Subject: Re: All these licenses and business models
From: David <>
Date: 20 Jan 2002 13:31:44 +0000
20 Jan 2002 13:31:44 +0000
On Sun, 2002-01-20 at 04:03, Karsten M. Self wrote:
> on Fri, Jan 18, 2002 at 01:47:21PM +0000, David ( wrote:
> > On Fri, 2002-01-18 at 13:43, Risto S Varanka wrote:
> > > > My issues
> > > > 1: I do not want to be a services only company, in fact I want to sub
> > > > contract most services (maybe even zope type model).
> > > > 2: I do want to be able to sell the system for a profit.
> > > 
> > > Then you probably don't want to give out too much freely. One risk
> > > is that you could give a starting point to a vigorous open source
> > > project that would become your competitor. 
> > 
> > That is the real problem for me.
> This is a frequently cited concern.  IMO it's largely a non-issue.
> First, any potential competitor faces an identical equation.  A
> proprietary company may try to chase you from a market.  However a free
> software competitor faces essentially the same problem:  you can
> undercut their software sales price.
> Your bytes are free regardless, if you're following a free software
> model.  You're looking elsewhere for a differentiator.  *That* is going
> to be a product or service which:
>   - Has a significant marginal cost of production.
>   - Favors a long-term credible relationship.
> To most minds, this translates to services, hardware, or a marriage of
> the two, in addition to software.  For the second time, I strongly
> recommend you look at my comments to FSL regarding the Software 500
> list, and the number of pure-play software companies in the first 20
> companies on this list.
> The answer is 1, its position is 2, its 2000 revenues are exeeded by
> 107% by the top-ranked firm.  Pure-play software is largely *not* a
> viable strategy.  At median, it's an US$30m/yr company strategy (mixed
> model median is $56m/yr).

I have looked at this email and you obviously spend a lot of time with
these issues, forgive me, I am new to the whole licensing thig. 

My idea with Open Source / commercial type license is two fold.

1: Open Source for the following reasons from your email (above) 
One significant problem IMO with the FSB model is perceiving free
software development as the analog of proprietary development.  Largely,
it's not.  

You're not building a large staff of dev + QA + marketing + sales +

Rather, there's a development effort, mixed in- and out- of house.  QA
is largely shifted to the user base, preferably through a "stable" v.
"unstable" dev tree.  Transparency matters tremendously.

I wish to get the solution 'out there' without having to build this
large team in house.

2: Generate enough revenue from this software distribution to enable the
creation of services / hardware components etc. for the solution.

I am trying to get this off the ground again after having spent a
fortune trying commerically to do this (without much luck). Whre I am
theres little or no VC moneys and that scares me anyway (how to persuade
a VC in Scotland Open Source is a good thing). 

Another aspect is (and an important one), what would make a software
inventor / developer good at running services, surely there good at
creating something but as in busines the initial entrepenour is usually
not the one to run a larger company (so exits). What is the software
developer supposed to exit with in this scenario ?

Thats the main crux. If it's a case of having to generate teh revenues
seperatly then is it a case of create a usefull thing for nothing and
give it away (I think what if Dyson had done this wiht the hoovers he
created or teh chap with the wind up radios etc.).

I am sorry if people are getting fed up with me not getting this, I am
very keen on getting the correct business model and benefiting
develoeprs and end users as best as I can. In my opinion businesses buy
producst, they are used to it and can cope with it (with little
thought). It's prbably different for some of us to consider other things
but perhaps not the small business (whic is my customer, through a
reseller, who want to make a margin).

> > > You could consider GPL licensing for some parts of the system,
> > > because then commercial competition can't take the code into their
> > > closed prodocts. (Unless they break the law, which might happen...)
> > > If you want others to develop and maintain some parts of the code,
> > > so you save a bit on development costs, give those parts out as open
> > > source. If you give out that code under LGPL, you can take all the
> > > modifications and patches you get from community, and build your
> > > proprietary components on top of the LGPLed parts.
> > > 
> > What do you think of GPL and having develoeprs sign an assignation of
> > copyright statement that would allow GPL releases and also proprietary
> > releases (similar to what SUN are trying with StarOffice).
> There are varying opinions on this strategy.  The opinions that matter
> are those of the developers (or their employers) you're dealing with.
> There's a strong argument to be made for copyright assignment (it makes
> certain aspects of copyright litigation more straightforward, according
> to numerous legal authorities), but assymetric rights assignment raises
> some hackles.
> > > You could also give the code for free or for charge to other
> > > companies, under an NDA or a limited license. You could include a
> > > clause that allows the other party to treat the code under
> > > GPL/LGPL after a specified period, like 2 years or so. (Think of
> > > this as a kind of guarantee they get when they purchase your
> > > product.) If the license is permissive enough, other companies
> > > could use your product as part of theirs. 
> > 
> > I will look into the delayed public license again for this purpose. Does
> > anyone know where this is.
> It's a concept, not a particular license.  Peter Deutsche has described
> this, IIRC on FSB.
> Peace.
I am also looking at
with some interest.


["application/pgp-signature" not shown]