Wikipedia:Forking isn't as harmful as we think

From Wikipedia, the free encyclopedia
Jump to: navigation, search

The Wikimedia platform was created in 2001, during an "archaic" era when CVS was still the dominant software for version control. In the intervening time, CVS has been highly criticized within the Open Source community, most notably for its relative inability to perform "merges".

Modern version control is generally handled by a radically different approach, embodied by Git (software). Git functions on the assumption that forks are inherently helpful, provided sufficient technology exists for sharing information across forks ('merges').

Wikipedia software and Wikipedia thinking is still run by the old CVS paradigm. It is important for the Wikipedia community revisit WP:CONTENTFORK in light of this new paradigm in Information Science.

An author forks even one's own code to implement new, original or strange ideas that have low chances to succeed yet are worth trying - but without risk to have a lot of cleanup if the experiment fails. Similarly, Wikipedia forks may be risky experiments done by initiative groups that just feel responsibility for this large and successful project. Without the right to fork the GNU project we would not have GNU/Linux as we have it now, as the original kernel, the core of the system (GNU Hurd) has been hopelessly behind. Also failed projects provide valuable information that the idea does not work as expected.

Would it be better to have real user names? Allow ads? Allow Java applets, flash animations? Local server hosting? More freedom for bots? At any given time a number of forks are trying these and similar ideas. Of course, time to time somebody just clones the content and expects to get traffic for Adsense. However such attempts have zero chance of success and are not topic of this essay. We think, to be a real fork, a Wikipedia clone must have its own active community understanding that do they want to try differently, and why.

There is an assumption that user-created content forks are inherently harmful to collaborative creation.
Post-Wikipedia developments in Version Control and Social Media strongly suggest the opposite conclusion-- user-forks may greatly benefit overall quality.
Our guidelines represent our "best guesses", they're not mean to be dogmas. Guidelines should always be re-evaluated in light of new evidence.

See also[edit]