I've also mentioned having some fun with the Seattle Times Historical Archives, which I subscribed to for a month or so -- I had some unusual surnames and got the fun of discovering that it wasn't just one of my uncles who smashed up a car while drink driving (and finally got some details on the drug conviction of one of the other uncles -- and let me just say, wow, they used to prosecute and write up in the newspaper very tiny quantities of Mary Jane).
I eventually subscribed to genealogybank.com, which gave me enough information on a friend's parents' who died in an airplane crash to really make some progress on her tree. I haven't found a ton of stuff there outside of Seattle, and can't really tell if that's because I am not using it effectively, or the set of papers they have doesn't overlap with my tree very well.
I knew about newspaperarchives.com, however had decided not to sign up for another service, since I wasn't having tremendous results. Then I was looking at it again and realized they had a few decades of the Steinbach Carillon, a Manitoba newspaper for an extremely Mennonite area. I signed up on the off chance I'd get an obit or two, and am still slogging through the flood of information. It has truly moved this part of my tree from nearly-impossible to shooting-fish-in-barrel.
The basic problem with this particular group of people involves (unsurprisingly) the naming patterns of the community, their reproductive patterns, and their historic relationship with the provincial government. They reuse names like crazy (they won't just name three stillborns in a row the dad's name until they get one that survives for a few years -- they'll reuse it if the kid dies as a teenager). They had a _lot_ of children (in the first generation born and raised in Manitoba, it was not uncommon for a dozen children of one set of parents to make it to adulthood and all marry and have kids of their own). And as near as I can tell, they didn't reliably report births to the provincial authorities. I could be wrong about this -- it's possible the births are all recorded and not yet online, and there are very real problems with getting the right version of a name (Aganetha/Agnes/Neta, for example, or Marie/Maria/Mary, and I will not even start on the John and Henry variations). While there are provincial surveys and national surveys available, the name punning is, er, punishing to slog through.
To make matters even worse, however, this community came over starting in the mid-1870s from Russia (what is now Ukraine), and people kept coming over for a while, interrupted by things like major wars. While part of the community (most of mine) went to Manitoba, a larger group went to the American Midwest (especially Kansas), and there was substantial back-and-forth migration, especially in the 1920s and 1930s, resulting in families that start in Canada, have a couple kids there, move to Kansas, have another kid or two there, then return to Canada, and have still more. Oh, and the name-punning problem spans the entire group.
One of my great-uncles did unbelievable work putting together a family register in the 1970s. While he clearly viewed this as a labor serving the family and its religion, he did not leave people out just because they were no longer Mennonites, or because they had switched from the Kleine Gemeinde (Evangelical Mennonites) to some other variation. While I have some issues with the work he did (zero geographical information. I mean, NOTHING. Yikes.), the more I work with this information, the more impressed I am with what he was able to accomplish and publish with pen, paper and a typewriter, and a private printer. He was defeated by people who outright lied to him, but he didn't make many mistakes.
But it dates from the 1970s, which means a lot of people have since died, thus in principle giving me an opportunity to use obituaries to piece together all the people who married into the descendancy that he focused on. newspaperarchives.com, by including the Steinbach Carillon, is the first real chance I've had to access those obituaries.
I'd be a little more excited, if the print quality of the original paper weren't quite so bad. It isn't the archives fault -- their other papers are quite clear. Ah, well. Be grateful for what we get, right?