Object oriented programming has continued to make strides in recent years, with Java, C# and scripting languages like Python and Ruby being popular. Those who work with me at Elluminates Software know that I am a computer language junkie, and I tend to like to understand from actual experience what language is best for different tasks. I have always felt that learning computer languages is much like learning spoken and written languages we use for human communication in that there are some people for which just knowing one is enough while others find the semantics, syntax and differing conceptualizations fascinating.

At the same time practitioners of software and systems engineering have been experimenting with cloud based computing, which is essentially like outsourcing your IT activities to an organization and treating the cloud's services like a black box. Companies from Google, Amazon, Rackspace and Microsoft and others are all involved in cloud computing. The idea beyond not having to manage the sometimes massive hardware and software infrastructure that goes along with building a cloud service is that you might only pay for what you use in the cloud and that the cloud's capacity would be elastic enough to scale how you needed it on demand. Of course, being Internet based, there are unreliable aspects from networking outages to architectural failures that the users of the cloud services generally have no way to directly manage.

I also have concerns about security from both an intellectual property and data confidentiality and integrity point of view. Those issues need to be addressed with any serious implementation of cloud based computing.

In an effort to learn more from a startup point of view on cloud computing and the Ruby language, I spoke with Ben Brinckerhoff, Cofounder of Devver.net. Here is an excerpt of our interview:

Erik Levy: Can you provide some background on what you are working on?

Ben Brinckerhoff: Around April of last year, my cofounder Dan and I were working on a startup called Seekler.

Ben Brinckerhoff: It was a Rails-based site and we were doing lots of automated testing to assist with our development. So, after we would write some code, we would kick off a test suite on our computers that would exercise and test a large portion of our project

Ben Brinckerhoff: It was good because it would catch bugs. But it was bad because it would take between 3-5 minutes each time we would run the tests. If we only ran the tests once a day, that's no so bad. But we were running them 10, 15, 20, 25 times a day.

Ben Brinckerhoff: It started to interrupt our workflow.

Erik Levy: Sure.

Ben Brinckerhoff: So, we thought - why not let a lot of computers split up the work?

Ben Brinckerhoff: It was beginning to be clear that Seekler was not growing like we wanted, so we were thinking about our next step. Dan had a week long trip coming up and we discussed what I should work on. He said, "Why don't you try building that testing thing we've talked about? Just to see if it would work."

Ben Brinckerhoff: So, over that week, I hacked together the absolute worst prototype ever. When he got back, I had pulled out old computers from our closets, old laptops, etc and had it running on all of those. I think you had to manually copy over some files before you could run the tests, and it only worked for our project.

Ben Brinckerhoff: But it ran the tests in 30-45 seconds. I think it was running on like 6 machines at that point.

Erik Levy: So, that opened your eyes to the possibilities?

Ben Brinckerhoff: Exactly.

Erik Levy: Do both of you code or just you?

Ben Brinckerhoff: We both do.

Erik Levy: And was the previous effort solely founded by the two of you as well?

Ben Brinckerhoff: Yep, it was just the two of us.

Erik Levy: As some background, how did you meet?

Ben Brinckerhoff: Sure. Dan and I met at the end of the 6th grad in Springfield, IL. We had gone to the same schools for awhile, but had never been friends. This is a somewhat confusing connection, but my dad's secretary's son was best friends with Dan's older brother. I think my dad's secretary forced her son to hang out with me a few times as a favor to my dad. Well he was older and much cooler than me, so he was like, "You know who you should hang out with? My friend's brother."

Ben Brinckerhoff: Originally we got together to talk about comic books and draw. Thus, began a long line of nerdy pursuits.

Erik Levy: But you are based in Colorado now?

Ben Brinckerhoff: Yes, Boulder. We were living in Denver, but moved to Boulder when TechStars began.

Erik Levy: Why Ruby?

Ben Brinckerhoff: A few reasons. First, it's the language we know best from our previous work. We knew we could build Devver much, much faster in Ruby than in any other language, in part because it's a great language for building something from scratch that needs to be flexible, and partly because our expertise is there.

Ben Brinckerhoff: But fortunately, it also works well for our business - the Ruby community tends to be very test-focused. When you create a Rails project, testing is baked in by default. We felt that this project would find better traction with Rubyists than others.

Erik Levy: Are you self funded, VC funded, et cetera?

Ben Brinckerhoff: We recently closed our Series A that included a major San Francisco VC firm as well as a group of stellar angels.

Erik Levy: Was that something you had done before?

Ben Brinckerhoff: Well, TechStars provided a little money for the summer. But before that, we were entirely self funded since December of 2006, which had pretty much exhausted our savings. We did sell once piece of IP to help us in that time, but basically, we lived in a dump and ate a lot of pasta.

Erik Levy: It doesn't sound atypical.

Ben Brinckerhoff: Yeah, I think it's pretty typical. It's rare to talk to someone who did a startup that doesn't have some awesome stories about their early living/eating conditions. It's fun to tell those stories, but of course, we were there by choice, so we really had no reason to complain.

Erik Levy: Did you have to write a business plan?

Ben Brinckerhoff: We actually never got around to writing a true business plan. I think some people say that the business plan itself isn't so valuable as the process of making it, because it forces you to think about a lot of important issues.

Erik Levy: I think that would be interesting to the readers since a lot of advisers and VCs say they want a plan.

Ben Brinckerhoff: The process of creating our presentation for Investor Day (and the TS process in general) really forced us to consider those questions.

Ben Brinckerhoff: Yeah, I mean, it's different for different advisers and VCs. In our case, we'd thought a lot about those questions. Our presentation reflected some of that, but mostly we just had it in our heads. We could answers those questions in detail when asked.

Ben Brinckerhoff: We did create an executive summary to send out to people, but it wasn't that great. I think our investors and mentors just felt that it was more important to grill us to get the details they wanted instead of having us write it up in a formal document. YMMV.

Ben Brinckerhoff: But I would say it's good to at least look over the outline of a typical business plan and make sure you can answer all those questions in depth. It's a good tool for pushing you to consider what you don't know (yet).

Erik Levy: A related question I imagine would be if the Ruby community is large enough to make money off of and also I would think security comes to mind since providing data to the cloud means you could loose control of it. Obviously, looking to Microsoft, Google and Amazon and many others, there is a belief that clouds are here to stay and a lot of developers push some IP to the cloud already.

Ben Brinckerhoff: Good questions: On Ruby, we hope so! No really, Ruby is sizable now (roughly 1 million devs worldwide) and growing quickly. Predictions are famously difficult to make, but to the extent that you believe them, it's estimated that there will be 4 millionish by 2013. We think that if we provide a service that clearly saves Ruby shops time and money, we'll be able to make a profitable business.

Erik Levy: Do you find security is an issue or do you assume people who will use the service are comfortable with cloud-based openness?

Ben Brinckerhoff: Security is another great question. Some people won't trust their data in the cloud, period. I frankly don't think defense programs (who I wouldn't expect to be huge Ruby users, but who knows?) would use us.

Ben Brinckerhoff: That said, more and more people are realizing that if the benefits are clear and the track record of the company is good, they are willing to take a calculated risk. Consider the success of Amazon's AWS, Github, 37Signals' tools, etc.

Erik Levy: So, it comes down to trust at a basic level?

Ben Brinckerhoff: I think more and more shops will trust the cloud in general. But we need to make sure they trust us. Part of that will be making sure our customers understand our security model. Our first users are very concerned about security. We basically were very open about it and gave them a complete description of how we would secure their data, every step of the way.

Erik Levy: Where do you want Devver to be in the short term? Are you still focused on product development or more marketing (or both)?

Ben Brinckerhoff: We're really focused on producing an awesome product right now. We're working with our early users now to make sure the experience is great and we'll be expanding the number of users in the months ahead. We're developers and for us, the best "marketing" is other developers recommending a tool to us. We know that making Devver awesome to use is necessary for success and that no amount of marketing will convince developers to use a sub-par product.

Erik Levy: What advice would you give someone or a group of people thinking of starting a startup?

Ben Brinckerhoff: If you are considering doing a startup, my advice is just to do a worst-case analysis. Many people look at startups and think that it's some gigantic risk, but the reality is that most people considering a startup could reasonably easily find a job as good as their old one if the startup fails (sometimes even better, because of the experience the startup gives you). If your current job is begging you to stay, why not take the risk? The worst you can do is to go back to where you are right now.

Erik Levy: Really appreciate your time Ben. Good luck!

Ben Brinckerhoff: Thanks!