Some of the data itself is from the public record, but compiling all of it yourself would be prohibitively expensive. It would take many trips to various libraries, churches, courthouses, etc. in lots of far-flung places in different cities, states, and countries. And that's all before you start indexing, searching, copying, or scanning any records.

Then consider they're hosting all those billions of records, have developed decent software for both building family trees and documenting them with their database of sources, and continually improve their software and increase their database holdings. The only reason they can charge so little is the economy of scale based on their huge userbase.

For me it's mostly just a hobby. I love history to begin with, and my family history is specifically interesting to me personally. Finding out where my ancestors lived, what their lives would have been like, and why they immigrated to America is fascinating to me.

I've read that genealogy is the second most common hobby in America, after gardening. Of course, that depends on how a hobby is defined. I consider reading my favorite hobby. However, for some people, i.e. Mormons, genealogy has religious significance. That's why Salt Lake City, Utah is the biggest destination for many people doing genealogical research.

I've also read that, which really interests me. Don't you find it fascinating that your (or at least some modern Turks) ancestors possibly spoke Hittite or another Indo-European or even Semitic language? I recently paid for a genealogical DNA test, which will tell me more specifically where my ancestors came from. My paternal ancestors immigrated to America from Germany in the 1880s (then Prussia), but there are several clues which point to them actually being ethnically Lithuanian. The area where they lived has been part of Lithuania, Poland, Russia, and Germany in just the last few hundred years.

Actually, every person on the planet is descended from immigrants if you go back far enough.

Census records are released 72 years later, to be exact. Also, although the US Census isn't available until then, many state listings of much more recent births, marriages, divorces, and deaths are available.

Too much data is far more preferable than too little. Back in the old days, before many sources were available online, most searching had to be done by hand, paging through entire paper records until you found a possible match. The earliest online sources were often just indexes to paper records. It made it much easier to find the records, but then you still had to travel to a library, courthouse, church, etc. that might be in another state or even country.

The reason the algorithm is so promiscuous is that it's probably using Soundex to match names. That's a good thing though, as I've found some of my ancestors names spelled in lots of different ways.

Definitely agree, the anonymous reader didn't try very hard. And records for living people are necessarily scarce because of privacy issues.

Actually, just creating and editing your family tree on Ancestry is completely free. And there's no limit to the number of people, AFAIK. The paid service is just for accessing their database of sources and connecting with other users who might have overlapping family trees.

I pay for ancestry.com because they have a huge database of sources online, and probably the largest number of users out of any other family tree websites. It's much more convenient for me to search and view their literally billions of records from the comfort of my own home than going to my local library or LDS center. If I did that, I'd have to copy their records and scan them myself. Ancestry's already done the legwork for me. Also, since they have so many users, I've connected to several other people doing research along the same family lines, which saved my quite a bit of time.

But that's really a non-starter with ancestry.com, they let you export your family tree as a GEDCOM file any time you want, which is the de facto standard file format for genealogy.

I agree wholeheartedly! Currently, I use ancestry.com for my main family tree, but regularly download a GEDCOM file and import it into GeneWeb (locally hosted) for offline access. I've tried several other programs, both open source and proprietary, but GeneWeb is my favorite by far. It doesn't make the prettiest family tree website, but it's easy to use and its functionality is great.

Even though the primary developer, Daniel de Rauglaudre, isn't developing it further right now, it has more features than I use regularly, and it's highly customizable. I'm somewhat tempted to learn OCaml, just to be able to modify this software.

Exactly! If I want to make my information available only to some people and not to others, that's MY right. And if a "large social-networking site" helps me do that, good for them, and I am more likely to use them than I would a system that says: "you MUST make ALL information public, or not share it with anybody at all."

I agree with Berners-Lee on his other points, but not that one. And I don't know why he would lead with that one.

I can't tell if you're being pedantic or stupid. I'm leaning toward pedantic asshole at the moment, as you are hyper-technically correct but not actually answering the question which anybody over the age of 5 understood was being asked.

Yes, CDs are "digital" and they are being sold "online." That's not what UnknowingFool was asking, of course. He is correct that this is the first time the Beatles' catalog will be offered online for downloading, legally.

Very easy. You sign up to buy their insurance on-line. They grab your useragent string when you make the purchase, and they stick that in their database of information about you, nice and associated with your customer number.

Then they run a query on claims made. They filter the result to show only claims made by people who bought insurance on-line. Then they average the claims amount by browser type.

Or, they do a random query to pick 10,000 people who bought their insurance on-line last year. Then they count number of claims by browser type.

Insurance is a game of statistics. Yes, some browsers lie in their user agent string. Yes, sometimes people use other people's browsers. But MOST browsers are what they say they are, MOST people use their own machine pretty regularly, and if the insurance company is really doing this (which I take no position on), then you can bet they have statistics to back up their belief that there's a difference in insurance risk, ON AVERAGE, between users of different browsers.

The insurance company couldn't care less if the correlation holds true for every single instance. They know that it doesn't, in fact. But if it holds true often enough, then they can use that data to offer some people a better price on the insurance, because they're statistically less likely to file a claim.

Exactly. I rather tire of seeing the self-proclaimed geek elite decrying these users as "stupid" and "ignorant." No, they just have different value systems then the uber-security-conscious. Lots of people in rural areas regularly leave their doors unlocked. Just because a hacker COULD get access to their account at a Starbucks doesn't mean that the odds of it happening at any particular Starbucks at any given time is terribly high.

Was it idiocy for the folks at this Starbucks to stay online on Facebook even after being warned by this hacker? Clearly from the warning he provided, he wasn't intending to do harm to them. You're a nice rural family sitting around the coffee table, and a nice man sticks his head in your door and says "just wanted to let you know, your door is unlocked." Do you expect the folks to get up and run around and lock every door in the house?