Some people eat, sleep and chew gum, I do genealogy and write...

Wednesday, September 25, 2013

Why a Wiki Works

With the FamilySearch Research Wiki passing the 75,000 article mark, it is about time to make a statement about the viability of wikis. By the way, this topic also applies directly to the issues raised by several detractors of the FamilySearch.org Family Tree, since the Family Tree is also a wiki-like program. I suppose you could debate whether or not the Family Tree is or is not a wiki, but it certainly conforms to the definition of a wiki, even if it is lacking a few of the more common characteristics of the same ilk.

For the moment, I am abandoning the issue of whether or not sharing your genealogical data on a unified family tree is or is not a good idea. There are obviously vocal proponents on both sides of this issue. In this post, I am examining the issue of the viability of wikis and explaining why I believe they work and work well handling and organizing huge databases. In doing this, I am, of course, addressing the issue of whether or not the FamilySearch.org Family Tree program will do or even can do, what it is supposed to do, that is, supply a central clearing house for the world's genealogical data.

So what make a wiki tick (or squeak or whatever it does)? As I have said quite a few times before both in print and in classes on wikis, the answer is counterintuitive and relatively complex. In short, there is no simple answer. I guess my initial question would be why do people think that it does not work? I could understand this irrational reaction if wikis were newly minted and just making their appearance on the Web. But the first wikis have now been in operation since 1994, when Ward Cunningham invented the concept and gave it its name (he gave the name "WikiWikiWeb" to both the wiki, which ran on his company's website at c2.com, and the wiki software that powered it). See, of course, Wikipedia: History of wikis.

Whether or not wikis work is not something you have to take on faith. We have almost twenty years of experience and the world's wikis are operating very well, thank you. So why do people continue believing the wikis won't work? The response involves human nature and the fundamental beliefs of humans about their own human nature. The detractors automatically assume that allowing everyone access to a database will degrade the quality of the data. They apparently believe that human nature will always produce a flawed product. Ergo, all wikis must be flawed. Actually, the opposite is true. Giving everybody free access to the data ultimately makes the data extremely reliable and accurate. The more data in the program, the greater the ultimate accuracy. In addition, the accuracy of an active wiki improves over time while traditionally managed databases usually degrade over equivalent time periods. Part of the reason is that it is basic to human nature, if given the opportunity, to constantly strive to improve and progress towards perfection. The dichotomy of the destructiveness of human nature as opposed to the view that it is basic to human nature to work towards perfection is one of the most basic of human controversies and has been discussed by philosophers and religious leaders since the beginning of public discourse. As you can tell, I come out on the side of those that believe perfection is possible. See Wikipedia: Perfection.

For this seemingly contradictory principle of the integrity of wikis to work, requires a distinction between dynamic or active databases and static databases. If you have a stack of documents (books, photos, etc.) and load a description of each item into the database, the accuracy of the included data is determined at different levels. At one level, the accuracy is determined by the integrity of the data seeded in. Since the database is static, there will be no changes to the data after it is included in the database and so the accuracy, if you will, of the database is fixed at whatever arbitrary level the seed data had obtained. This was one of the basic flaws of the New.FamilySearch.org program. The data set was flawed and there was no way to make corrections. Users could only add more, sometimes contradictory data, thereby degrading the accuracy of the whole until it became, in some cases, totally unreliable. It was unreliable because there was no way to determine which data was accurate and which was not by reference to the program itself. Each user had to determine whether or not the data set he or she was presented with was accurate or not. The only safe tactic in this situation was to assume that all the data was unreliable. This is also a major basic flaw of all online individual user based family trees.

I learned this principle in my early university courses in drawing and painting when I was a Fine Arts major. As you began working on a drawing, the more you worked on your own drawing, the less likely you were to see the flaws. You had to stop, put it away for a while and then come back to the same drawing (or painting) to see what you had done wrong. I still find the same thing when I re-read some of my blog posts. It is this process of continued examination that induces progress.

Wikis fall into the category of an active database. Entries that are incorrect or simply wrongly entered, can be corrected by anyone who sees the problem. In our modern jargon, this is called cloud sourcing. This also reflects the actual reality of genealogical relationships. The descendants of any given ancestor all have equal access to information about that ancestor. In a sense, this is not collaboration, it is mutual cooperation in a common goal. Collaboration implies voluntary cooperation, in wikis the cooperation is not voluntary. The success of wikis such as the FamilySearch Research Wiki demonstrates the viability of this type of cooperation.

However, there is another level to the wikis that is being ignored by the detractors. Wikis can be moderated. In the case of both the FamilySearch Research Wiki and the Family Tree, these are moderated wikis. In both programs, FamilySearch has made it clear that certain types of behaviors are unacceptable. See FamilySearch Wiki:Policies. People who violate these policies can be barred from the wikis. This point has been made over and again during presentations made by FamilySearch representatives talking about the FamilySearch.org Family Tree program. Since I have been actively engaged in moderation of the Research Wiki for some years now, I am in a good position to know whether or not this works. I can unequivocally say that it does work. The conflicts about content predicted are possible but is quickly terminated. In those cases one or all of the participants in the controversy can be barred temporarily or permanently from the wikis.

Another reason conflicts are rare and can, in the vast majority of the cases, be resolved quickly is that only a very small percentage of all of the people involved in any given topic (or family line) will actually participate in adding and correcting the information. Generally, only the most dedicated and involved individuals contribute and correct. Most of the unreasonable people will not take the time to contribute or even know of the existence of the Family Tree.

Now, whether or not you believe what I say is immaterial to the reality that wikis do work. A unified family tree is not only possible, it is a working reality. In my opinion, it is the only presently known practical way to address the huge amount of genealogical data with any assurance that the inherent flaws will be corrected at some point in time. Whether you wish to share your data with the Family Tree is your own personal decision. From my perspective, the loss to the Family Tree will be minimal because it is highly unlikely that your research is in anyway unique or could not be duplicated. But your loss will be immense. You will lose the cooperation of all of your other interested family members, especially those of whom you are not aware. You will also spend considerable time duplicating research done by others.


2 comments:

  1. Good humor in this great post! This is what I call a very good way of driving home the point. Keep it up guys and I sure like this place…
    laser gum surgery florida

    ReplyDelete
  2. I believe in Wikis, in the wisdom of the crowd, etc, etc. However, the truth is that in my own experience re genealogy - *there*is*no*crowd*. Maybe you have a different situation in the LDS church, but out here...

    "it is highly unlikely that your research is in anyway unique or could not be duplicated. But your loss will be immense. You will lose the cooperation of all of your other interested family members, especially those of whom you are not aware"

    Oh, if only that were true... My research could be duplicated, yes. (Actually, that's not true either - most of the stories about people in the recent past will be lost if not written down. A good reason why we need a working wiki.) As for "all of your other interested family members" - if only, if only. So far as I know, there have only ever been 4 people doing primary research into the story of James BRUCE, weaver of Dundee (abt1754-1819). Two of those are dead and the third (who helped me immensely) isn't actually a relative, so the chance of him handing on stuff is limited.

    Similar stories are found across my tree. Yes, there are several cases where I have had immensely useful collaboration. But in many cases, I seem to be the only descendant with an interest in genealogy. And so I would really, really like to use FS FT to pick up other potential collaborators.

    But unless and until FS turns FamilyTree into software that helps me contribute my stuff in a way that helps both *me* and *everyone*else*, then it's never going to get good reports off people. The issues are not the attitudes that "Wikis don't work" and "It's my data" (though clearly they exist) - the issue is that FS FT is a long way from being a useful collaborative tool.

    ReplyDelete