Subject: Free software businesses
From: DV Henkel-Wallace <gumby@henkel-wallace.org>
Date: Tue, 13 Jun 2006 21:11:13 -0700

I guess I'm feeling prolix this evening.  Or I'm giving this mailing  
list one final chance, or something (not that I mean to imply that  
mail from me will raise the S/N ratio).

Turnbull alleges that there are some core business issues with FS  
businesses that make them different from others (which makes sense  
and I don't significantly disagree).  I have come to believe however  
that these issues are in some important sense at the margin:  
successful FSBs are first and foremost "B"s and FS is just a  
vehicle.  It would be hard to find many who would disagree,  
especially those who jump in immediately half-cocked, but I doubt  
enough people truly grok it.

I'l repeat what I've said for decades: "there's no such thing as a  
'Free Software business' any more than there is such a thing as a 'C  
business' or a 'SQL business' "

Turnbull then postulates some distinguishing issues to be resolved  
(though to be fair he makes no claim that they might be exhaustive).

His biggest one is the perennial favourite: that somehow FS  
businesses are at a disadvantage compared to proprietary software  
businesses because of the lack of royalty power (he uses the phrase  
"mitigating the lower power to generate future revenue?").  I no  
longer think this matters.  Sure, someone selling licenses can pick  
up some incremental revenue, but in the vast majority of cases (see  
below) FS replaces an existing package rather than innovating to any  
significant degree; in those cases you just have to follow the path  
of other people in commodity businesses, whether true commodities  
like grain or steel (there are companies large and small in both  
those segments who prosper) or those who surfed the commodity trend  
(e.g. tons of small software providers, gardening services, chip mfrs  
or even Microsoft, for quite a while the underdog).

For _none_ of these businesses is the FS the core element, though  
it's sometimes dressed up that way, just like companies like  
Microsoft and Apple pretend to be technically inventive or Disney  
good storytellers.  So what?  Stop trying to somehow find the magic  
recipe to square the circle and just get over it.

Now sometimes FS actually is innovative, usually by accident when it  
slips into a new undiscovered niche (e.g. Apache).  Sometimes the  
innovation _is_ the niche, as with Linux.  These are rare cases,  
proportionately I would bet as with proprietary software or any other  
product.  Of course the advantage is ephemeral, again just as with  
any product.  See any good Apache support companies lately (or any  
significant innovation for that matter?).

So talk about business sure, or perhaps talk dispassionately about  
how FS can be a tool to increase the probability of the second case.   
Which leads to another point of Turnbull's:

> 6.  How does an open source programmer morph herself into an
>     entrepreneur?  Is it generally possible to start with your
>     favorite software and find customers for it, or does that only
>     work if you're lucky?

Well, the second question is just a restatement of the first issue.

How does an FS programmer morph into an entrepreneur, well, as you  
can see I don't think that's an FS question at all.  But perhaps this  
could be a sympathetic place for such a discussion if anyone actually  
spent time talking about _that_.

Oh yeah, more perspective:

>> "g" == DV Henkel-Wallace <gumby@henkel-wallace.org> writes:
>
>     g> Thanks guys for speculating for what I might or might not have
>     g> said.
>
> You're welcome!  I'd much rather hear from you, but you've been rather
> busy of late.

So I am very very busy and should not be writing these long messages;  
I'm (gasp) actually producing _stuff_.  As in in a factory, following  
ISO9000 (OK, it's been renamed), etc.  Every person in the company  
has a laptop, plus we have one machine for accounting, and three to  
run three instruments, and oh yeah a cheap standalone fileserver  
appliance.  I think the appliance may run some sort of Linux, but  
it's essentially a sealed box.  Maybe the gateway does too, who  
knows.  Otherwise, apart from me, every machine runs Windows.  Some  
run NT 3.

Free software/"open source" just isn't on my radar, and if someone  
offered me something and that phrase appeared in the first visit, I'd  
probably just show 'em the door.  I'm small, and cheap, and moving  
really fast, and need major innovation, and I don't have time for  
that crap.

Too bad, eh?  What have you got for me?