Marc Fleury, founder of the successful JBoss open-source project and company, is largely considered one of the great open-source pioneers.
Not many people can claim to have have built a project that continues to inspire tens of thousands of downloads each month, plus the commercial envy of Larry Ellison and $350 million from Red Hat.
Fleury can, but he's not resting on his laurels. Having upended the application server market, Fleury is now funding OpenRemote, an open-source home automation project that was inspired while Fleury was shopping for a "geek chic" home automation system and discovered that it cost far more than he thought it should.
I reached Fleury at his Madrid home on Monday to discuss a wide range of issues. But the longer we talked, the more we focused on the value of open-source communities.
After all, how many developers do you know who have the aptitude to contribute to low-level home automation controls? This is a very different market from, say, operating systems for servers, personal computers, or even mobile phones. While many people use remotes to control everything from their TV to their lighting system, virtually no one knows how to program them.
Hence, I asked the question, "How does community even apply to this market?" His reply:
Marc Fleury
Automation is a primitive software industry, though it's advanced in hardware. There are very few standards. It's 100 percent proprietary. Everything here is deployed in a very proprietary fashion: business models, hardware, software. Everything.
Electricians are our installer community. They're great at what they do, but they're not software developers. It's a fairly rare skill to be able to develop this kind of software. But then, open source has never really been about thousands of developers, contributing to a project. It's about millions (or thousands) of users.
Yes, you need a few good guys, which we've funded through an off-shore Chinese development group. (JBoss, incidentally, was much the same: we funded the development exclusively early on.) The open-source dynamic kicks in when you have a big body of users.
I felt like I was hearing heresy, what with the dearth of paeans to community, but then Fleury has never seemed overly concerned with catering to consensus. This is the guy who dressed in Flavor Flav (Public Enemy) style at Javapolis 2006.
But Fleury wasn't disparaging the value of open-source communities. He was simply being realistic about the value and nature of such communities over the life of an open-source project. Until you have established an exceptional project, you can't hope to attract users. And if you lack users, forget about trying to find a significant body of committed developers.
So what does OpenRemote need more than anything else now? Users. And why? Because users attract installers, and the product with the most installers wins. Period.
Fleury noted:
Home automation is a highly fragmented market, which makes it hard for any company to become big. There are no standards here, so you have hundreds of little vendors. So, if you really want to have a standard, then you need to integrate all of those vendors.
But open source helps to alleviate this, attracting a user community that can hold off competitors while attracting a partner ecosystem. Open source serves as a rallying cry; a rendezvous point. We make the system modular and provide the integration points, and then work to attract an installer community to take advanage of these. It's not a short-term strategy: we're in this for the long haul.
The interest in driving a community of users is that we will breed the next generation of home automation installers. If we can scale the awareness of OpenRemote through a community of users, then we can rise above the noise and get installers to take us seriously. They don't want to spend time on a small player. As we gain mass, we should start to see some of the more traditional benefits of open source, like debugging and development assistance.
Not that OpenRemote hasn't had outside contributors. Fleury noted that as awareness of OpenRemote has grown, the project has attracted a trickle of outside contributions. He expects this to grow over time, but said he won't be concerned by a lack of significant outside contribution until the five-year anniversary of the project. (He also noted that one developer is worth 1,000 users, so he clearly recognizes the value of outside contributions.)
For now, OpenRemote employs virtually all of the developers who work on the project and has recently acquired the exclusive rights to iKNX, an iPhone stack that works with the widely popular KNX automation hardware. If this sounds like JBoss' strategy, it's because it was.
And, like JBoss with Java application servers, the time for OpenRemote may be ripe. Despite the morass of nonstandard technologies and bit players in the automation market, OpenRemote's open-source approach just might have a chance to unify the market. It's now possible to put a $200 computer in the wall, which suggests it just might be feasible for OpenRemote to open source a deeply proprietary and fragmented industry.
Would you bet against Fleury?
Follow me on Twitter @mjasay.
Marc Fleury (of JBoss fame) and Mark Spencer (of Digium fame) have a significant side project: turning us all into couch potatoes, but doing it with open source.
(Credit:
OpenRemote)
Their project, OpenRemote, is the world's first open-source remote for home automation. Given the plethora of systems (entertainment, lighting, etc.) that need to feed into a home automation system, opening up a central repository of controls to enable developers to plug in new ones feels like the right way to go.
Part of the OpenRemote team's vision is tying into the world we already live in, not merely making our future more Jetsons-esque. It's therefore great to hear about OpenRemote's work on an application for the iPhone and iPod Touch, despite the lack of launch-timing specifics:
(We have) put together a first end-to-end prototype of the OpenRemote system. The first milestone is targeting infrared with X10 and KNX following soon...In short, the first prototype we have running comes with an iPhone native application communicating with OpenRemote Box (ALIX hardware) over a Wi-Fi connection. The incoming commands are then translated to other media, in this case infared, and sent out to the corresponding device.
This is a great milestone for us--it's a proof of concept that actually works (imagine that!), and it establishes the infrastucture we will build on for other automation system integration.
It's awesome to hear about a working application prototype from OpenRemote for Apple's App Store, especially since the OpenRemote Web site has been quiet for months as the team has been updating it. Now I just need OpenRemote to write posts for this blog during my vacation, and I'll sign up to be its first customer.
Follow me on Twitter @mjasay.
I wonder why Marc Fleury and Mark Spencer decided to call their open-source home automation project OpenRemote. Reading the most recent update on the project, it's clear that these guys are making progress, but its progress toward a much bigger future than "remote" connotes.
The site talks about controlling audio/visual setups, full-home automation, etc. The sky is the limit.
Or is the name the limit? To me, a remote is something I point at my TV. The Marc/ks want me to point it at my house, and who knows how others will extend it?
Ultimately, it probably doesn't matter. What does "Fedora" have to do with operating systems, or "Openbravo" with ERP? Nothing, yet both projects are doing well. Of course, neither name limits itself to one aspect of the project, either.
Am I wrong, Marc/ks?
There are relatively few markets that would benefit more from open source than home automation, with its myriads of different electrical nodes and associated complexity.
It is this opportunity that led to the creation of Marc Fleury and Mark Spencer's OpenRemote project, and that recently led them to release the Beehive database, a "Web-based open-source application to collect, format, and distribute home automation codes."
Similar to the Volantis Mobile Device Database which serves as a central repository for the growing array of disparate mobile devices (i.e., data on screen size and resolution, keyboard, etc.), Beehive promises to be a central repository to manage the profusion of home-automation codes.
From the OpenRemote release:
Until now, no Web-based open source central database effort of this scope existed to bring cohesiveness to a fragmented home automation, or domotics, market. Beehive is seeded with 100,000 codes that are compatible with 2,500 devices. Anyone can browse through Beehive, download whichever codes they need, and contribute new codes.
"Today, there is simply no central database for these kinds of codes--only scattered collections in different and proprietary formats," said Christian Bauer, Beehive project lead. "Beehive attempts to change this. We believe there is a need for a truly open, unified way to collect and share all code formats and enforce a clean database schema for easy consumption by both professionals and hobbyists alike."
It's an ambitious effort, one worthy of and conducive to open source. The same sorts of people likely to be fiddling with home-automation setups (as opposed to buying expensive home-automation setups) are the same people who are capable and interested in contributing back to an open-source project focused on home automation. Beehive is an important step in this effort.
It's good to see Marc Fleury go public with his new project, an open-source home automation project (not yet a company) called OpenRemote. We had talked about it back at Open Source Goat Rodeo 2008, but he seemed to be taking his time to actually release something.
The problem that I see with this idea? We already have an excellent Linux-based home automation company called Control4.
Control4 was started years ago by a pair of serial entrepreneurs, Will West and Eric Smith. That broadband in your hotel? They are almost certainly the ones who put it there (iBahn).
I used to be involved with Control4 during my time as an Entrepreneur-in-Residence with Thomas Weisel Venture Partners, an investor in Control4. The company was doing fantastically well back then (2003/2004), and I'm betting it has easily cleared $100 million in sales by now, and has signed up every major distributor and OEM (original equipment manufacturer) one can imagine.
This is a very, very well-run company. Is there room for an open-source competitor?
... Read more
I just wrote about Linus Torvalds, and how his personality feeds into the success of Linux. I then came across Mark Hinkle's "Five Reasons Why JBoss Founder Marc Fleury is My Hero," and thought Mark was spot-on in his assessment of Monsieur Fleury.
I didn't know Marc well when he was at JBoss. My familiarity with him came after. He had dinner at my house earlier this year and joined the open-source crew for the inaugural Open Source Goat Rodeo.
I particularly appreciate two things about Marc, which Mark Hinkle calls out. First, his legacy didn't end with JBoss. He hired people that have since gone on to start or heavily contribute to a range of other startups: Ringside Networks, Loopfuse, Alfresco, Pentaho, Appcelerator, etc. Much as I respect Red Hat, you'd be hard-pressed to find much in the way of open-source off-shoots that it has fostered. Ditto for Novell.
... Read moreI read the news that SpringSource has named itself the "first proper" Java application server product in a decade, and I was left scratching my head. Over the years I've heard just about everyone call themselves an "app server" at some point (Funambol went through a spell when it was a "mobile application server" [PDF] and ActiveGrid was a "grid application server", or something, as just two examples), and the only two times it made sense to me (in the open-source context) were with JBoss and Geronimo.
How did SpringSource become an application server? I thought it was a framework.
So, apparently, does Marc Fleury, who had some blunt counsel for SpringSource's founder, Rod Johnson:
To me this is a VC driven move. Spring is a natural consultancy, being a development framework, but they have been struggling with their sales in the runtime. So voila, we now have a box drawn around an OSGi kernel, the Spring framework and Hibernate/Tomcat, and it has a name: it's an application server. It is the same thing you had yesterday for free, except it is now under the GPL and a proprietary subscription license.
... Read more
(Credit:
Matt Asay)
Last night my wife and I had "the bad boy of open source," Marc Fleury, over for dinner. He was in town for the Open Source Goat Rodeo (to be blogged (and YouTube'd) separately later today). I was fortunate to get his time over dinner.
I've known Marc for a few years, but not as well as I would have liked. I didn't really get to know him until after JBoss was acquired and he became somewhat less involved in the industry, taking a well-deserved break.
He may have once been a "bad boy," and he might still be such in his spare time, but over dinner and with my kids he was great. You can see him here overseeing the education of my kids: More math and science, less of the Asay staple of literature. :-)
... Read moreOddly enough, the answer to that question ("What's really going on with JBoss?") is probably best answered by someone outside Red Hat: Marc Fleury. Marc isn't shackled by the need to keep corporate secrets, though perhaps he's a little biased.
Marc offers several data points that suggest that JBoss adoption and monetization is "going through the roof." But he also takes on two potential aspirants to the JBoss throne: Ruby on Rails and SpringSource (I've sanitized Marc's comments to suit my Puritan sensibilities):
... Read moreReading Marc Fleury's post on the subject of open source and proprietary software (a response to my post on Benchmark's investment in Engine Yard), you'd be tempted to believe that the world is growing more proprietary. Reading InfoWorld's response to Marc, you'd be certain that yes, the world is definitely closing off.
Unfortunately, the data suggests the inverse.
It seems quite clear to me that the software industry is rapidly, in some cases, and gradually, in others, opening up. Very few can get away with foisting a heavily proprietary model on the market anymore. Were a startup to launch today with a great new idea for a proprietary database/application server/etc., they'd probably fail to get funding and would certainly fail to find a great deal of customer traction. (I should know - the founders of my company, Alfresco, attempted to do just that before starting Alfresco.)
Even Marc's words fail to convince because his example (100% open) speaks much louder than his belated blog entry on the subject. The blog isn't paying his bills. Open source is.
I think the reason for the disconnect between reality and words is because we're looking at the same data in different ways. Let me explain.
... Read more
