The Destruction of the Enterprise Software Industry Is Not Imminent


Gene Quinn at IPWatchdog made this provocative statement on April 2: "Open Source Race to Zero May Destroy Software Industry." I don't think it's likely, but it is a trend worth watching and a blog post worth reading.


For starters, I don't see a "race to zero," whether caused by open source terms and conditions (Ts&Cs) or by anything else. I believe Red Hat has been successful because it offers some great products and markets them effectively. Red Hat's Ts&Cs do not appear as a big market differentiator in my research. To be fair, there are surely some IT folks that will buy only open source, just as some people will only wear Italian shoes or drink French wine. But that is not the bulk of Red Hat's (or Sun's or Alfresco's or Compiere's) customer base. Red Hat has said that it wants to "run" half the servers in the world by 2015 and half the desktops/access-points sometime thereafter. Believe me, it does not want to reach that goal without being compensated by you.


There is some open source software that is truly free from a Ts&Cs perspective. But Red Hat's and Sun's offerings are not in this category, and such software requires top IT talent to deploy and administer it. If one were to follow the IPWatchdog thought to its logical conclusion (all software is only available with open source Ts&Cs), then the software industry would simply be replaced by the "software deployment and administration" services industry. For you, it's "Pay me later instead of now."


And then there's the second part of the "race" idea. If there is a "race to zero," it is the tortoise vs. the snail on a molasses track. The market to make open source software available in a way you can use it without a rocket scientist is already dominated - subsequent to Sun's acquisition of MySQL - by the major IT providers. Red Hat is the only outlier. The half of the server and desktop/access-point census that Red Hat does not want to "run" still leaves billions of servers and desktops/access-points as an opportunity for these large suppliers.


What is likely to happen is buried in that generic concept I used, "access-points." Pick your term: appliance, mobile device, etc. What we think of as desktop software will become totally embedded. But it will be 2020-2025 before that trend dominates. In the meantime, there is plenty of headroom for software of all types, no matter what Ts&Cs you use to license it.