In a wiki direction
I’m back! :)
I’ve switched over to using a wiki for my productivity needs (to-do lists, research notes, etc.), and I’m more and more convinced that it’s perfect for genealogy research. It’s freeform, laughably easy to store, and the hyperlinks make it a piece of cake to stitch things together where needed (and when you think about it, researching genealogy is just that: you stitch evidence together to form people and then stitch them together to form families and lines). And wikis are collaborative, and most support revisioning.
The only real downside I can think of is that there aren’t any set fields for things like names and dates, so searching is basically just a full-text search unless you have some kind of parser that picks up fields automatically. (And it would be easy enough to add some codes to the wiki markup to let users specifically designate portions of text as dates or names or places or what have you.)
And in reality, doing it this way is far more flexible, which makes it much nicer when you move to the international playing field, dealing with hundreds and thousands of different languages and cultures. The rigidity of existing record managers has been acceptable, but we’re in a new era now, and we need something organic, something that can grow and mold itself to the needs of the user. And users have vastly different needs.
Posted in Design
Comment feed for this post
I’m also impressed with wiki potential - both for research and to collaboratively collect the stories of our ancestors. I’m surprised genealogists haven’t jumped on Wikipedia.
Have you thought about expanding something like the XFN microformat (http://en.wikipedia.org/wiki/XHTML_Friends_Network)? Potentially something like this could be used across the ‘Net to show relationships . . .
I agree that the freeform style is good for genealogy *research*, for keeping notes, trains of thought and future research paths, I think it hinders the storage of genealogical data as a whole.
The set fields that you point out as lacking are often the very fields that you or others are looking for, and I think the “record managers” that you refer to still have a purpose.
Perhaps we need a hybrid, with structured records for known data, but a freeform block, either per individual, per family, or per dataset, for research notes.
Back to the fields: MediaWiki, and perhaps other implementations, does support the idea of templates, which can aid in a common data structure. You can see examples of this throughout Wikipedia, under animal entries for their classification or under country entries for their statistics. I’m not sure if you can specifically search these fields, but it provides a visual structure, if anything.
–
Crwth
Moultrie: Good point — wikis are great for collecting stories. Wikipedia itself isn’t well-suited for genealogical research, I think, other than in the same way any encyclopedia would be (places and some basic facts about them), but projects like WeRelate.org are aiming at what you might call a genealogy Wikipedia. As for XFN, I’ve thought about it but only briefly; I’ll have to look into it in more depth one of these days… :)
Crwth: First off, crwth is the old Gaelic instrument whence my surname, Crowder, comes. Crwth became crowd, and the person who played the crowd or crwth was called a crowder. I was reading Sir Orfeo in my Middle English class a few months ago and ran into a mention of it, in fact. So, cool nickname. :)
You’ve got a good point here, namely that wikis lose some of their usefulness when it comes to structured data. I agree. In the past, the research end of things has generally been neglected in favor of structured storage, which is why I’m pushing for more freeform methods, but it would obviously be pointless to swing completely over to the other end of the room. Errors in extremity are errors regardless of which side they’re on. That said, I’m not quite sure where the golden median lies, and even with structured data we’re still not quite where we need to be — the only standard we have (GEDCOM) is over a decade old, and every attempt since then to replace it has failed. And I don’t think that those failures necessarily mean that GEDCOM is fine the way it is and doesn’t need to be renovated.
Templates would be good, especially for beginning users. (I’ll be writing about that — wikis and beginners — soon, incidentally.)
An additional reply on the wiki idea… It may be useful for stories, biographies, or research notes, but I don’t believe it’s applicable to genealogy data. I believe that genealogy system needs as much structure and validation as it can possibly have.
A few examples from the genealogy database that I designed, and have on my website… When entering a record, the system I have will not allow you to put in a death date earlier than a birth date. I have routines setup to cross-check dates of a person with the dates of their parents, spouses, and children to make sure it’s at least feasible. Data entry tries to make sure place names are entered consistently to enable color-coding, mapping, and statistical reports.
In a free-form wiki, I don’t see how any of this is possible, even if there are templates.