Yeah, I haven't done anything in quite some time, but I still have all of the stuff saved at home. Some of the data (newspapers, COOP records) was either hard to find or process, but it's all out there. The Blizzard Book was a neat throwback -- many of the accounts of the storms it lists from before 1900 can be found on free(-ish) newspaper archives from the New York Times and Google.
I based any of my maps from before the mid-90s entirely on COOP observations, which are inconsistent and geographically sparse, so they're very approximate. I'd do the maps differently nowadays. At the time, I overlaid multi-day totals on the map and tried my best to make contours that took into account the local tendencies and variations we experience. Later I attempted to create a mapping program in Python, but I was unsatisfied with the results (I needed to deal with way more data to make somewhat decent maps than my skills or time allowed, and it's really hard to make a program that considers local geography when plotting contours). It could absolutely be accomplished, though -- anyone that knows how to deal with masses of data, which can be pulled from databases in commonly used formats, could make a decent reconstruction of daily weather across the eastern US from 1893 to the present. I vaguely recall a product from some NOAA branch that sort of did this, but the execution wasn't exactly how I'd envisioned it.
Someday I'll revisit everything and do it my own way.