One of the big, open source successes in both Southern California and the software world was Gluecode Software, which was based in Los Angeles and acquired by IBM in 2005. One of the firm's co-founders, Adam Lieber, is now the founder and CEO of Webtide (www.webtide.com), which is also based here. Adam also had a stint at Mission Ventures, and ran worldwide sales for open source middleware at IBM. We caught up with Webtide to learn more about the firm's open source web server software, Jetty.
Where does Webtide fit into the software world?
Adam Lieber: We make a web server called Jetty, which is our centerpiece, and after twelve or thirteen years of hard work is an overnight success. It's smaller than other alternatives, and is much easier to add to existing software and devices, is lots more scalable than alternatives, and works with both rich web or back end database, web services queries, such are used in transactional banking or gaming type applicatoins. Jetty is used in everything from Eclipse programming environments to a number of commercial application servers out there, to open source projects like Hadoop, the highly sponsored project from Yahoo which showed the fasting sorting of terabyte of random information recently. We also work with Google's App Engine and widget toolkit. It's extremely widely used. Webtide is designed to be a company which is the expert on using Jetty in companies. We provide three lines of service, one is called Developer Advice, which is not a common service out there--where customers can have confidential discussions with our architects through a secure forum, so that they're not posting specifics about their customers or projects publicly. It gives them an escalation path only dealing with us. We give them options and advice on the best course to reach their goals, such as responding quickly and getting to market faster. They can choose to implement our advice themselves, or we can do that as well for them as a microproject. We also provide production support, and we're also very proactive in monitoring items that might affect them, supporting the specific versions of software they are running. It brings support to open source, instead of forcing them to just update to the latest version and hoping it fixes their problems. Instead, we provide support for what they are using. Clients, especially those using embedded devices, might be using Jetty versions which are years and years old, and we still support them. On top of that, we build components that can be included in applications, such as Cometd a messaging service, which we've done reference implementations for, we do AJAX push, to do live events across the Internet, and provides reliable, face communications to large numbers of people across the world.
We've got people around the world, in three states in the U.S., Australia, and Italy. Our clients are similarly distributed across industries ranging from banking to online services, to devices. We've even seen people running Jetty duringly on a Windows Mobile phone. We're also growing and hiring.
Where's the open source market nowadays--a few years ago, people were playing around with business models, how is yours working?
Adam Lieber: I think the business model for any particular open source company depends on the open source. There's all sorts of variations on what type of services you might offer, how you might charge for them, and in terms of how you sell them, too. It really depends on who is going to use the software, and the industry setting--there's not one right answer. Jetty is a horizontal use components, which is particularly appealing to development teams. In other places, on the administrative side of things or in a more transactional mode, other models might be more appropriate. But, because we're a core component of the apps out there, the advice model really works. Because we've really hit a strong note, with Jetty used in thousands and thousands of domains out there, our technology is our calling card, and our service is sold to people who value them. I don't have a long sales cycle, because I'm talking to people who have already chosen Jetty. It's just sign up here, so we can get them into the system. We're also not directly competitive with other commercial offerings, and in fact we have commercial software vendors who test their software with Jetty whenever either side does revisions. People haven't come to us just because we're cheaper, they use us because we're the best technology and service combination for them to bring their solution to market. It's not a matter of looking for expenses that can be cut, people use us because they can't put a multi-gig, application server on a cell phone.
Your software is tied somewhat to enterprise development - how has a relative slowdown in enterprise software affected you?
Adam Lieber: There are people with a range of technical requirements, a range of business requirements, and different license requirements. All of those three vary by customer, but we're seeing that we're headed towards a mix of service offerings and a mix of licenses between commercial and open source. Both will stick around, but the universe is definitely working with open source in their development cycle. We can't endlessly purchase individual components to go into applications, or inexorably, new applications would be much more expensive. You've got to be able to stand on the shoulder of the giants that came before us. Our goal is to be a common underpinning of the Internet, not just the open Internet, bu the Internet. Our open source powers Hadoop, Jetty powers Google--if you look at our community site, the list of products that use use is exhaustive. The list is over 35 pages long of places it's been embedded, and that list is not all of them. Certainly, the use of common components will speed software development, as people realize it's not necessary to reinvent the wheel.
I see you were at Gluecode, and also were in the venture world at Mission Ventures--can you talk about how you've bootstrapped this firm and how it compares?
Adam Lieber: If you follow my career path, you'll see I saw all sides of the table. I was a VC at Mission Ventures, Rustic Canyon and Palomar Ventures VC-funded Gluecode, I was at Gluecode and among the first four people there, and we were acquired by IBM with 350,000 people. I also worked in sales groups for Websphere, and bootstrapped Webtide--all right in a row. They all were very different companies. Webtide is a completely, engineering heavy organization, and wee need to keep it that way because all of our products are customer and support driven. We've profitable and cash flow positive all the way through, whereas a VC-funded company tends to run losses while they build out their market. The growth path is different, and we had to budget to make noise for ourselves in the market. That's starting to change as we've grown and our footprint is more recognized. When I joined Webtide from IBM in 2007, I also discovered that the basic costs of running a company is a lot lower than running a company was in 2003. Even in the past four years, you've seen a lot more software-as-a-service offerings. There's on-demand platforms for accounting, customer relationships, and things like that, and the options in each of those categories have let us have low monthly costs. Far and away, our biggest expenses at Webtide are salary and compensation related costs. Aside from the people, our company costs nothing to run, it's really breathtaking efficiency when you are as distributed as we are. Our also requirements is that you need certain type of people in that mode, since they all work from their home countries without physical offices. They have to be self motivated, and highly communicative, because we're not running into each other at the water cooler. So far so good. Without venture capital funding, it means we've grown a little more deliberately, hired sparingly, but also left us in a place where when lots of the market has cooled, we've never had to cut personnel. In fact, we're hiring, because we can't see any scenario where we'd have to cut someone in 12 months, and already know where their revenue is going to come from. It's a very different mode, and definitely interesting to be back outside of a megacorp.
You mentioned earlier that your workforce is quite distributed, any hints on how do you handle that?
Adam Lieber: Skype and IRC is our main communications channels in our company. I see the names of our folks pop up during the day as we're chatting, and those messages give me the impression of the activity and energy going on in the company. Interestingly enough, that means at our main trade shows - EclipseCon, JavaOne--as much as we are there to meet current customers and prospects, those end up being our largest physical gatherings of Webtide personnel during the year. We gather across countries and continents, and catch up face to face. I don't think we can permanently stay that distributed, but right now, a couple of big meetings a year is enough to keep us going, with the right personalities involved, and universal enthusiasm for our technology. I'm always amazed in search engine alerts in my email, about who is talking about or using Jetty. At EclipseCon, the person who won the prize for the most fascinating use of Jetty came from JPL here in Pasadena, which is affiliated with NASA. It seems that the systems that prepare and send commands to the Mars Rovers are Jetty based, and that's the kind of thing we're very glad to hear about, and think is very cool.