Re: Tim's paradigm shift

Matt Asay wrote:
> (See
> http://management.itmanagersjournal.com/management/04/06/10/1932254.shtml?tid=85.)
>  In it, I argue that open source has less to do with source development,
> and more to do with open distribution.  That is, open source succeeds
> because it drives down the costs of distribution, sales, marketing, and
> QA, allowing its corporate adopters to undercut established players on
> price, in true Clay Christensen fashion.

I agree with your general point here, at least as it applies to 
enterprise software used in corporations and other organizations. (Based 
on your message and your article I presume that's the type of software 
you are talking about.) As you no doubt know from working at Novell, the 
cost of sales and marketing for traditional proprietary enterprise 
software is very high if one goes about it in the conventional way, even 
for a channel-based distribution strategy.

> So, while I gushed on and on about open distribution, Larry posed a
> troubling question:  "If that's such a big deal, how is it any better
> than simply giving the product away for free (no cost) over the
> Internet?"  I didn't know, and still don't.

In addition to Taran Rampersad's and Jamie Lokier's comments, I'd offer 
the following thoughts: The key to corporate adoption of FOSS products 
(and the "open distribution" strategy you're talking about) has been 
bypassing entirely the traditional "push-based" enterprise sales 
channels whereby products get sold by salespeople calling on IT 
management, and relying on a "pull" model where the product gets 
downloaded and adopted by the actual people who will be using the products.

For that class of people the "buying" criteria are much different (in 
some cases almost totally different), and I think it's clear from 
experience that source availability is one of those criteria, especially 
given that for a lot of enterprise software the actual "users" are in 
fact developers. Regardless of whether one would actually use the source 
or not, lack of source is a major risk factor if you're a corporate 
developer using a product procured outside of the normal enterprise 
procurement system, where there's no pre-existing vendor relationship 
you can fall back on if things go wrong. (This is leaving aside the 
question of whether vendor support is actually effective or not in all 
or even most cases.)

I think the other things mentioned by Jamie and Taran are important as 
well, including the potential availability of external help through a 
user community and consultants. However I think lack of source by itself 
would be such a major risk factor for a "free" (gratis only) enterprise 
software product that it would greatly impede adoption.

A good example here is a product I was involved with at CollabNet, 
openadaptor <http://www.openadaptor.org/>. It's a pretty specialized 
product targeted at Java developers building transaction systems within 
financial services firms and similar organizations, and pretty much all 
of its developers work for a single company. It hasn't set the world on 
fire, but it has seen adoption by a number of companies both large and 
small, including several that a proprietary software company would tout 
as major customer wins. I certainly couldn't have seen that happening if 
the product had simply been offered as a gratis binary-only download 
with limited rights; given that openadaptor is designed to be embedded 
deeply within business-critical systems, it's exactly the sort of 
product where the relevant risks almost mandate requiring availability 
of both source code and the necessary rights to take advantage of that 
source code.

Frank

-- 
Frank Hecker
[email protected]