Howard Stearns’s post, below, about How Croquet will Take Over the Universe (Bwah-hah-hah) got me thinking about OpenLaszlo and our own plans to take over the, um, er, well, our plans for success.
Laszlo Systems, Inc, signs my paycheck, but 90% of what I do for that check is related to OpenLaszlo, the “Rich Internet Application” platform that is given away for free. Just as Howard suggests, Laszlo Systems makes its money by selling applications and services on top of the platform, not from selling the platform itself. Laszlo Mail is the first such product, and others are under development. The OpenLaszlo platform, which Laszlo Systems Inc subsidizes to the tune of several full-time developers and one full-time documentation guy, generates exactly zero dollars for the company.
Laszlo Systems, Inc, is a startup in which I have a relative pantload of stock options. So I want Laszlo Systems, Inc, to succeed, which means that Laszlo has to convince deep-pocketed customers to buy Laszlo applications. In order for Laszlo applications to be acceptable to potential customers, the customers must be convinced that the underlying technology is sound and that it will be around for the long haul. That implies that OpenLaszlo must be seen to be thriving. There must exist a rich ecology of corporations that have a financial interest in keeping OpenLaszlo healthy.
Trust is the substrate upon which the open source ecology can grow. The best way to ensure that trust, of course, it to make OpenLaszlo really, truly open; to make it abundantly clear to potential developers that Laszlo Systems is not self-dealing, not trying to control the platform for its own benefit.
Laszlo is the fourth startup I’ve worked at. I ain’t rich yet, and I ain’t getting any younger. So I want *this* to be the one we get right. Wetmachiners Howard, Gary and I all worked for, and got virtually incinerated by, Curl, which, like Laszlo and Croquet, developed a potentially web-transforming technology. Alas for us, Curl screwed the pooch, as they say; it pissed away all the opportunity that that technology could have given them (us) by messing up this fundamental process that Howard wrote about.
A Curl post-mortem would be a dreary topic and would probably generate, among the few people who cared about the subject, more heat than light. But I think it would be safe to say that Curl failed because they never came up with a strategy for creating a complex ecology with lots of stakeholders who had a strong interest in ensuring Curl’s success. I don’t know if anything could have saved Curl, but during those days, around January 2002, after we had hit the iceberg but had not yet lost the ship, there were plenty of developers who were trying to convince management to give away the source to the platform and overhaul the company’s business plan to selling applications. Who knows what might have happened had Curl gone open source? All I know is that that’s exactly what Laszlo did just about one year ago– it was in late summer 2004, if I recall correctly, that the Board of Directors gave the go-ahead for this course correction — and things are a whole lot more fun at Laszlo now than they were at Curl back when we were taking on water. Over a three month period I called person after person into my office and gave them their pink slips. I personally gave the bad news to 14 people, and then got it in the neck myself.
Curl still exists, by the way, but not so’s you would hardly know it. When Howard, Gary and I worked there, the payroll had about 140 people, 14 of whom reported to me. Last I heard there were about 8 people still on staff, bonded vassals to some Japanese concern that bought out their intellectual property a few years ago.
Way back in the Pleistocene Era, when teh intarnets was just aborning, long before the Web was a gleam in Uncle Tim’s eye, Sun was a little nothing company. (I joined shortly before they went public, in 1986, and stayed there for nine years.) Sun had developed a protocol called NFS, network file system, which they proceeded to give away. It became a standard and put them on the map. Sun went on to promote the idea of “open systems”, often more in slogan than in fact, but nevertheless at least somewhat in fact, and grew like a weed. Sun used this “open systems” mantra/strategy to hit “proprietary” companies like IBM and DEC in their soft underbellies.
Around 1990, when faced with being squeezed out of the play by overdependence on the Motorola 68k family architecture, Sun came up with their own new chip design, called Sparc, which Sun claimed was also an “open” design. Sparc never really caught on outside of Sun, and I don’t know how much of the reason for that was because the chip’s design was not in itself compelling. But it also seemed to me that Sun never convinced many potential Sparc adopters that it could be trusted to not self-deal.
I was at Sun during the early and middle years of the Unix Wars. Which, if you enjoy studying the minutia of the Hundred Years’ War, you may find interesting but otherwise forget it. Anyway, out of the Unix wars arose such things as the Open Software Foundation, and the CDE, common desktop environment project, to be jointly engineered by Sun and HP. These things foundered on the rocks. There was no trust.
Sun’s own soft “proprietary” underbelly was exposed by Linux, ironically enough the scalpel was in IBM’s hands. Now Sun claims that it is once again the openest of the openest. We’ll see how this plays out. But in any event, what they have on offer is no longer a transformative architecture. They’re just selling tools for doing the same old IntarWeb thing.
I fret a lot about the rate at which developers join the OpenLaszlo movement. Now, it’s likely that more developers downloaded OpenLaszlo this month than have downloaded Curl since 1998. That’s small comfort. I won’t be happy until OpenLaszlo has downloads more in the Mozilla neighborhood. There are many thousands of OpenLaszlo developers; ten thousand are registered on our developer forums. I would like to see hundreds of thousands; I would like to see as many people using OpenLaszlo as now use Java. Remember, we’re a tiny group. How do we get there? How do we get developers to play around enough to get hooked?
Clearly we have to make it easy for developers to get started with OpenLaszlo; it has to be easy to download, easy to learn. But that’s not enough: witness the underutilization of OpenLaszlo on Wetmachine, not to mention the OpenLaszlo and Laszlo Systems sites themselves. Cobbler’s children, etc. The point here is that even those of us who know how cool OpenLaszlo is, potentially, have not transformed that potential into actuality.
Beyond being intrinsically cool (Curl is cool, after all, but that didn’t save them); beyond being free, and open, and easy to download and learn, more than that, a new technology needs a few “killer apps”; a few demonstrations of things that you can do with it that you can’t do with any other technology.
When the first personal computers came out, people thought they were really cool, but nobody had any idea what you would use them for. Storing recipies? Planning your garden? Really. People had a general sense that these machines would be nifty and perhaps become indispensible, but we didn’t know how it would happen. Until Visicalc, that is. Visicalc changed everything. I remember its effect quite well; I even remember a job interview I had at Visicorp when Software Arts was a tiny company and nobody who wasn’t an accoutant had any idea what a spreadsheet was.
Recently I was talking about this phenomenon with somebody, who said, “name one OpenLaszlo application that you absolutely must have on your computer, that you absolutely have to have.” His point was that there wasn’t one. There were lots of nifty OpenLaszlo applications, but none of them was “killer.” He didn’t have to explain that developer adoption rates would be slower than we would like until that “killer app” showed up. You can’t just talk developers into adopting a new technology; you have to seduce them.
Well, I’m happy to report that I met that OpenLaszlo killer app the very next day after having that converasation. It’s a music-discovery gizmo called Pandora. It’s in a closed beta now, but it will be opened up to the public this week, and I want to tell you that Pandora will knock you out. The day after I first used it I went out and bought a new set of headphones. I’m listening to Pandora now. I’ve been listening to it 5 hours a day every day since I discovered it. I am addicted to this thing and I must have it on my computer. I’ll be blogging about it as soon as it becomes openly available. I think you’ll like it, and I think you’ll see that it’s the kind of thing that would be a giant pain to implement with any other technology, if you could do it at all.
I’m very optimistic that Pandora will really open people’s eyes to the possibilities of OpenLaszlo. This in turn will speed up the creation of the OpenLaszlo ecology, and that will make it easier for Laszlo Systems, Inc, to sell its wares–provided of course that the applications are excellent, as I fully expect them to be.
The funny/cool thing is, Pandora was built on top of OpenLaszlo without our help. We didn’t even know it existed until they informed us of the Beta. (That implies among other things that our documentation was sufficient to the task. Now that just makes me feel good all over.)
As to whether OpenLaszlo is a transformative or merely a linear improvement on the status quo, (see Howard’s discussion on this point) I’m not the one to ask. But I do think we’re at the forefront of something.
I don’t have any opinion at all about Croquet. To me, for now, it is a theoretical concept. It’s a very intriguing theoretical concept, but it’s too abstract and remote for me to get a handle on.
All I can say is that I hope that when and if it does take over, er, revolutionize computing there will still be room for OpenLaszlo too.
NOTE: I’ve updated this a few time since first posting.