Subject: Re: Mandatory donations or build from CVS...
From: Keith Bostic <bostic@sleepycat.com>
Date: Fri, 12 Oct 2001 09:54:00 -0400 (EDT)

A couple of things I can tell you from experience:

+ Shareware doesn't work (and that's what donation for download is).
  I know several people/groups that have tried this, and it has paid
  off for exactly zero of them.  To paraphrase Mike Olson, "You can't
  afford to let your mortgage payments depend on the kindness of
  strangers."

+ The difference between "free" and "free, but please give me $5" is
  infinite.  War story: Sleepycat Software put up a "tell us what you're
  doing" HTML page in front of our download.  We asked 3 questions:
  name, email address and "how are you using the software?" You would
  not believe the abuse we took.  It was sufficient to make us remove
  the page.  And all we were doing was asking questions, not asking for
  money.

Now, on to what I believe -- a lot of very smart people have tried to
figure out how to make money from Open Source software.  (I want to make
it clear that I'm talking about the software itself, not other things
like consulting, or documentation, or proprietary add-ons, or whatever.)
The only ones that have been successful are the ones that have found a
way to retain IP rights in the software while still giving it away.

Everybody else trying has either found sponsors (read patrons: and
there's nothing wrong with that, it's how much of the world's great art
was done), or barely made enough money to buy beer.

> From: burton@relativity.yi.org (Kevin A. Burton)
>
> I think the Internet community needs to *seriously* get over the free
> ride they have been getting for soo long :)

This isn't a joke -- you're absolutely correct -- but I'm sure it's not
shocking to you to find out that the Internet community isn't interested
in changing its habits to make developers more money.

I believe the root problem is the Open Source community's refusal to
approve a license returning money to the copyright holder.  Even
something like "if you sell our software, you have to give us a portion
of your profits", has not been acceptable.  And, I think there are three
side-effects of this choice:

    1: Nobody makes money off of writing Open Source software (again,
       writing the software, not wrapping something around it)

    2: Many (most?) of the people investing the time require to build
       a large Open Source software package thrash for a considerable
       amount of time trying to figure out a way to make money...  but
       fail.

    3: Open Source software packages generally lack "quality factors"
       (for example, good documentation, and test suites).

Of course, these side-effects don't apply to everybody, or even equally
to different groups, but I believe they apply fairly widely.

There are components of the Open Source view (for example, access to
source code with the right to modify) that are separate from the right
to freely redistribute without compensation to the copyright holder,
and which are valuable in and of themselves.  We should acknowledge
there are reasonable levels of Open Source and adopt a big-tent policy.

All that said, I've wandered off into the weeds  -- this is not the Open
Source community's orthodoxy, and I understand and accept that. :-)

>> sell. http://www.theoretic.com/ransom . "Hold your code for Ransom"
>
> Yeah.  I personally don't like this model.  It seems too aggressive :(
> I have kicked this around in the past.  Even talked to RMS about it
> at length.  I would like to seem how some initial experiments turn out.

I'm interested in experiments too, but I believe the idea is flawed.
To me, the interesting sentence in the proposal is the following:

    The public gets fully open source software, and the programmers
    are fairly compensated for the real work they do, not the amount
    of "copies" they sell.

This assumes all programmers are the same and should be paid equally --
yes, you can make the initial price set by the programmer reflect the
programmers skill/experience, but that makes it difficult for the
consumer to determine a "fair" price to pay, and it gets around the
stated goal of being compensated for the "real work they do" -- if I'm
a great programmer, and I charge a billion dollars for a piece of
software, does it matter to anyone that I got the money up front instead
of a dollar a copy?  See also: old joke about an engineer charging a
company $100,000, and upon being asked for an itemized bill, responded:
$1 to chalk an `X' on failed component, $99,999 for knowing where to
chalk the `X'.

What ransom software does is load the entire cost of the software on
early adopters (which is the exact opposite of what you want to do,
since the early adopters are the people that debug it for you), and
practically, if not theoretically, bound the amount of money you can
make from software as early adopters won't pay that much individually.
Finally, developers need an ongoing revenue stream to avoid bit-rot;
money to fix bugs as they are found, new features, and so on.  You can
do that in a ransom model, by adding additional ransom for subsequent
releases, but the difference between the ransom and proprietary software
models disappears with periodic releases: it's simply a maintenance fee,
the same as you pay for almost all commercial software.

Regards,
--keith

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Keith Bostic
Sleepycat Software Inc.		bostic@sleepycat.com
118 Tower Rd.			+1-781-259-3139
Lincoln, MA 01773		http://www.sleepycat.com