The Simple Economics of Open Source

[Gordon McMillan]

Robin Becker wrote:

> In article <1255786824-35980039 at hypernet.com>, Gordon McMillan
> <gmcm at hypernet.com> writes
> >Why does everyone miss the main point?
> >
> >I (and you) are literally 100s of times more productive because 
> >of open source. Imagine having to either invent it all yourself, 
> >or duct tape together expensive half-assed closed "solutions". 
> >Shudder.
> 
> Developers who share code may actually do so for other than friendly
> reasons. If M$ released win2000 source into the world it might actually
> slow down Linux development.

Doubt it. It would probably yield an emulation layer (c runtime 
lib) that worked.
 
> I prefer not to worry about the exact electronics/chemistry behind my
> computer. So in that sense my 'open sourceness' has limits. I guess this
> argument has to do with the relative value 'developers' put on the
> tools. On win32 VC++ is valuable; while gcc/egcs is really useful on
> unices.

What problems would you be solving without Python (TCL ...)? 
Remember the late 80s / early 90s when big companies were 
pushing 7 figure middleware that is now obsolete (thanks to 
internet technologies)? Remember when IBM truly believed the 
world would communicate through SNA? Remember when 
damn near any real app cost at least $10K?
  
> >
> >Developers who share code are *enormously* more productive 
> >than those who don't. So the range of problems that can be 
> >attacked increases geometrically, so computers penetrate 
> >new areas, so geeks stand a better chance of making a living.
> 
> This is a competitive position to take. Geeks presumably compete with
> non-geeks for resources.

So? The article was about why developers like open source. 
Those wierd people with real lives didn't enter into it.
 
> In addition the inter geek co-operation may actually be a form of
> competition for other things like 'mindset' 'noosphere' etc. Painters
> exchange ideas, concepts and criticism freely, but compete for
> customers.

Sure. And my competitive position is greatly enhanced vs 
those developers who rely on commercial products. I never 
used the word "altruism". I was just pointing out that the 
article and the initial responses there to completely missed 
what I see as the main benefit. In most cases, cooperation 
remains a good thing, whatever the motivation.
 
> >
> >Win-win situations are rarely profitably analyzed by 
> >competitive models.

> Wrong! Non-zero sum games are widely studied in many areas of
> mathematics, economics and control theory. 

Did you see any in that article?

> Just because we can both win
> doesn't mean we're not in competition. 

And just because we're in competition doesn't mean we 
should shoot each other <wink>.

> Win-Win games tend not to be so
> stable as minimal-loss minimal-loss (Nash) games and for that reason are
> of less interest. It requires a constant flow of information (co-
> operation) to achieve best-best so that is rarely achieved or sustained.
> 
> Win-Win for the group of geeks is a sub-game in the wider community. As
> a group the geeks strive to turn their knowledge into something useful
> to get income. It may be that inter geek co-operation aids the group as
> a whole; certainly that cannot be simply asserted as there exists a
> large number of geeks who work only for $/£ and who co-operate with the
> prince of darkness in Seattle. I don't believe any comparative study of
> commercial vs opensourcerer geeks has yet been carried out to see which
> group does better.

It doesn't take a comparative study for me to know that, 
despite having an NT box and MSVC etc., I do 90% of what I 
do with open source stuff.

Whether mankind is inherently selfish and evil is another 
question.

i-think-they're-just-sleeping-ly y'rs

- Gordon