Skip navigation

I am in tokyo

Posted by marc.fleury May 30, 2005


I am in Tokyo just for one day. This is no "lost in translation" in fact it reminds me more of blade runner (tis a rainy morning here).


We are announcing a partnership with NEC that covers all of Japan. NEC will support JBoss and Tomcat on its line of servers for the Japanese market. We signed this deal last week so I am here for the press conference.


After 14hours of flight, it is 5AM right now, I am massively jetlagged and I am indulging in "navel gazing" blogging. Funny thing is that I am flying right back out this afternoon so today will be the longest day in my life. I start with a meeting in the morning and finish with a meeting in SFO in the afternoon. As Bob Bickel put it "You are seeing tomorrow, today. But by the time you get home it will be yesterday I guess..."



This article by Dan Lyons of Forbes was picked up by Slashdot. The gist of the article is that OSS advocate and Linus friend, Larry McVoy claims innovation CANNOT happen in OSS. He is venting his frustration with OSS users and the cost of supporting the Linux kernel development (est 500k/year). Basically he is dropping support for Linux and criticizing the business model heavily claiming the model is not sustainable, not economically viable, period.


I know Dan, the journalist, he is one of those clever guys that I keep briefing but who never really writes anything about us :) I talked to him last week about the IBM announcement but he kept coming back with a line of questioning around "OSS can't innovate". He was working on that article already.


Of course McVoy gets teared to shreds by the script kiddies on slashdot for claiming that developers should make a living and that creating innovative software requires pay. I must admit that I actually think a lot like McVoy on some of the points he makes. I am not all doom and gloom like he is, in fact I KNOW that the professional open source business model works and will argue that POSS is inevitable as a business model. However after 5 years of professional open source, I know how hard it is to do and how frustrating the "opinions" of many clueless so called "OSS advocates" can be.


Making a living at OSS is tricky, taxing and leaves me wondering at night. We pulled it off and are making money at about 130 employees but it is a trick. We are destroying value for over-priced software vendors who’ve been milking the licensing system for the past 30 years— we create tons of value for the IT consumer—and whether or not we participate in this wave it is happening anyway— this sort of whining by McVoy is like people complaining about the horse and buggy business going down hill because of the invention of the rail-road. AS LONG AS THE MODEL IS SELF SUSTAINED ECONOMICALLY, the system is defendable.


One of the arguments detailed in the thread is about getting hourly pay for consulting services. Namely, write the core software for free and then pimp yourself for hourly wages. Honey, get me my gun! The communists are on our lawn! This line of thinking could turn software developers into the equivalent of low paid burger flipping slaves, which would be bad obviously. If we end up, as a professional category, as highly paid hourly professionals, say like doctors or lawyers in the US then that would be good but still it would be a far cry from the fortunes that were made in the license model.


30 years ago Bill Gates III wrote the vribrant "open letter to hobbyists" , I reread this last week and I can't say how much I RELATE to the points made by BillG, which incidently are exactly the same as the points made by McVoy. Namely that developers that are writing great software should get great pay and that it is a good thing for the industry to have full time dedicated people. Developers working for free is still a romantic deception embraced both by the press in general and the developers THEMSELVES. Why this is so controversial (developers require pay) really puzzles me, maybe it is human nature to cling onto romantic deceptions and ideologies?


I get pissed off everytime I have to articulate this but it is surprising how many people believe open source is about free development, for the umpteth time FREE SOFTWARE != FREE DEVELOPERS.


Some business people don't get that point, which is funny because it is about money, I am growing convinced that the developers that flame on this topic are either burnt out hackers who don't want anyone to make money or kiddies in college where communism still looks good since mom and dad are the ones paying for the rent and the groceries.


You can only go so far as a hobbyist. Some people wrongly assume that all software can be done on a volunteer basis, and they usually point to Linux as an example of software that compounded over time, without financing, into something decent. The thinking goes that if you give enough time to OSS then you will achieve the same result as dedicated professional full time teams. Also Linux has not innovated as much because the target was to copy Unix—this is not the case with OS middleware…


Here at JBoss, we are exploring a particular strand of self-sustained software production in open source. Our revenue scalability is non-linear with people (that is good). The important question all the business models are addressing is "IS THE OSS SOFTWARE BUSINESS MODEL ONLY A LOSS-LEADER MODEL OR CAN IT BE A SELF-SUSTAINED AND PROFITABLE ENTITY".


See, when it comes to philosophy I am much much closer in spirit to MSFT, which says that developers should make money, tons of it if possible. IBM on the other hand says that developers should wag their tails for low wages and IBM laptops. IBM today, just as they did 40 years ago, still sees software as a tactical loss leader-- something to make hardware and services palatable. OSS development is funded by profits coming from elsewhere, hardware, services, govt and proprietary software. We, and others like MySQL, continue to prove that professional open source is a standalone category.


As a counter point to McVoy though I will argue that OSS has a track record of proven innovation. While it is true that OSS thrives in commoditized, standardized environments (Posix, HTTP, SQL, J2EE), professional OSS also sustains innovation. We at JBoss provide anecdotal evidence of this with work like AO/EJB3/Hibernate/JBossPortal/JBossCache. There is tremendous work done in the open source community, pushing the state of the art in terms of technology and this is particularly true of the Java camp. We drive innovation and standardization.


I believe that some of the innovation comes from the OSS licenses themselves by allowing tighter feedback loops with our users that then become contributors to the code base. To me open source is a powerful model because of the community of USERS.


But the point remains that to do this seriously, professionally, in a sustainable fashion you need to make a living. What is clearly compromised imho is the "instant billionaire" club. I remember the first time I saw Torvalds on a panel and someone asked "why isn't there an open source billionaire", and I immediately thought "cause you are distributing FREE SOFTWARE dummy why else?"... and there still isn't an open source billionaire today. There are very few billionaires period. Your average MSFT developer certainly isn’t one, maybe Paul Allen…millionaires are an achievable category in OS and for JBoss.


I for one, don't believe there will ever be an open source billionaires club, there are and will be many multi-millionaires though. If we execute on our plan without screwing up, we will create a large batch of OS millionaires… we care about the developers and people who create real value in companies getting rewarded.


We never saw OSS, passion and money as being mutually exclusive. We aim to have it all.




While we cared about the IBM news last week, basically that they are coming for us, most people didn't. One person said "bah, you already have critical mass of brand". Another dismissed the news as the death of their project. The press cared for a couple of days and that was it. In 3 JUGS and 4 customer meetings one of our lead developers participated in last week the question of IBM came up ZERO times.


I am starting to think we have overreacted to the news.


What compounded this is the first signal we picked from our sales field organization. While the whole thing drew a yawn from our user base, one account told us IBM is fudding the LGPL, that is how they are going to attack us.


I am shaking in my boots :)


IBM IS AFRAID OF US, IBM doesn't like independent developers that kick their ass in their own markets. We had the number one market share at 34% utilization and IBM second at 33%. And as another of our lead developers put it "that's OK, since we are not afraid of them".


So onto the particularly slimy and smelly goo that is already coming out of Armonk. So IBM is fudding the LGPL and praising BSD style licensing. Why? Cause they can steal the latter, not the former. If you want to know more about this, I explain it in more detail in "From GPL to BSD to LGPL and the issue of business friendliness".


The thing that cracks me up is this: how can they criticize the LGPL with one side of their mouth and praise Linux, based on the more stringent GPL, with the other side of the mouth? It is hilarious and mind boggling if you think about it for a second. It shows

  1. either hypocrisy if they understand what they are really saying
  2. or confusion if they don't,
I will be happy if they continue with that line of messaging. The bullshit flags will be flying in the press.


How can they tell the market that GPL software as in the case of Linux is OK when it competes with MSFT but NOT OK when it competes with their own software? One explanation I feel strongly about: IBM loves open source as long as it doesn't compete with it and only when developers behave like a bunch of good dogs and wag their tails when thrown a salary and a couple of IBM laptops at them. For the other part, real open source, professional open source is dangerous and they are afraid.


I mean, do they really believe their customer base is that DUMB? I am going to enjoy watching them struggle to articulate that message and try to make sense of the pile of wombat drops of a company they just acquired. How long before IBM sweeps it under the rug? He he.


Boys you are going to have to do better than this if you REALLY want to attack our number one marketshare. Something tells us you don't, as the Gartner report pointed out, since it would compete with your own offerings.


Peace, Love and FREE SOFTWARE.




So today was one of those days that you always remember. Let's see, I got up at around 8:30 and was at the office at around 9AM. Then my cell phone started going crazy and very quickly I knew something was wrong. I went through the rolodex of paranoid scenarios that I have in my head and remembered the "IBM will buy Gluecode" rumor we had heard through two sources. Sure enough, IBM bought Gluecode.


First things first, regarding IBM's move, this is quite the PR coup.


This clearly says "IBM wants to kill JBoss". After months of claiming that they don't see JBoss on their radar, they go out and do a highly strategic deal that is squarely aimed at slowing our momentum down. See, it is pretty obvious that the situation with open source has gotten a little bit out of hand. When we came out with our number one market share, I wondered how IBM would react in order not to not to lose the distribution wars against us. IBM is deathly afraid of mass distribution plays in their market. They have their own paranoid history with Microsoft and OS/2. We believe this is a repeat of the OS/2 play :) In this case, IBM is starting with Apache Geronimo, which has become a running joke among open source and industry cognoscenti--with few downloads, little development activity and no community to speak of.


This is also, and perhaps more importantly, a huge move against BEA. It says "Open Source is the here and now;" it validates the model and now customers KNOW that OSS is here to stay, whether with us at JBoss or IBM or anybody else for that matter. It will happen. The inevitable rise of open source beyond Linux has today been validated by IBM.


And finally, with this move, IBM is definitely giving the finger to Sun. That one cracks me up. You remember the movie "The Usual Suspects" and that heist they do on the New York police department? Well, I got to hand it to IBM for craftiness because this is the most aggressive standards stance I have seen them take against Sun. If IBM controls volume distribution in middleware, Sun and the Java Community Process will go the way of the dinosaurs. I actually believe IBM has just forced Sun's hand.


So let's talk about that. This move is so funny and its ramifications so far reaching that it boggles the mind. First the obvious stuff. NO ONE wants IBM to have a monopoly on Java standards and distributions. That will consolidate a lot of people around JBoss. I believe that this will likely turn into a net positive for us. WE are the leaders, the number one marketshare. The business model makes sense and we don't have conflicts of interest like IBM does. WE are a clean play. IBM has a vested interest in keeping their open source offering an inferior product. Already we are seeing "bait and switch" slides spewing out of IBM. They are trying to position this as the "low end" and Websphere as the "high end". The market is tired of fake open source.


JBoss today controls the low-end mass distribution market and is a serious player in the very high end. We have one message around one platform and THAT IS the JBoss Enterprise Middleware System (JEMS). Bottom line we are not worried on the product set end. We are offering the OSS SUPER-PLATFORM. We will continue to gain traction and leverage our number one marketshare. While we are focused on improving developer productivity with our work on simplifying Enterprise Java at the EJB3 standards level and in our own JEMS product suite, IBM's announcement adds complexity for their customers and partners, not simplicity.


We have very solid relationships with HP and Novell and our other partners. As most of you know (and this is also for our employees) we are committed to staying independent and profitable, but the net-net is that this move by IBM could very well strengthen our position in the market overall.


Time will tell, but this is actually fun. JBoss was born with a target on its back and thrived in what was a hostile environment. If evolution has anything to teach us, it is that hostile environments accelerate mutations and natural selection favors the fittest organisms. We are the leaner, faster party, with the true professional open source community base. This move on IBM's behalf will send us into over-drive.


Today will go down as the day IBM came out of the woods and declared its intentions against us. I want to welcome them to the party, open source is a difficult business model and terrain. On the competitive front, we are the established player and this is our turf. If they are ready for war on our turf, then fine, bring it on!