wrought/open-source-anti-patterns-by-byran-cantrill.txt

## open-source-anti-patterns-by-byran-cantrill.txt
Found this very interesting talk by Bryan Cantrill https://en.wikipedia.org/wiki/Bryan_Cantrill on Corporate Open Source Anti-Patterns: Doing It Wrong (youtube https://www.youtube.com/watch?v=NhgXQFk9noI & slides http://www.slideshare.net/bcantrill/corporate-open-source-antipatterns). His slides are verbose, a little rough, and hard to understand, so I took some notes and made a summary, trying to simplify:

Side-note - He pushes a few debatable viewpoints (common flamebait such as GPL vs other licenses), so take these phrasings with a grain of salt (wink)
Summary

Inverted Thinking:

Misconception that the benefits of open source are assumed to be primary ($). In actuality, benefits are typically secondary, tertiary, and one should consider the cost of sharing is pretty much $0.

Wishful Thinking:

Assumption that all potential users are potential customers. Folks who wouldn't have bought your software won't pay for it if it's open source, but they might use it. Assume the benefits of open source are: marginal gain in terms of bug fixes, platform support, community--plus non-linear network affects due to openness and usage.

No source:

Mistake to market a newly open sourced project without the source code (for whatever reason, including FUD). This is a bad idea, you waste the marketing push without any valuable feedback, and it may even harm your reputation.

Forkaphobia:

Fear of folks taking the software, forking, and dividing up the community. But, there is a forking paradox that the easier a software is to fork, the more difficult it is to fork the community. Good experimentation happens with easy forks, safely downstream. When forking is difficult, only dissenters (few) bother to fork.

Sub-Anti-Pattern: Governance orgy (from Forkaphobia)

Difficult forking, especially exacerbated by forkaphobia, breeds focus on governance and formalism,. This wastes times and encourages the formation of factions (thus winners, losers, politics, etc).

Ersatz Democracy:

It's easier to construct governance than to use it. Democracy is great unless the company or community has no intention to observe it.

Eschewing Leadership:

Good open sources projects have good leadership. Make technical leaders visible, internalize the truism "organizations don't innovate--people do."

Eschewing Ownership:

Giving open source software to a foundation can be a signal of abandoning software that is worth little, and giving ownership can mean (even legally) giving leadership as well.

Competitive Paranoia:

The false belief that competitors will rip off your software once it is open sourced. The "not-invented-here" (NIH) mentality can be very strong, and those companies that do use your software are not your competitors. (License Data.)

Anti-collaborative Licensing:

Position that copyleft licenses (GPLv1/v2/v3, AGPL, etc) will protect your code from being stolen by competitors. License incompatibility (GPL vs MIT) can make it difficult for your software to integrate with other technologies and form collaborators. (License Data.)

Dual-licensing for Profit

Freemium licensing model with two licenses (open, proprietary) encourages bad behavior such as for the company to demonize the open or "community" version in order to sell proprietary licenses.

Demanding assignment

Copyright assignment to a corporation, possibly motivated by legal advice, can be very volatile and create power asymmetry in the community. If assignment is made to the corporation, at least view it as a social contract.
	Found this very interesting talk by Bryan Cantrill https://en.wikipedia.org/wiki/Bryan_Cantrill on Corporate Open Source Anti-Patterns: Doing It Wrong (youtube https://www.youtube.com/watch?v=NhgXQFk9noI & slides http://www.slideshare.net/bcantrill/corporate-open-source-antipatterns). His slides are verbose, a little rough, and hard to understand, so I took some notes and made a summary, trying to simplify:

	Side-note - He pushes a few debatable viewpoints (common flamebait such as GPL vs other licenses), so take these phrasings with a grain of salt (wink)
	Summary

	Inverted Thinking:

	Misconception that the benefits of open source are assumed to be primary ($). In actuality, benefits are typically secondary, tertiary, and one should consider the cost of sharing is pretty much $0.

	Wishful Thinking:

	Assumption that all potential users are potential customers. Folks who wouldn't have bought your software won't pay for it if it's open source, but they might use it. Assume the benefits of open source are: marginal gain in terms of bug fixes, platform support, community--plus non-linear network affects due to openness and usage.

	No source:

	Mistake to market a newly open sourced project without the source code (for whatever reason, including FUD). This is a bad idea, you waste the marketing push without any valuable feedback, and it may even harm your reputation.

	Forkaphobia:

	Fear of folks taking the software, forking, and dividing up the community. But, there is a forking paradox that the easier a software is to fork, the more difficult it is to fork the community. Good experimentation happens with easy forks, safely downstream. When forking is difficult, only dissenters (few) bother to fork.

	Sub-Anti-Pattern: Governance orgy (from Forkaphobia)

	Difficult forking, especially exacerbated by forkaphobia, breeds focus on governance and formalism,. This wastes times and encourages the formation of factions (thus winners, losers, politics, etc).

	Ersatz Democracy:

	It's easier to construct governance than to use it. Democracy is great unless the company or community has no intention to observe it.

	Eschewing Leadership:

	Good open sources projects have good leadership. Make technical leaders visible, internalize the truism "organizations don't innovate--people do."

	Eschewing Ownership:

	Giving open source software to a foundation can be a signal of abandoning software that is worth little, and giving ownership can mean (even legally) giving leadership as well.

	Competitive Paranoia:

	The false belief that competitors will rip off your software once it is open sourced. The "not-invented-here" (NIH) mentality can be very strong, and those companies that do use your software are not your competitors. (License Data.)

	Anti-collaborative Licensing:

	Position that copyleft licenses (GPLv1/v2/v3, AGPL, etc) will protect your code from being stolen by competitors. License incompatibility (GPL vs MIT) can make it difficult for your software to integrate with other technologies and form collaborators. (License Data.)

	Dual-licensing for Profit

	Freemium licensing model with two licenses (open, proprietary) encourages bad behavior such as for the company to demonize the open or "community" version in order to sell proprietary licenses.

	Demanding assignment

	Copyright assignment to a corporation, possibly motivated by legal advice, can be very volatile and create power asymmetry in the community. If assignment is made to the corporation, at least view it as a social contract.