Sorry to be a curmudgeon, but I'd prefer not to see folks referencing the rip-off thief of an inaccurate website known as the Pocket Watch Database.
Can you enlighten us?
Here are the issues that I have:
1. The site uses NUMEROUS sources that are quite obvious if you're familiar with the literature available. The Hampden references are a direct rip-off of Henry Burgell's work, which was derived(with permission) from Jim Hernick and Bob Arnold's Hampden book(BTW, Bob says he's going to get around to revising the book "one of these days" and I hope he's able to). The Illinois stuff comes from Russ Snyder's database, along with the Erhardt Illinois book. Russ wasn't aware of it until I brought it to his attention, while a good friend of mine has the copyright to the Illinois book and is working on a revision of it.
The problem, though, with using these sources is that NONE of them are cited-the information is simply there, and again it's not used just a reference but as a blatant copy.
2. This really wound me up. The NAWCC is in possession of the Hamilton ledgers in their library. About 30 years ago, they were microfilmed and specific pages were supplied at no charge to NAWCC members or a small charge to non-members. About 7 years ago, 5 folks(two of whom are very good friends of mine) donated $1K each to have the ledgers digitized, and they are online as an NAWCC benefit for members. Again, they are also available for a small fee to non-NAWCC members. The database has a COMPLETE transcription of the ledgers available free for all, and again with no reference to the fact that the ledgers are property of the NAWCC or even that the information came from them.
3. The site invites corrections by observation. Unfortunately, this system is inherently flawed. I hate to sound snobbish about this, but observations of watches out of the ordinary really need to be vetted by folks who know the brand well to differentiate between legitimate factory oddities and put-together pieces. The author of the database takes observations at face value.
Also, I still try to keep information accurate, and, as an example, I had a round with the author over the first run of Waltham 1892 Vanguard model movements. I specifically collect movements from this run, so I feel like I know them fairly well. I have a half dozen and have handled probably 30 since I specifically watch for them at shows. The ledgers list the run as 17j and pendant set. As they stagnated, some were changed to 21j and some were changed to lever set(and of course some got both changes). Early on, the database was changed to 21j and lever set based on a single reported observation, and it took a LOT of emails back and forth(along with a lot of photos and other work on my part) to get the point across that the ledgers were accurate but that watches were factory up-jeweled and otherwise changed. The problem was compounded by the fact that the 17j Vanguards weren't marked with a jewel count, and many observations "assumed" 21j when nothing was marked("why would they make a 17j Vanguard?")-I had to spend time pointing out the number of uploaded photos that actually showed 17j watches. It was FINALLY noted as a split run, which was a reasonable enough compromise but still not totally accurate.
4. This one is probably a bit more nit-picky, but it pertains specifically to the Illinois section of the database. Hard core Illinois collectors(and yes, I'm one of them) break down watches into multiple variants. Frankly, if I'm looking up a Bunn Special, I don't care about how many were made that were marked "Bunn Special"-I care about how many were made in that size, jewel count, pattern, and marking variant.