FamilySearch Wiki:Accuracy and Collaboration

Accuracy
Newcomers to a wiki often ask "What happens when beginners and experts collaborate? If a credentialed professional writes an article and a beginner edits it, won't the content be ruined? Why would experts want to subject their writings to editing by novices?"

Some edits are simpler than others. Most of the seed content on this wiki came from research outlines published several years ago. These publications contained many archive and library phone numbers, Website URLs, and postal addresses that were out of date. They also contained many computer numbers, a sort of call number that the Family History Library no longer uses. Clearly, the task of updating contact information and deleting obsolete reference numbers doesn't require credentialed professionals, so we set our missionaries to work and they completed the job rapidly.

Other edits require some genealogical experience. When missionary Vida Pittman was adding a FamilySearch publication to the wiki, she corrected errors and also updated information that had been obsolete for years. This is a great example of how volunteers can help FamilySearch staff increase the rate at which content is improved.

Wiki content authoring isn’t about command and control; it’s about the merit of ideas. When multiple authors collaborate on content, they sometimes disagree. This is true whether an authoring team is a group of credentialed professionals or a variegated community of professionals and volunteers. Each page on the wiki has a place where users can discuss conflicts regarding the page's content. When authors are unable to solve conflicts, they will be able to submit their disagreements for mediation. A full history of changes is kept for every page so users can revert a page to an earlier version. When vandals surface, sysadmins can revoke their authoring and editing rights and block their IP addresses if necessary. Content accuracy, then, is moderated by the community. Wiki authoring isn't about who you are, but what you know.

Collaboration and speed
The material in wiki articles often iterate (change). When an author wants to see whether a wiki article needs editing, he should check it. In a wiki there's no concept of an author "checking out" a document and preventing anyone else from writing on it for a week while he edits it. In a wiki, rather than overhauling a document from stem to stern, an author makes a small change to a section, then saves, then makes another small change, then saves again, summarizing the change so others can validate it.

Wikipedia was developed in this way, and it works. You'd think a bunch of volunteers couldn't build a site which rivals the accuracy of Encyclopedia Britannica, but they've done it. Many eyeballs make any bug shallow. We are smarter than me.

The writing process and single sourcing
Wiki pages evolve thousands of times faster than paper publications, static Websites, or multimedia presentations. Some folks within large organizations (like the LDS Family History Department) desire a "single sourcing" model where content can be written in one place and distributed to these other media formats. Some even want the content to be authored first in XML and only then transferred to the wiki and other media. This reflects a misunderstanding of the speed at which a wiki evolves. As a wiki matures and its number of contributors reaches critical mass, its content iterates very rapidly and its quality comes to rival that of formal paper publications. At that point, the wiki becomes the tool of choice for anyone who wants current, accurate content. Since authors can contribute to a wiki anytime, they find themselves using it as their information repository. Since the community tends to iterate content 24/7, an author can post a new article on Pet Topic X and let the community improve it while he sleeps. Months later, when the author is assigned to write a paper publication, static Webpage, or multimedia presentation on the topic, he finds the content already updated, ready and waiting on the wiki. To an author, the wiki experience is like having a whole team of research assistants. So it makes sense to write on the wiki first where the content will attract edits from the community, and then roll the content to other media formats when necessary.

"Protecting" pages against other writers
Writers accustomed to a command-and-control publishing paradigm are sometimes challenged in transitioning to a community paradigm. We can learn a lot from Wikipedia. In Wikipedia, the only types of content that are “protected,” or locked against community editing, are those which are highly controversial or those which tend to attract vandals. For instance, articles on the Gaza Strip might be protected due to persistent battles between Palestinian and Israeli authors. On FamilySearch Wiki, we may choose to protect a page on Mormon research if we find anti-Mormon vandalism becomes a persistent issue. We will do the same for other topics which may draw major controversy. But 99% of the pages on FamilySearch Wiki won't need this type of protection.