wikipedia, etc.

    OPEN SOURCE SOFTWARE GOES BACK TO THE BEGINNINGS OF SOFTWARE
    “open source software” is just the most recent label attached to a behavior pattern that’s existed essentially from the beginning of software, assuming the pattern is:

    a) People contribute source code or designs or effort (sometimes work on standards committees is more important than writing the code) to a common pool, or provide it “free” to others, with no specific remuneration from that transaction.

    b) This has long been supported by:

    - governments, including research labs & funders
    - universities
    - corporations
    - associations ranging from official long-term bodies down to informal ad hoc groups that may never even meet, or whose work covers a short time
    - individuals, of whom many would be hard to label “amateurs”, as they were some of the world’s top computer scientists / software engineers. Some undergraduates (amateurs?) have made real contributions over the years, and by now, there might be high school students (although I haven’t ever managed any of the latter).

    Some kinds of software work very well as volunteer / open source efforts. Others simply do not happen except by paid commercial efforts over long periods.

    I conjecture that the interactions might be more complex than for some other forms of content creation. Here’s a short set of answers to the questions I posed in #7.

    1) In computing, how long ago did individuals start writing/designing software/systems, distributing it to others (or sharing in a common pool), without getting paid in $$ for the transactions?

    I can track that back to David Wheeler’s creation of the subroutine idea at Cambridge for the EDSAC, around 1948. Others started contributing subroutines to a common pool, and “user-contributed� software has been a typical category in mainframe computer centers ever since, especially at universities and R&D labs.

    Alternatively, around 1952, John von Neumann gave away plans for the Princeton IAS computer, and over a dozen were built, with variations. RAND’s JOHNNIAC variant was cloned by Iowa State University.

    2) When did universities support such efforts?

    From the above, at least since 1948.

    3) When did corporations?
    Informally, during the 1950s, when operating systems, languages and tools were commonly written outside computer companies, traded around. Computer companies often encouraged this and supported the efforts financially. IBM’s SHARE user group started in 1959, and Digital Equipment’s DECUS in 1961.

    4) When did governments?

    Governments have long funded computing efforts, back into the 1940s, sometimes by specific contracts, but quite often by supporting/funding “volunteer� efforts. COBOL got started that way (1959). The US Government’s role in making the Internet happen is probably stronger than most people credit it for, as not only was there direct funding by ARPA/DARPA for obvious Internet work, but the Berkeley 4BSD UNIX work, from which much Internet software is descended, was DARPA-funded.

    5) Which of the following are/were “amateurs�?
    (just to pick a few I know)
    Eric Allman
    Vint Cerf
    Charles Forney
    Brian Kernighan
    John Lions [to represent Oz]
    Ken Thompson
    David Wheeler
    (put another way, is “amateur� the right term with regard to software?)

    No computer professional would *ever* label any of these as amateurs, even though much of their work wasn’t done commerically, at least not directly.

    Eric Allman would be the closest, as he wrote sendmail when he was a graduate student, but he was already a very good software engineer.

    Vint Cerf was a Stanford professor when he and Bob Kahn designed the Intenet’s TCP/IP, and then later moved to DARPA. I’ve lost the 2000 email I originally received, but they had interesting words about Al Gore’s role.

    Chuck Forney was the Assistant Computer Center Director at Penn State. In the mid-1960s, engineers at IBM Houston had written “HASP” which was distributed as source code. User sites loved and modified it, and there was a SHARE user group project (run by Chuck) that gathered and selected improvements. This was supported by IBM, who even bought the beer. They even had their own songs, including “Should old Chuck Forney be forgot, and HASP songs sung no more.â€? No user ever made any money from their contributions, but this was quite similar to the rationale for Apache: people needed something, and it was easier to work together.

    Ken Thompson was the original father of UNIX, which was always “open-source� within Bell Labs, and Ken (and his closest Co-worked Dennis Ritchie) did his best to make it so outside as well, although sometimes over Bell Labs’ lawyers’ dead bodies, since AT&T was not allowed to be “in the computer business.�

    There was a concerted, successful 1970s effort to get UNIX source into universities, essentially free. Ken did a sabbatical visiting professor period at his alma mater, UC Berkeley, around 1976. That connection is one of the reasons Berkeley has long been so involved with UNIX, and there were many informal interactions. Berkeley UNIX had much work done by grad students, but with some long-time experts (like Bob Fabry) involved. Much of the work (especially 4.xBSD) was paid for by DARPA to be able to spread Internet-enabled software around.

    Brian Kernighan was one of Ken’s colleagues at Bell Labs. Among many things, he wrote the “Software Tools� books, which essentially made UNIX tools available on non-UNIX systems, and led to the Software Tools Users Group (STUG), which enthusiastically contributed over many years.

    John Lions was a beloved Australian computer scientist at UNSW, with long connections to Bell Labs and UNIX. He is most famous for writing “Commentary on UNIX 6th Edition, with Source Code�, in which he printed the 6th Edition kernel source, accompanied by line-by-line commentary, an invaluable teaching aid, widely distributed, often by informal copies. Although this mortally offended Bell Labs lawyers, it was encouraged by Ken and Dennis and others, and we actually used it for internal Bell Labs classes, albeit without telling the lawyers. John was a scholar and a gentleman, and I remain happy about <a href=� http://www.usenix.org/publications/login/1999-4/award.html�John Lions Award and my car’s $6000 license plate.

    6) Was Linux started as an innovative operating system? Can you compare it to any previous OS’s, especially one whose name doesn’t appear even once in John’s article?

    Linux of course started as a license-free implementation of UNIX, compatible enough that it runs programs from the 1970s. Like most useful software, it has some innovation, and in this case, reimplementation of long-existing functionality, or in some cases, incorporation of existing code. In some sense, the biggest innovation is in the Internet-adapted development process and social structure, although some of that mirrors the exact way that the HASP project worked, or to some extent, UNIX inside Bell Labs.

    I.e., someone creates software, others get interested and contribute, but there is a central core group that exercises architectural and editorial control. I.e., this is what Linus Torvalds and his “lieutenants� have long done for the Linux kernel. It’s been a long time since Linus could be called an amateur.

    7) Without looking them up, what are SHARE? DECUS? STUG? USENIX? ALGOLW? ASSIST? PL/C? SPASM? WATFIV? SNOBOL? HASP? MTS? MAD?
    SHARE and DECUS are/were the IBM and Digital Equipment user groups. STUG was the Software Tools User’s Group. USENIX is the UNIX Association. All of these had mechanisms for encouraging people to trade code.
    ALGOLW, ASSIST, SPASM, WATFIV, SNOBOL, and MAD were all 1960s/1970s software packages created by universities (or Bell Labs for SNOBOL) whose source code was widely spread around universities, at minimal or no cost. HASP (late 1960s)was explained above, as an IBM/user-group joint effort. MTS, The Michigan Terminal System, was an operating system for IBM mainframes, with source and work shared among the universities who used it.

    CONCLUSION
    Computer software has long had a tradition of open sharing, encouraged by universities, governments, and even corporations. Even for-profit software companies sometimes get involved in when that makes sense.

    Over 6 decades, as computers have become more accessible, the pool of potential contributors has expanded greatly, which is all to the good.

    Still, most software that most people use is written by professionals, regardless of how they get paid or why they’re doing it.

    The goodness of the current low bar to contribution is that anyone can try, so you don’t have to work for IBM or Microsoft to have your software used and get recognized for it.

    The badness is that there is even more crap, <a href=� http://en.wikipedia.org/wiki/Sturgeon's_law�Sturgeon’s Law still applies.

  14. Andrew
    May 20th, 2008 at 12:11 | #14

    Great comments John Mashey – very interesting

  15. May 21st, 2008 at 11:43 | #15
  16. glenn
Comments are closed.