'Community' is an overhyped word in software
A recent developer survey from Packt Publishing asks the question, "Does open source really needs individual contributions from developers to survive?"
The short answer? No. That also happens to be the long answer.
Much is made of the importance of community in open source, specifically, and in software, generally. But "community" is perhaps the most overhyped word in software, one that doesn't deliver nearly as much value as marketing people would like you to think. John Mark Walker called this out back in 2006, but apparently many people missed the memo.
In open source, we tend to think of community as a group of people that actively contribute code to a project. But this myth was debunked years ago. Most people don't contribute any software, any bug fixes, any blog mentions, or any anything to open-source projects, including those from which they derive considerable value. They just don't.
Sure, there are counterexamples to this, but they're the exception, not the rule. And it's by no means clear that open-source projects like Linux, MySQL, DotNetNuke, or (Insert project name of your choice) would shrivel up and die if their "communities" disappeared.
As Linus Torvalds recently told me,
Most users are "free-loaders." There are relatively very few people who actually give back in source code or in bug reports, so anybody who argues against free-loading in open source is a moron.
Not that open source has cornered the market on morons or the hollow hype surrounding "community." It's true, for example, that Apple's iPhone is particularly interesting because of the "community" of applications that run on it. It's equally true that Apple routinely frustrates and angers that community as an inconsistent and sometimes despotic gatekeeper...yet iPhone sales show no signs of abating.
Or take Microsoft. A lot of noise is made about Microsoft's reliance on its channel for up to 97 percent of its revenues. Yet Microsoft is rolling out SaaS applications with margins that are reportedly weak for channel partners, but it's doubtful that this "anti-community" action will hurt Microsoft in any tangible way. The other benefits of working with Microsoft outweigh this apparent demerit.
The reality is that communities of users and, to a lesser extent, developers, tend to form around successful products, the products of which are almost always developed by individuals or individual corporations. It's hard to corral a broad group with common interest, aptitude, and vision for a single project, especially if it's also a requirement that this group agree to work for free.
Hence, even so-called "community-developed open-source projects" like Gnome are primarily fueled by corporate interests. That's just how it is. Always has been, always will be.
So, if you want to rely on a community to build your product for you, good luck. You're going to need it, as experience suggests that hard work by a committed core team develops great software, whether its Linux or Microsoft SharePoint, not some committee masquerading as a community.
Community is what happens once you've done at least some initial work right. Community is mostly made up of onlookers. There's value there, yes, but it's not what most people think it is.
Follow me on Twitter @mjasay.
Matt Asay brings a decade of in-the-trenches open-source business and legal experience to The Open Road, with an emphasis on emerging open-source business strategies and opportunities. Matt is vice president of business development at Alfresco, a company that develops open-source software for content management. He is a member of the CNET Blog Network and is not an employee of CNET. Disclosure. You can follow Matt on Twitter @mjasay. 
http://news.cnet.com/8301-13505_3-10201008-16.html
http://news.cnet.com/8301-13505_3-10171661-16.html
Mind you, I agree with this particular article. Unless you are willing to put a lot of your own time into producing something to give away, the only 'community' that really matters is the user community willing to pay for a product.
Turns out (as expected), that only .03% have actually done anything as far as reporting an issue, posting on the forums, and/or interacting through various channels. To make this clear, 99.97% have done nothing as far as contribution (I was being generous by counting every forum topic as a contribution, even though most are asking questions and not assisting the community).
+1 for Linus's comments. Freeloaders compose most of the community. It's how the game is played.
Roy / Magento
I agree - to a point.
A 'user community willing to pay for a product' doesn't mean 'users willing to give you money for bits on a disk (or a download)' - those days are dead for all but a few in the club that already exists - Oracle, M$, etc. etc. In fact, the 'club' has rigged it so that there can be no future entrants, hence the rise of open source.
You're correct in that the only user community that matters is those willing to pay - pay for services, support, customization, etc. - all of those much lower margin activities that now compromise the future of the software business for most of us.
Cheers,
- Bill
"In the animal kingdom, the rule is, eat or be eaten; in the human kingdom, define or be defined." -- Thomas Szasz
I'm glad you brought this topic up. I cringe every time I hear an open source company spout on about how many hundreds or thousands of developers they have based on forum sign ups or downloads. I wouldn't go as far as to say that this kind of hype is damaging open source but it is easy for opponents to point to such claims and add - you can't believe anything else they say.
I want to point out that even though the term "free-loaders" sounds negative, a project can get still good value from them. The freeloaders, by being freeloaders, tell you that you are doing something right or that you are missing the mark. We measure them in terms of downloads, version checker hits, forum questions and can gauge where they are successful and where they give up in their evaluation. Even the freeloader questions, like "how do I ..." tells us where we need to make the product easier. The real key is not to expect all of the community to contribute directly but to use the information they give you to help build better product faster.
I agree completely with your statement "The reality is that communities of users and, to a lesser extent, developers, tend to form around successful products..." I always here people say it the other way around. The good thing for us is that as we get more successful, we tend to pick up some very good community people along the way. From the hype standpoint, it appears like we have few truly active community members when compared to the huge number of downloads and forum members. The reality is we have easily quadrupled the number of developers and have 40% of our bugs reported by people we do not pay. For a start up company, that is still pretty big value and it is spiraling as we get more successful. When I talk with other community managers, I don't think we are atypical. The value of community is very high and it is real but there has been so much hype, that the real numbers don't look so good. As for undersexed, who isn't?
Doug Moran
Disclosure: I'm the Community guy and one of the founders of Pentaho.
No fair - You changed the title of this article - I hope you are not going all "politically correct" on us.
1. Probably just an oversight. The PR release regarding the survey doesn't actually ask "Does open source really needs individual contributions from developers to survive?" It asks "Does Open Source Need Individual Donations to Survive?"
http://www.prweb.com/releases/packt-publishing/open-source-donations/prweb2345184.htm
The 'donations' in the release and in the actual survey refers primarily to monetary contributions and not code contributions. This is highlighted in both the press release and the survey overview.
The survey does note that contributions can be in the form of time, money or code. In the actual survey, the question originally asked is "Have you ever donated to an Open Source project?" The answers are quite interesting.
http://www.packtpub.com/article/packt-open-source-survey-results
2. Interestingly, of the 4500 people surveyed, only a total of 34% are 'Developer' or 'Programmer'.
3. More interestingly, while the majority (50%) 'prefer' to use Open Source exclusively, a good 32% claim they are more interested in using the best tool for the job, open source or otherwise, and only 13% declare categorically that they will use open source only.
Disclaimer : This is my understanding of the survey results. I am not involved with Packt. The survey results leave a number of questions unanswered. For one thing, it might be useful to know what the answer analysis is for individual job roles, e.g. what are the specific responses from 'Developers' or 'Managers', instead of lumping all the responses into a cumulative result. There is usually only so much that one can read into any survey. An interesting read, nonetheless.
Regarding 'Community' :
1. As has been pointed out, 'Community' can be one of those ephemeral things that can be hard to define. That said, a cursory browsing of many open source projects would suggest that the current definition of 'community' includes users, core developers, third party developers, theme designers and so on.
2. A number of open source projects (a couple of popular web CMSs come to mind) have achieved their popularity because of an available supply of useful third party plugins or customizable themes. These add-ons might not be part of the core product but can surely be considered valuable contributions to the project nonetheless. These non-core developers and designers might not be actively contributing to the core software but are ostensibly part of that software?s community.
3. If wide-spread use of a particular open source software is an indication of that software's success, then what would be more desirable as a 'community' :
- a hundred core developers contributing to the project, but these hundred developers are the only users of the software; i.e. you have a community of a 100% active developers/users
OR
- a core group of ten developers with a million non-contributing users using the software; i.e. a community of 99.999% 'freeloaders'
Which would be commonly regarded as the more successful or valuable open source project? Which has the more active 'community'?
4. It could also perhaps be argued that the number of active developers is merely a function of the total number of users of the software. Generally speaking, it could be that the more users you have, the more developers you tend to have from this community. Even if this correlation is subject to diminishing returns, you might end up with a critical mass of active developers given a large enough general user base.
5. Perhaps more open source projects need to achieve a certain critical mass of users before they can reasonably expect a certain number of active developers to emerge from that community. This could easily be a chicken/egg argument as well (i.e. you need enough active developers to develop a compelling enough software to attract a large enough user base to begin with, etc) and if anyone knows of any research or statistics regarding this I would be interested to find out more.
- Darren
As I mentioned in my comment on Dana?s blog, my company, Black Duck Software, owns and operates Koders.com, a free online database for open source software and other downloadable code. Koders.com currently holds 2 billion lines of code, which is growing as we speak. Developers use Koders to find and use code for their own purposes. In most cases it?s for a profit-oriented business endeavor.
I see Koders.com and plenty of other examples illustrating that there is a healthy pragmatic community of open source developers. Does the open source community exist in the pure, altruistic sense of the word? I agree with Matt that it doesn?t. But isn?t that stating the obvious?
Matt I have always liked your blog because it gave a solid review. Lately however it has started to read more and more like a PR column for the companies you consult to. All round the world, commercial software companies are creaking under the financial strain. This is not so for open source companies. We don't scream when our revenue drops, because it simply does not dominate our thinking.
The open source business works with or without revenue. It is a whole lot nicer with revenue, but it is not us doing the whinging because we have to downgrade from a double cheeseburger. There is a viable open source industry out there and hundreds of great products function perfectly on this basis. Matt no longer understands this, he writes subjectively because he has become trapped in his own model, trying to run open source business on commercial structures. Matt you are experiencing your own personal strain of commercial software and you are taking it out on the open source model. Please go back to giving us objective blogs on open source and lket you paying customers sort out their PR.
Those of us that have been around a while in the software business, both in and out of the open source business, understand the convergence of the open source and commercial business models - its was inevitable. In a world where the suits and the dungarees fight, the rule is - the suits always win.
The open source model IS a business model. You just have to play it carefully and choose your customers and financial partners. Don't expect your community to look after you. They don't - they look after themselves. It is only by picking the right common ground that it all comes together..
The trade off is simple. In the open source model your marketing is effectively paid for you and your logistics gets sorted out by the cloud. Free simple rapid delivery. No need to invest in licencing departments or marketing campaigns. But you can't have it both ways - without the sales revenue you can't be taking clients to football games or hosting cocktail parties at trade fares. You have to make do on much less. Most importantly you can't expect your users to be responsible for expanding or supporting your business.
Matt you are quite obviously not a developer yourself. The Community DOES contribute. In the TurboCASH project, our development is largely dependent on our upstreaming and downstreaming relationships. Many other self sustaining open source components go into making TurboCASH - Firebird, ZeosLib, Reportman to name a few. No one does anything for nothing - duh. But seeding the market with thousands of users (paid for by Sourceforge and our word of mouth customers), sooner or later it pays a single user to finance the entire development of a single improvement. The trick is to keep those development steps small enough so that you do not have to drag your few paying customers into the entire process of the development cycle.
Thousands download our products for free. Hundreds make a great living off integrating TurboCASH. It is an insult to them calling them "freeloaders". They are not freeloaders, they are an army of unpaid zealots promoting our brand. The recession is not the time to be rejecting the free model. This is the moment when the walls come tumbling down. Low revenue hurts the commercial companies a lot more than it hurts us. When the smoke clears on this recession, it will be the open source projects that stand victorious. Matt, the problem that you have, is that you have lined up on the wrong side of the burning barricades.
Ask not for whom The Bell tolls - The Bell tolls for the software tollgates.
But how can one measure the impact of the many online and offline conversations these people have as they go about their business using the product in question? Community is certainly a term which requires more careful consideration. I would say the use of the word 'free-loader' betrays a significant oversight.
- by hurricane_d August 26, 2009 6:16 PM PDT
- I agree that most community members are not used to the idea of posting in forums for help--especially if they are used to calling someone for help with proprietary software. Those same users help spread the word in other ways (I am one of those). Recently, I recommended we go with the open source Alfresco content management system, instead of Sharepoint. This was after years of free loading Linux and getting comfortable with that and other free software. Now, I actually donate $$ to Sourceforge and I post comments to the Alfresco forums. It can take years to get used to the open source way of doing things for the majority of people. No one should be expected to become a developer or contributor right away.
- Like this Reply to this comment
-
(15 Comments)