$B of licenses… easy (and the market already accepts this model), $B for support …harder but possible (there is acceptance within mid-large enterprises), $B for services … much much harder (services = people’s time, $1B is a lot of people and a lot of time)

For someone to buy a license or support you have to offer enough value beyond just good, open code.

To get to a billion dollars, you need to offer enough value to enough customers that you get volume. At a billion dollars with value and volume you’re basically becoming an industry standard. So yes, I think Linux distros are prime candidates – they address the entire server market. Open source CRM/ERP apps… smaller market, entrenched “standards” already there, and a difficult market to get volume. Maybe sometime out in the future, but not soon.

So they all seem to play together – tough to say just volume alone will ever get you revenue (ahem… Apache/Eclipse/Mozilla/OpenSSL – how much did Apache/Eclipse/Mozilla/OpenSSL make last year? they had plenty of volume). And it’s harder to say just offering open sourced value will get you there (CentricCRM is a great CRM app… have you bought it?)

Besides… no open source project will ever reach a billion dollars – Larry Ellison always buys them or their underlying technology before they can hit $500M

i would also push back on the eye balls comparison, simply because where eye balls – IMO, anyway – had no real economic basis, volume does. real world businesses such as JBoss and MySQL are proving that the economics are real, if different than what we are all accustomed to.

you bring up a very valid point, however, when you remind us that volume is not the *only* path to ubiquity. it may be, as i assert, be the surest, but there will unquestionably be opportunities in niche areas of the market for products to survive and thrive. i tend to think that few if any of the specialized players have a realistic shot at the billion mark, but then i don’t regard that as the criteria for success anyhow

I just think stating the success of open source is all about volume is too simple and sounds too much like eye-balls. For an organization to be successful they have to know what is their core competency; open source makes this even more important, due to the low barriers to entry. Maybe it is not value but if you define volume as downloads then I just don’t see it being sustainable.

I also believe the world is moving towards more specialized/vertically oriented software. In this world large volumes may not exists but you can still be successful. I actually believe open source is making this possible. At Eclipse we are seeing more and more specialized IDEs that may not have been possible before Eclipse and open source.

Certainly there are some very good examples of volume creating the network effect, Eclipse and the other you mentioned being valid points. However, moving forward I don’t think you need to have the massive deployment like Eclipse to be considered a success.

Stephen O’Grady apparently spent last week at the wrong conference. He came away from OSCON wondering if open source companies can ever reach the $1 billion mark in sales. He writes:As stated at OSCON, I believe both that open source…

agreed. even Microsoft has been shipping BSD software for a long time. but while the closed v open distinction maintained above is obviously artificial, i’d maintain that there’s blended and there’s *really* blended. i would not, for example, lump Microsoft into the blended category because they are not poised to embrace open source to the degree that, say, IBM is.

ultimately, the point of the above is certainly not to argue that blended is not the way forward. it is rather to point out that those that are not blended will have a difficult time growing to 1B in revenues.

“the irony is that, as per brandon wichard’s musing, there are companies “built on OSS” such as Google, making tons of money. but they are not major code contributors to oss. o’reilly’s architectures of participation.”

well, i don’t know that i’d agree with that. while i agree with those who express skepticism at dibona’s comments from OSCON saying that “our code is so specialized it wouldn’t have an audience” i’m not sure saying that GOOG doesn’t give back to OSS is accurate. by sponsoring SOC, they actually give back to a variety of important projects in important and meaningful ways. they’re also reported to have folks working in house on things like OO.o. so i don’t think that assertion is necessarily correct.

Ian: interesting pushback. value rather than volume? i’m skeptical.

let’s look at eclipse. eclipse provides a value, of course, but is and has the value been that far ahead of commercial or open source competitors? i don’t think so. i think Eclipse was way out front in volume, and amassed the largest community as a result which locks people in via the network effects.

is MySQL in the position it is because it offers a great deal more value than, say, Postgres? JBoss more attractive to partners than Geronimo because it offers more value? PHP so much better than Python, Ruby et al?

i don’t think so. i think the argument can be made that value begets volume – though the argument would have a lot of holes – but over the long term it seems as if success in the open source world is defined more by volume than value.

What I don’t buy is that open source is about volume? This sounds too much like the dotcom days of chasing eye-balls. Sure it makes sense if you are in the volume business but maybe open source is really about enabling more specialized software, ie. the Long Tail.

Maybe we should think of the ‘v’ as being value creation. Open soource provides a more effective model for creating value since it allows you to 1) specialize the value creation, 2) lower barriers to entry so that it is easier to reach a niche and 3) reap opportunity cost by not recreating commodity infrastrucutre.

Just a thought…

its a solid analysis but i feel its important to note that ALL software companies are mixed source software companies now. show me a software company and i will show you a firm that is using, and likely contributing to OSS.

the irony is that, as per brandon wichard’s musing, there are companies “built on OSS” such as Google, making tons of money. but they are not major code contributors to oss. o’reilly’s architectures of participation.

we’re seeing a related bet between nick carr and chris wagner (or is it lee gomes?) about profit driven versus peer driven content sites.

surely success will be defined by a mix of both? we shall see.

the reality is that the software business is now a mixed model business. BEA would call it blended.

Brace yourself for impact- you got that right.