Free market, Open Source and risk mitigation

Image via Wikipedia

I had an interesting chat at lunch with Christophe Laganne (eWeek Europe) and Olivier Rafal (IT News) about various updates from nuxeo and we quickly came accross the Autonomy acquiring Interwoven topic in particular, and software market consolidation in general. I had a chance to expose my views about risk levels and risk mitigation when investing in software (“investing” as in “choosing and deploying a software platform to run a critical part of your business”). I thought I could take this opportunity to develop a bit on this topic, here.

Risk mitigation is a major concern for CIOs when choosing the software that will make (part of) their business run. Choosing this kind of software (and with ECM this is usually the case) is a life-long wedding. Risks have to be well considered and mitigated if something goes wrong or it can result in a really painful, if not fatal, experience.

I think Open Source can play a lot in software-related risk mitigation strategies, for a simple reason: choosing a successful (as in “actually deployed and used”) open source software and/or vendor is much safer than almost any software from a billion-dollars vendor. I know I have obvious interest in this case, but I’ll try to make an honest argument. :-)

A software is a market

One can consider each software as a market in itself. Choosing a software is entering this market. Vendors, service providers, customers and users are actors of this market.

With proprietary software, when you select a software you enter into a controlled market, ruled by one main actor. The software vendor’s own monopoly. When a proprietary vendor goes down, is acquired or simply releases a non-compatible new version, customers have a risk to be forced to spend a lot of money adapting / migrating (usually meaning re-doing) their projects. And the size of the vendor is a detail: it doesn’t really have any impact on what happens for customers. How would you feel as a Hummingbird customer? Or as a Stellent one? Or as an Interwoven one? I would be worried. And I would be worried too if I would be using software from OpenText, since market conditions have turned them from a predator to an attractive prey.

I don’t like monopoly. It usually hurts — trust me, I’m French, I know a lot about them, we had plenty around… :-).

With Open Source Software, it’s a totally different story. When you select an open source software and/or vendor, you also enter a market. But a free one (as in free market, not free beer ;-). There is no monopoly anymore. When a software vendor goes down, stop supporting a software, change direction or — better for them — get acquired. It actually impacts the market by creating new needs around the software: customers want support, bug fixes, improvements, maybe even new versions or just more confidence in their providers. And as in any free market, clever entrepreneurs can make some money providing those services and products to demanding customers.

Free market at work

3 years ago, we were a tiny open source vendor and our software, CPS, was a successful open source content management platform, based on the Zope app server, widely deployed in the public sector in France and Europe (when I say widely, I mean thousands of instances in production, many of them being still actively used and maintained).

At some point, we were planning to make a major new version of the software, to transform it into a full-scale ECM platform. The choice of the technology came and we considered migrating our code base from Python to Java. Debates were intense and the decision was difficult. But for many reasons we decided to switch to Java. It was a tough move and a big risk for our company, hence for our customers. But we thought it was the right way for us and our customers in the long run. So, we’ve started to work and announced our intentions to the public. We’ve explained the move to our customers, committed on 3-5 years (minimum) support for CPS — Three years later, we are really happy of our decision back then, we've made a tremendous enterprise-grade ECM platform, and we are still maintaining CPS and supporting our customers that are still using it.

After the announcement, some customers and users were concerned about our move, thinking that we would not be enough committed on the support because of our new technical direction. And I can understand that, actually. So what happened? Well, several companies (Nuxeo’s partners or companies hiring some of our ex-employees), started to offer standalone support for CPS and some even sending patches and participating in the maintenance of the software. Free market at work.

Most of our customers kept their confidence in our company — thanks to them — and continued to work with us, but some choose another way. And it’s great. The move was obviously a risky move for our company and it could have failed. But our customers were — and still are — protected from our failure thanks to the good ol’ free market. Even if in that case, good for us, a radical solution hasn’t been required for our customers since we’re still there, live and kickin’ with a great ECM platform. The market has, at least, served as protection of our customer investment and added confidence in our market. Maybe it also helped us to stay committed on the support of our legacy platform.

And there is a number of examples where Open Source has protected customers from vendors’ troubles. Take a look at db4o, acquired by Versant, at Sleepycat and Inno, acquired by Oracle, at Zimbra, acquired by Yahoo, or at Xen, acquired by Citrix. I’m not worried as a user of these software neither as a customer of some of those vendors. It would have been a very different story if they were proprietary ones.

Great benefits for IT buyers

What happens in the proprietary word? If your small vendor decides to rewrite its software onto a new platform and fail: you’re screwed. And what if your big vendor gets acquired by a bigger one: screwed too. Same story if your vendor just wants you to upgrade because you’re 2 versions late and he doesn’t want to support the version you’re running on anymore.

Open Source offers a unique way to secure your investment in software, not relying on wishful thinking but on actual well-known and proven economic mechanisms. And this might be a major improvement brought by Open Source: transform a monopoly controlled market into a free market. Of course, it’s more difficult for vendors but it brings so much more value and confidence to customers, software users, IT buyers.

I really think this perspective should be considered if you need to choose a software on which your business relies to run its operations.

I would be pleased to engage a discussion if you have comments on this topic. I’d love to develop further.

Let them free! ;-)

EB.