Friday, August 28, 2009

Which search engine is "best?"

As Google, Yahoo, and now Bing duke it out to be everyone's search engine of choice, you can now do a blind comparison of the three for yourself.

Visit http://blindsearch.fejus.com/, enter your search term, and you'll get back a column for each search engine on what was found. The order of the columns changes with each search, but by voting for the one that gave what you consider to be the "best" results, you can find out which is which.

All claim to put the better matches to the top of the results list, but of course that's in the opinion of the software folks and how they decided to evaluate the results against your search terms. Your opinion matters, too. If you find one of them 'thinks' like you do, that search engine might become your first choice when you find a new ancestor!

Not only may you find some references to your family or your current brick wall you didn't know about before, but by experimenting with the search terms you use, you will discover how to choose better search words for each of the three.

Try it! It's kind of fun.

Sunday, August 23, 2009

Common sense and census indexes

I'm a firm believer in using all available censuses and the data they contain. But there's also a good reason to use common sense.

Minnesota is home to many different ethnic groups, and has been for centuries. However, if you are researching in Minnesota, be aware that the Ancestry.com indexes were not done by people who knew about Minnesota ethnicity. For example, the letter "I" used for 'Indian' as a race indicator looks a little like "J." Under instructions to use full words for the index, the transcriber reported the race to be 'Japanese' for some people were really Dakota Indians. In another, the letter "H" for 'half-breed' was expanded to Hindu for many entries for both Ojibwe and Dakota Indians. In the first case, it was a mis-read of one letter; in the second list, it was choosing the wrong term. Both kinds of errors can mess you up if you don't use a little common sense and look beyond the index.

Of course, it's always worthwhile to look at the original source, but especially if you aren't having luck with the index. If people working with hundreds of entries can make errors like these, what did they do to your ancestor's name or to the remainder of the information they indexed?