Re: Rethinking Licensing (was Re: [ba-ohs-talk] open source +patents = puzzled)
Jack Park wrote:
> So, I'm going to go public with some of my thinking along these
> lines. [snip] (01)
Jack- (02)
Thanks for the insight into responses vs. answers. :-) I agree, and I
don't think there are any easy answers, just hard choices. (03)
=== on micropayments === (04)
I like your effort towards rethinking. I especially like your question
at the end regarding the BA value proposition -- that is a good question
to always revisit periodically for any effort expecting support. (05)
However, sorry, I don't think micropayments are the answer. See:
"The Case Against Micropayments"
http://www.openp2p.com/pub/a/p2p/2000/12/19/micropayments.html
The article discusses several reasons, the biggest one is that the user
cognitive overhead of dealing with micropayments is too high. A
transaction might cost a fraction of a cent, but it costs you time and
attention to think about it (even to decide whether to click or not,
knowing you need to pay), and that makes every micropayment
fundamentally expensive. The article also suggests several old
alternatives (Aggregation, Subscription, and Subsidy) to micropayments. (06)
While you may be able to come up with a proprietary license which
successfully implements a MLM scheme, I don't think anything you specify
would ever be called "open source":
http://www.opensource.org/docs/definition.html
or "free software":
http://www.gnu.org/philosophy/free-sw.html (07)
Another way to think of it is, have you ever had to get a large legal
department to approve a software license? It can take weeks, even months
to have a new one reviewed (even if it might only take an hour or so of
a lawyer's actual time). Your new license will have to reviewed before
people can use your work, perhaps effectively in practice rendering your
work under it unusable. (08)
=== on choice of licensing === (09)
Realistically there are three free licensing alternatives for the OHS:
BSD, GPL, or something in the middle (Apache seems fairly close to the
BSD end). I think things in the middle are usually too hard to think
about in this case for too few benefits and there are too many choices,
so I'd pick one end or the other, if for no other reason than wanting a
license compatible with the GPL (which the BSD license now is).
Essentially, for quicker adoption of standards and infrastructure at
this point I might pick BSD-ish (e.g. Python) [if I don't fear Microsoft
"embrace and extend"],
http://www.gnu.org/philosophy/gpl-american-way.html
for applications I'd tend to pick GPL (e.g. a garden simulator). As I
see OHS as more of an application at this point, GPL works for me for
it. (Others might see it more as infrastructure to be rapidly adopted
and might come to a different conclusion.) In any case, there can and
will be multiple efforts under multiple licenses, although some perhaps
may be more blessed by Bootstrap than others. I think that is the best
way to look at it -- however "permission to use" indirectly prevents
that from happening, by privileging Stanford and BI unduly and
discouraging voluntary contributions of code for liability reasons. (010)
== a creative question === (011)
You obviously have a lot of creativity to deploy along this financial
viability line. Perhaps for fun you could try working from this "extreme
case" assumption:
imagine I had delivered yesterday an OHS implementation that was GPL'd
and was built to this specification:
http://www.bootstrap.org/alliance-980.htm
How could you or others make money off of it (respecting the GPL
license)?
[And if you feel the need to keep the answers proprietary, at least
share the count of them or the likely revenue $$$...] (012)
If you can come up with enough plausible positive answers to that
question of how to make money off of a GPL'd OHS system, then why not
then GPL the OHS during the development phase, and dispense with the
over-reaching "permission to use" for contributors? How much more value
is a non-GPL'd OHS really providing you? And at what cost -- as there's
possibly a lot of GPL-leaning developers working under Linux who might
like to have a connection to a source of wisdom like Doug. (013)
=== maybe you just wont make much money at it === (014)
As for your specific project ideas, while technically they sound
fantastic, sorry maybe it won't make money for you directly to publish
under the GPL. (015)
However, to be realistic, even publishing it in a book would probably
not make more than the $10K or so typical advance for technical books.
Most books seem to be authored to give the writer credibility in the
field which translates to consulting revenues, grant funding, or
employment. And dealing with book publishers and their "standard"
contracts is a pain:
http://www.mediachannel.org/views/oped/bookcontract.shtml (016)
Sorry, at this point, there is so much free code out there, and so much
commercial (and free :-) stuff is buggy or incomplete garbage, I
probably wouldn't bother buying a book just to use someone's servlets
code. I buy books generally when I'm already using the code. In fact,
I've stopped considering code in books as a valuable feature when I
realized I can't typically use it in an arbitrary project -- so I could
never make a proprietary project with a snippet of code from a typical
programming book open source down the road, and I can't use it in
consulting either typically. By the way, ask Jason Hunter if he's made
over $100K from his book (enough return to think about putting a year
into developing software) -- I bet the answer is no. So all he may have
done is ensure most people can't use his code. What a loss. Maybe if I
had seen his code in use, and started using it myself, I might have
bought his book, like I have bought books on Python after using it some,
Linux after using it some, CVS after using it some, HTML after using it
some, C after learning it using a borrowed book, and so on. Perhaps
Jason Hunter is a special case and his effort is profitable, but then I
would bet his publisher invested heavily in him, and he has personal
attributes and a drive that would have made whatever route he took a
success. (017)
Also, ask yourself, if you are offering code for money, will you feel
compelled to answer your support email every day? My wife and I sell
some software through the shareware marketing method, and even though it
brings in at most a few hundreds of dollars a month (in a good month)
which is a very good return for shareware, for this we have to pay the
price of checking email every day to handle orders, and guilt if we
don't answer support emails right away. Plus we need to look at all
sorts of spam (such as without titles or with vague headings) because
people sometimes ask for support with not very good email subject
headers. Plus you have an up and down emotional rollercoaster for all
sorts of other reasons. In our case, for our most financially successful
product, a "non-profit" competitor tried to grab our trademark and
registered a domain with our product name (costing us thousands in legal
fees out of honor more than feasible profit from winning, which we did).
Clearly the emotional and time costs to us of running a small software
business have turned out to be much more than any money we can hope to
make from it at this point. And all this is exclusive of time spent on
the product itself. Yet, we have had every intention of making a profit
on it, and we have continued to sell the product perhaps partly in hope
of striking it rich in one big swoop (like when a major software company
called to talk with us, and then unfortunately reorganized) or thinking
the next version might be the one that sells 20X better. So, when you
add up all these costs vs. benefits, there may not be much profit in
your venture as a proprietary thing when you look at it as an entire
business venture. (018)
Essentially, selling software is rarely extremely profitable anymore,
and probably never was except for some lucky few business application
specialists. Now rare can still mean thousands of success stories that
can be pointed to (out of millions of tries), but the odds are
nonetheless wildly against you. On the other hand, people have paid lots
to us on a regular and repeatable basis for focusing our attention on
their specific issues and needs, just like people pay lawyers and
doctors a lot for the same privilege. And, there are people making big
dollars working on software like our successful projects -- they are
just working as techies at Pixar or LucasFilms or DreamWorks and the
works as used on specific films to achieve specific emotional effects --
so essentially these people are making money from consulting (I doubt
the proprietary software holdings of these production companies will
ever produce much revenue for them compared to the investment though --
but it doesn't matter because the money is made on the film.) (019)
One reason software doesn't make so much money these days is there is
really too much of it. I've often though the software world would be
better off if 90% of people left the profession (since most programmers
mainly make work for each other, and create painful code to maintain --
see "the mythical man month" on communications overhead etc.).
Programming is a world where there are over 100X orders of magnitude in
individual productivities (and where individuals can also easily have
negative productivity in non-obvious ways) and where many or most
projects fail in a significant way (compared to say bridge building).
And how many of each type of application does one really need? HOw many
spreadsheets or wordprocessors or OHS systems? One hopes to make money
on the new thing, but how many new things can there be long term? What
we need is less code, not more. Less versions of Windows and Office, not
more. My biggest beef against most systems I look at is they are just
way too bloated and my first desire is to make them lighter. Think about
lawn care companies -- most of the money is in cutting grass, not
putting it in. If lawn care companies got judged for height of grass
(like programmers getting judged for lines of code) we'd all be living
in jungles. Well, you know what, we are living in a software jungle. :-)
And maybe the money is in selling and using machetes. That's not to say
I don't think new code isn't needed sometimes -- if for no other reason
than the license (proprietary) is not right for current needs. An
unfortunate corollary of this would be am implosion in tech stock market
values. (Oops, that happened already didn't it? Might be more to come
then for current companies -- unless like IBM they transition more to a
services model.) (020)
=== maybe a profit doesn't matter for this particular project === (021)
Philip Greenspun's take on making money on the web:
http://philip.greenspun.com/wtr/dead-trees/53002.htm
"Anyway, the response from my new acquaintances is invariably the same:
"How are you going to make money off your Web site?" ... If they knew
I'd splurged for a $5000 Viking stove, they wouldn't ask if I was going
to start charging my brunch guests $5 each. If I told them I dropped
$20,000 on a Dodge Caravan, they wouldn't ask if I was going to charge
my dog $10 for every trip. " (022)
=== maybe you already have or will make a profit === (023)
Another way to look at it is, the payment you get for free software and
content you write is all the other free software and content others
write. If you look at it on this basis, any investment spent making free
content and software is enormously repaid. The six person-years my wife
and I put into making a GPL'd garden simulator have been repaid a
thousand times over by getting GCC, Emacs, Apache, Python, and other
software and content (like this mailing list) and the whole world wide
web in return. (024)
You can also chalk up free software contributions to the same social
networking that happens when you volunteer at the local ACM or Rotary
chapter. (025)
You can also ask yourself (and your lawyer), how are lawyers paid so
well when the law is essentially public domain? Would we be better off
if laws were copyrighted by their proposers, judges decisions by the
judge, and evidence by the prosecution? Maybe then lawyers and judges
would be paid even better and justice would be quicker and fairer? (026)
One last comment, 90%+ of IT dollars are spent inhouse on requirements
gathering, analysis, testing, customization, security, and training.
(Don't have a link for that, but I've read it several places -- exact
percentage may be off, might be even 95%.) (027)
=== perspective rant on deeper issue behind licensing discussion ==== (028)
There's a fundamental unspoken assumption in most copyright and patent
discussions that:
A) content needs to be paid for to be created and is thus always a
financial investment, and
B) content needs to be paid for to be distributed, otherwise no one will
see it. Both of these assumptions are no longer true given the partial
success so far of the Unfinished Revolution. (029)
Here's my perspective. In the best of circumstances, if some vested
interest like RIAA give ground gracefully or elected official have
backbone like David Villanueva Nuñez of Peru,
http://www.theregister.co.uk/content/4/25157.html
we have the chance to transition to an age of abundance for many
material and digital things (there will may never be enough of some
social things like parental attention split among siblings). The issue
is, how do we make a peaceful transition from where we are now to the
worlds of material abundance described in James P. Hogan's novel "Voyage
from Yesteryear" or Theodore Sturgeon's short story "The Skills of
Xanadu" or in Neal Stephenson's (darker) novel "The Diamond Age"? And
then to make it personal, how do "I" make such a transition in my own
life? I feel it's OK to have one foot in both worlds for a while (like
doing some paid work and some free work) if you think that ultimately
the system is ratcheting itself up to a higher level of functioning as
long as we all slowly adapt as a society, just like environmental issues
are now becoming much more accepted slowly as an important consideration
in designing industrial processes. (030)
However, having said that, I think it is hard for specific projects to
have one foot in both worlds (like a license that is "sort of" free).
Usually I think it might be best to have two separate but perhaps
complementary projects if that is the intent. (031)
Alas, that is in the best of all possible worlds, and I think it quite
possible this will be a shooting war in a decade or two, just like the
drug war grew out of control into a shooting war driven by the perhaps
unintentional synching of the interests of large drug dealers (to reduce
competition) with the interests of drug policy enforcers (to keep their
jobs and gain promotion). Even if the drug war recedes, the fact that it
went on for decades is a tribute to these strange social forces
involving feedback driving the legal & political system to such negative
policies and to the point where, for example, U.S. judges criticize the
drug war policies in private but not public [see Reason article
mentioned below]. Again, James P. Hogan's novel has some good advice on
how to win such a shooting war over copyrights and patents (mainly by
showing people a better way and giving them better opportunities, one
person at a time). Just like the drug war has damaged unduly whole
segments of US society and promoted corruption and, worse, cynicism
among the police,
http://reason.com/0201/fe.ml.battlefield.shtml
the potential for abuses created by DMCA and other proposals is
enormous.
http://libertyboard.org/
http://www.eff.org/ (032)
>From the Reason article, just to see where these things go: "I was
involved as an expert witness in the Donald Carlson case, which was on
60 Minutes. In that case, a multi-agency task force, outfitted in
high-tech guerrilla gear, crashed into the home of a Fortune 500
executive and shot him down in his own living room on the basis of the
word of an uncorroborated informant. Nobody was penalized for it. In
fact, the people who did it were eventually promoted." Do we want to
work toward a world where this is likely to happen over a (paid)
informant tip off about illicitly copied proprietary OHS code? (033)
My point -- ultimately the only way copyright and patents can be
enforced to the extent they are being made into business models these
days is through the coercive use of police power such as Stalinist
Russia used to prevent "Samizdat",
making such "accidents" as befell Donald Carlson certainties, with
selective enforcement used to control dissent, and essentially
criminalizing much of human behavior down to ignoring TV advertising
when you go to the bathroom:
http://writ.news.findlaw.com/commentary/20020509_sprigman.html
>From a Richard Stallman interview:
http://www.memes.net/index.php3?request=displaypage&NodeID=650
> The U.S. though is not the first country to make a priority of this. The Soviet
> Union treated it as very important. There this unauthorized copying and
> re-distribution was known as Samizdat and to stamp it out, they developed a
> series of methods: First, guards watching every piece of copying equipment to
> check what people were copying to prevent forbidden copying. Second, harsh
> punishments for anyone caught doing forbidden copying. You could sent to
> Siberia. Third, soliciting informers, asking everyone to rat on their neighbors
> and co-workers to the information police. Fourth, collective responsibility –
> You! You’re going to watch that group! If I catch any of them doing forbidden
> copying, you are going to prison. So watch them hard. And, fifth, propaganda,
> starting in childhood to convince everyone that only a horrible enemy of the
> people would ever do this forbidden copying. ... The U.S. is using all of
> these measures now. (034)
The bottom line: these copyright and patents laws and enforcement are
growing so out of synch with how people view information and have
handled it traditionally through storytelling and sharing that such
draconian laws cannot ultimately be compatible with a just and fair
society. Overly restrictive laws promote disrespect for the laws as
happened with marijuana in the 1970s and 1980s and is happening now with
MP3s. See this speech written in 1841 by Thomas Babbington Macaulaon
which "became the basis of copyright policies in the English speaking
world for well over a hundred years":
http://www.kuro5hin.org/story/2002/4/25/1345/03329
Why should a public OHS effort done by volunteers be part of increasing
oppression? Make whatever compromise decisions you need to make as an
employee of a corporation to survive personally for the short term, but
as a person, as an individual, make some room for freedom for you and
the next generation too.
http://www.gnu.org/philosophy/right-to-read.html
Kids will have a tough enough time knowing that according to Moore's law
& Kurzweil & Moravec most current jobs might be filled by AI and robots
in 20-30 years time. At least let them have a free OHS to plan their
survival in style. (035)
== final comment === (036)
Doug drew us here via UnRevII to make an "open source" open
hyperdocument system (OHS) That's been clear from the start. That's what
we should make. That's what I want to make. People here can make other
proprietary things for whatever reasons, fine, but we should make at
least one open source OHS. (037)
-Paul Fernhout
Kurtz-Fernhout Software
=========================================================
Developers of custom software and educational simulations
Creators of the Garden with Insight(TM) garden simulator
http://www.kurtz-fernhout.com (038)