The usual disclaimer: Though I may speak here about technologies related to or released by my employer, the views expressed here are my own and do not reflect those of my employer. BTW, I hate having to put this disclaimer in front of everything that may remotely have to do with my employer. Smart people should be able to discern the difference between thoughts of a person typing on their personal blog vs. thoughts of that persons employer. That’s a rant for another day though.

Today I went on another genealogy kick. Fueled by one of mom’s bombs (“You know your grandma was married once before she married your grandpa don’t you?”) that she likes to drop every now and then, I went on a tear looking for links in my family tree that weren’t obvious to me before.

After a few hours I’d made little headway into fitting the new information into what information I already had. I decided that I needed to comb through census records manually (looking at pages of information laid out before me, sorry trees) to find non-obvious links between people and families (See those people next door with the same last name? Turns out you’re related!).  Unfortunately using the most well transcribed census database on Ancestry.com is an exercise in frustration.

Since I was not permitted to dump entire portions of transcribed census, my choices were to print the raw pages and transcribe myself or just work directly off the raw pages. Neither of these choices were very palatable to me so I went off searching for other sites that had transcribed census data. I happened to find the USGenWeb Census Project. This is a volunteer effort of people manually transcribing pages of census data so that others may freely use and search it. In theory this is a GREAT idea. In practice, there just aren’t enough people in the world who would bother doing this sort of manual labor. Thus many pages sit in their database, untranscribed and unsearchable.

If you think like I do, you’re probably reading this thinking about the obvious solution to this problem; OCR technology. Though it’s out there free for use by anyone and being applied by big companies to big things, nobody has bothered to apply this technology to the genealogy problem, which has been around longer than any PDF you can dredge up. Sure Ancestry has their crippled, for-profit, search engines but this limits what is otherwise public data to the set of people with the means to pay for it. This, though by no fault of Ancestry.com’s, is the heart of the matter.

IMO, technical people have this desire to always always move forward and do New Things. This is not a terrible thing because this is how innovation occurs. However this means that if a problem is solved in one way, nobody wants to be the one to return to that problem and solve it better. For example, if Ancestry.com were never invented someone would see all the problems that Ancestry has solved as new and challenging and would be applying all these newer technologies and ideas to the problem. Instead Ancestry exists and there doesn’t seem to be anyone anxious to revisit the problem except people who aren’t super technical for the most part. This leads us back to the people manually transcribing census data.

These problems that can be solved better or differently exist all over the place. See Brad with his proximity based garage door opening phone. See robotically cleaned floors. See all sorts of everyday activities that can be improved upon with the application of fairly common technology.

Take your laundry for instance (sorry people who subscribe to my FriendFeed, you’ve heard this one before). Right now most of us complete the task of separating clothes, measuring out detergent, starting the washer with the right settings so that the result isn’t shrunken sweaters and bleached jeans. If you’re over there thinking like me again, you can see the solution to the laundry “problem” fairly simply. Sorting based on detected colors built into the laundry unit. Programs that run based on the color detected. Detergent/Bleach/Fabric Softeners containers that disperse based on the program being run. Water temperature determined by the same program. All these problems have been solved but it will take someone to go back and apply and implement them properly so that people no longer manually do laundry when they don’t need to.

I suspect most of the people who read my blog are involved in tech in some way shape or form. I have a request for you. When you see something in the world around you and think “That could be done so much better/easier…,” try to devote even the smallest amount of time to resolving the problem and making it much better/easier for the folks that come after you. Meanwhile, I’m going to return to figuring out how to apply all the OCR tech out there to census pages and you’re welcome to join me. Here’s hoping some freely searchable census data comes of our efforts.

No related posts.