Now that large companies are getting deeply involved with open source, the sky is the limit in terms of innovation, says Matt Asay, a Novell Inc. technology evangelist and one of the people behind that company's Linux and open source strategy.
SearchEnterpriseLinux.com recently caught up with Asay to talk about the impact open source is having on the IT industry, and the increasing pace of open source innovation. Here are some excerpts from that conversation:
How would you size up the state of the open source revolution at this particular point in time?
Matt Asay: It's kind of similar to the chip industry. In the chip industry, basically there is 80% or 90% of the [intellectual property] that's just out there, that is common. And all of the innovation is in that last 10% to 20%, and that's really where the chip industry is focused right now -- innovating at the edge, as it were. That is increasingly going to happen in software. It hasn't happened traditionally because everybody has had to rewrite from scratch the ERP [enterprise resource planning] system or the collaboration system, etc. Everybody has just kind of gone off on their own little islands and written these things. Customers then were left to choose between these little bundles of isolation and then pay the system integrators or somebody else a lot of money to integrate it all and make sense of it. It's fundamentally a broken business model, but that's all we had.
Increasingly, that's not the case. We now have this pool of innovation, of open source code, that is just sitting out there, free, for anybody to use. It's free, often in the sense of cost and also in the freedom to go out and use it. You see that in Linux where you can have fierce competitors like Novell and Red Hat sitting side by side, breaking bread as we work on this kernel together. Increasingly, I see open source as again this common pool that everyone is contributing to. Instead of being focused on rebuilding that pool over and over again, we're just focused on extending in that 10% or 20% at the edges. That benefits customers because they get the 90% for free, and it benefits vendors, frankly, because it pushes us to be more innovative.
Microsoft has made some moves recently on the open source front. Is the so-called open source revolution forcing that company and other proprietary vendors to rethink business models?
Asay: Microsoft is the last holdout on that, but even they're starting to brighten up on it by releasing open source projects and getting involved in the open source community. They're lightly toe-dipping, but I think it will increase over time. I don't think that open source poses any threat to Microsoft. Linux poses a threat to Microsoft's Windows franchise, but open source is not a thing. It's not a product that sits on a shelf. It is a development methodology, and so they're as entitled to benefit from that as everyone. Microsoft has 10-plus years of thinking that anything not invented there is probably not worth inventing. But my conversations with [a Microsoft exec have] me thinking that they're actively exploring the topic. [I'm just] an outsider looking in, but I think they're coming around to it.
What progress is Linux and open source making in the area of server virtualization?
Asay: I actually think that virtualization and Linux is, or soon will be, superior to anything you see anywhere else through Xen. It's [because of] shared innovation. I think IBM has about 100 engineers working on it. You've got all these different companies with different agendas, with different strengths and weaknesses, but overtime, Wikipedia-style, it's becoming just fantastic. And I don't see how one vendor could compete with that.
I've heard analysts say that the combination of Xen, JBoss and Linux will soon give virtualization software vendors like Microsoft and VMware a run for their money. Do you agree?
Asay: Yes. But I don't even think it's even a 'run for their money.' To me, it's a no-brainer, [because that open source combo] is more stable and the feature set that it provides is equal to or superior to anything that you can find on the market.
Some say that open source development models won't work for highly customized ERP or customer relationship management applications. Do you agree with this?
Asay: If you think of open source as a collection of rogue developers who do drive-by development for you, then that critique is probably accurate. But, in my experience, if that ever was what the open source community consisted of, it's no longer the case. Open source has increasingly, but not overwhelmingly, become corporate-ized. So, you have the IBMs and the Novells, etc., employing developers who do nothing all day but write free software. More to the point of your question, you have companies set up around open source projects [like SugarCRM and Compiere]. These companies are shepherding their own projects. They're doing the bulk of the development work, but what they're getting back in return, is this global pool of quality assurance testers, people running betas for them and providing but fixes, etc. If you just leave it to the community to go out and create an ERP system, it's probably not going to happen. But the way open source is increasingly working is there are companies setting up projects, establishing communities around those projects and then selling through a variety of business models.
Is the pace of open source innovation picking up?
Asay: Generally, it's been thought in the past that open source doesn't innovate. I no longer believe that [because private companies are now putting major resource behind that innovation.] To the extent that open source is a collection of hippie or geek developers, then no, it probably doesn't innovate to the same extent as [a company like] IBM. I think that as Linux is showing, the pace of innovation in open source far exceeds that of any commercial product.