Archive of UserLand's first discussion group, started October 5, 1998.

Re: Patents, Open Source & Server-side Development

Author:Will Iverson
Posted:8/23/2000; 4:55:37 PM
Topic:Next survey: Are you an open source developer?
Msg #:20023 (In response to 20010)
Prev/Next:20022 / 20024

wgi: The reality (IMHO) is that 1. closed binary versions can't ever diverge from the original base, or they always eventually become dead branches, and 2. the distribution benefits by pushing back to the core will always eventually drown out the gain from the closed tree. /wgi

mah: Not true.

PostgreSQL has two proprietary forks in its history, Ingres (now owned by CA) and Postgres (now integrated into Informix). From the history, it does not appear that the original forks contributed to the continued development of the "Open Source" product. /mah

Hmm. Very good point.

I guess part of the issue is the scope of the engineering effort expended on the branch.

The thing I had in mind was the work on Tomcat. From what I understand, many companies take the Tomcat work, extend it, and ship it as a product, with few pushbacks. I think my point was that everyone can still get Tomcat, and for a lot of the stuff out there, Tomcat is as if not more relevant than the closed alternatives.

Also, Apache. You can get all sorts of Apache derivatives, but people still consider Apache to be Apache. If you get a binary only derivative, you're placing yourself at significant risk of eventually getting a dead branch.

I don't think I'll make any comment about the long term viability of Informix. I'll confess I know nothing about the adoption or success of Ingres. For disclosure, I use MySQL for development and other projects and recommend Oracle for those who want, need, and can afford it.

Perhaps it boils down to a risk assessment. Is the open source option riskier in some fashion? Is the commercial version going to be supported over the long term? Does it perform well enough? In other words, why do people go for the commercial version?

I guess that my point is that as a general rule, over a long enough period of time, there's a good likelihood that the open source version of a given piece of software will eventually include the functionality of the commercial versions, eventually displacing them.

I suspect that this plays a role in why such projects as the Linux kernel and Apache are playing such a role now. They are "good enough" for many tasks, and the commercial variants are in varying stages of displacement.

Having been involved with commercial product development for a while now, it's become a bit of a truism that all commercial software eventually dies. I guess the question is, when and how long...

Very good point...

Cheers,

-Will



This page was archived on 6/13/2001; 4:56:11 PM.

© Copyright 1998-2001 UserLand Software, Inc.