Running a business like an open source project

[friendsofkim]

I thought about how some aspects of open source community collaboration could be applied to a business:

• A business is a community of people with no barriers to joining (you don't need to ask permission to participate.)
• A core group of developers define the overall objective of the business
• There is a monthly development "pot" of money
• Anyone can submit feature requests, specs, bugs to fix.
• Members debate and comment on items. They "vote" on ideas they want to see implemented.
• The votes determine the 'bounty' for each item, as a percentage of the pot.
• Items could be approved by the core development team.
• Whoever completes the item to the satisfaction of the development team first gets paid the bounty
• The pot is replenished from sales of the product

The reason I've been thinking about this is that there is apparently a big difference in productivity, motivation, enjoyment and creativity between projects which are open, collaborative efforts (Linux, GIMP, Django, Bitcoin etc.) and corporate ones. I read recently that 2/3rds of corporate software projects fail... which is astonishing. What's more, people working on some projects often complain of poor management, lack of motivation, low energy and interest in the project.

One of the big attractions is the idea that it is entirely self-motivated. No one has to work on projects they're not interested in - so everyone goes where they want to be. The bounty system allows people to be paid for their work however.

[myrkul]

I like this idea. Very much.

[genjix]

I used to work for a group like this that worked on free software. There was 2 arms to the organisation: pay the bills workers (web dev, system setup, admin, ...) and research which I was part of doing 3D dev.

http://en.wikipedia.org/wiki/Agile_development
http://en.wikipedia.org/wiki/Cooperative

One of the problems I saw was if some people become unproductive or turn out not to be very useful then they become a drain on the organisation. How do you then boot them without creating bad blood? Some types of work are hard to objectively evaluate because they bring added value that isn't immediate to see (or tangible to measure).

My future plan is to make an organisation like this anyway since it allows everyone to work on whatever cool stuff they want and is more productive.

[ribuck]

I thought about how some aspects of open source community collaboration could be applied to a business

I really like this idea. But before people get too excited, there is strong evidence that in such situations, monetary reward leads to worse results for any task requiring creativity. Therefore, you can't expect to duplicate the motivation and productivity experienced by voluntary software projects.

See this video for an overview of the research on money and motivation:
http://www.youtube.com/watch?v=u6XAPnuFjJc

Rather than having a centralized bureaucracy in an open collaborative business, I think it could be run by allowing anyone to create and manage one or more of the "cogs" that make the business work. Each cog (person) chooses which other cogs to interact with, and the net result is the overall business.

For example, suppose you had a project to create forum software like the one that runs this forum. Fred might write the code that reads the database and formats the web pages. Susan might write a CSS stylesheet which Fred will use. But if Tom wants to write a better one, Fred is free to use that instead. And Susan is free to enhance her stylesheet until Fred prefers to use hers. Or Susan and Tom might choose to combine their efforts.

Sam would prepare some graphics, which Fred and Susan would be free to choose or not. And for file uploads (attachments), one of the bitcoin-based file upload services would be used.

I don't know how well I'm explaining that, but the idea is that there's no central committee making decisions. The people who are working on each part of the business are free to decide how they interact with the neighboring parts of the business, and the business as a whole emerges from the sum total of all those decisions.

[myrkul]

I don't know how well I'm explaining that, but the idea is that there's no central committee making decisions. The people who are working on each part of the business are free to decide how they interact with the neighboring parts of the business, and the business as a whole emerges from the sum total of all those decisions.

Decentralization has been and continues to be the trend in almost every field. People are realizing that top-down doesn't work as well as grassroots for almost everything. Off-hand, I can't think of anything which needs top-down organization, and a bunch of things which it hurts.

You'll still probably get leaders and followers, but the leaders will be people striving after their personal vision, and the followers people who want to help them succeed.

[ribuck]

I thought about how some aspects of open source community collaboration could be applied to a business

Friendsofkim, you might find this video about the Bettermeans organizational tools very relevant to your idea. The Bettermeans tools are free for projects whose tasks are public.

http://www.youtube.com/watch?v=MAlnMWlvw9g

[friendsofkim]

One of the problems I saw was if some people become unproductive or turn out not to be very useful then they become a drain on the organisation. How do you then boot them without creating bad blood?

@genjix Good point. A lot of large organisations can afford to carry dead wood, but smaller companies will be more severely affected by unproductive employees. What's attractive about the open source approach is that people self-select with respect to what they do, so they don't get bogged down in things they don't want to do. People also decide how much they want to work.

But before people get too excited, there is strong evidence that in such situations, monetary reward leads to worse results for any task requiring creativity.

@ribuck This is true to an extent - when you do something for pay it tends to be less personally rewarding, but it doesn't have to be. I think it's realistic to expect that once you introduce financial rewards within an open source collaboration project it will mitigate other motivations (enjoyment, sense of achievement, recognition etc.) But... the idea behind the open collaboration model is that it balances these different kind of motivations, instead of making work entirely about money.

The problem is that people can't work for free all the time, they need a source of income to live. The challenge is to make money making more free, collaborative and enjoyable.

Decentralization has been and continues to be the trend in almost every field.

@myrkul A trend partly driven by technology - but I wonder if decentralisation can go too far? In some circumstances centralisation seems essential, e.g. Britain in World War II. Decentralisation seems desirable where it is possible - it is increasingly enabled by technology. What seems important as well (particularly where hierarchy is concerned) is the freedom to exit from political systems and organisations (something Locke saw as essential to preserve the legitimacy of the social contract.)

@ribuck Excellent link - very relevant indeed. I was wondering what role software could play in formalising these processes. Any examples of successful projects using this system?

[kiba]

@genjix Good point. A lot of large organisations can afford to carry dead wood, but smaller companies will be more severely affected by unproductive employees. What's attractive about the open source approach is that people self-select with respect to what they do, so they don't get bogged down in things they don't want to do. People also decide how much they want to work.

Until there are things that are necessary to do but no one wants to do. That's why it's called work, not leisure.

[myrkul]

Until there are things that are necessary to do but no one wants to do. That's why it's called work, not leisure.

Undesirable jobs will slowly rise in offered reward, until someone snaps it up. See the video thread for an example.

[genjix]

Until there are things that are necessary to do but no one wants to do. That's why it's called work, not leisure.

Not so true. You do the 'work' on free software because you want the project to be good and it has to be done. Likewise everyone in this project wants the organisation to do well- that's motivation to do the 'work'.

It's just:
- Dead wood people.
- Aligning people's interest to keep the organisation healthy.
- Fostering an environment where people take initiative, not orders.

[friendsofkim]

Until there are things that are necessary to do but no one wants to do. That's why it's called work, not leisure.

The payment model increases financial reward for less desirable tasks.. so potentially it improves on a potential weakness of unpaid open source collaborations.

I think genjix also has a point: if you care about the goals of the project, you will have an interest in taking care of less desirable work.

[Anonymous]

There's no stable vision for the final product. There's room for a lot disagreement.