I have a lot more engineering freedom.
After launching InfraDNA to take Hudson to the “next stage,” JAXenter caught up with Kohsuke Kawaguch, to find out what we can expect from Hudson in the future......
JAXenter: When you announced you were leaving Oracle, you said it was to start up a company to take Hudson to the “next stage.” What is the “next stage” for Hudson?
Kohsuke Kawaguchi: It meant several different but related things.
One of them is expanding Hudson's adoption base further. See, Hudson is a very unique project for a company like Sun/Oracle --- it started as my hobby project, and it just grew from there organically. So it wasn't connected to the "bigger picture" of the company at all, and it was just one of those "nice" projects that hang around without any real attention. This has some upsides, but among the downsides is that there's no strategic business decisions made. In contrast, a small but nimble company like InfraDNA has a lot of opportunities to do that ---
I'm already having several conversations with different companies about integrating Hudson with their stuff. The broader adoption base will result in a bigger ecosystem, which in turn improves the speed of progress in Hudson. That's a good thing.
Or put another way, I think it's a good thing for the project to have a commercially sustainable / economically viable model behind it. Sun/Oracle was quite rational in just putting myself into Hudson and nobody else, because it wasn't making any real money. So it was mainly supported by their good will, and that's not healthy. But if InfraDNA makes some revenue, it should be able to have a few people assigned to Hudson, and it's in our interest to continue doing so. That enables a lot of interesting things.
Another part of the "next stage," in my mind, is to be able to work outside the constraints of a big company. For example, at Sun/Oracle, I couldn't use competing technologies without raising some eyebrows, and given the company sizes, they have a lot of competing technologies. Now that those constraints are a thing of the past, I have a lot more engineering freedom. The big item in this direction is a cloud-hosted Hudson offering.
Overall, I hope the Hudson community will enjoy more investments.
JAXenter: When did you first start to perceive Hudson as a viable base, to build a company on? Was your decision to leave linked at all to the acquisition of Sun by Oracle, or is a Hudson-based startup something you've been considering for a while?
Kohsuke Kawaguchi: Building my own company around it has been in the back of my mind for quite some time. It was one of those crazy dreams that I've been cherishing from time to time. But at the same time, I had good colleagues and a happy work environment, where I felt respected, so the idea never really came to the tipping point for a long time.
But as the acquisition drags on by EU, I really started thinking about my options, and when I looked at the idea more seriously, it seemed clear to me that this is doable. It wasn't like there was a particular moment where it started making sense, but when I looked, it was there.
Anyway, that's how InfraDNA came into being.
JAXenter: InfraDNA offers consulting services and development services, where you build plugins and implement new features in the Hudson core. You are also maintaining your role as the Project Lead for Hudson. Do you perceive any cross-over, between the Hudson development you'll be doing at InfraDNA,and the Hudson development you'll be undertaking as part of project Hudson?
Kohsuke Kawaguchi: When people want specific features developed in Hudson, rarely they want to keep those inside their corporate firewalls --- it makes no sense for them to maintain patches and build their own custom Hudson. It makes a lot more sense to push it back to open source Hudson, and let everyone else maintain and improve it from then on. So in that sense, maintaining/nurturing an open source project and offering custom development are very much compatible.
Another good side effect that I'm discovering is that I now get to really see all the mechanisms I put in enabling the plugin development from "the other side," from the view point of a plugin developer. This is giving me a lot of ideas about how to improve the core Hudson.
Yet one more is that now I'm talking to more users than ever, and that also gives me many insights as to how Hudson is used and what features would benefit those users --- I think this is one of the benefits for subscription customers; they hold special places in my mind and their deployments are extra visible to me, so their needs tend to drive the development.
So all in all, I see a lot of positive feedback cycles between two roles.
JAXenter: InfraDNA is still in the early stages, but how are things shaping up, so far?
Kohsuke Kawaguchi: I think so far it's been great! There's a healthy stream of business opportunities, and I'm certainly learning a lot about the landscape. The time management is becoming trickier, and I'm putting a lot more miles on my car now, but I suppose those are good things.
The only thing I'm not really enjoying is the paperwork, and boy there are a lot of those. But I guess that's my new reality.