Conflicting information in public databases

Discussions on Records, Recording, & Artists
Post Reply
CPBarnum
Victor Jr
Posts: 26
Joined: Sun May 06, 2018 1:04 pm
Personal Text: Longtime music lover, Grafonola newbie

Conflicting information in public databases

Post by CPBarnum »

I love collecting generally, and am relatively new to collecting 78s, but man -- I'm trying to put together a basic spreadsheet of my records, and my OCD gene is grinding its teeth.

What's up with conflicting information in public databases? One small example: Tommy Dorsey on RCA Victor 27520 (Star Dust / Song of India) is listed as 1940 in Discogs and 1941 in 45worlds (over on 78discography, meanwhile, dates of 1940 and 1937 are provided for the A and B sides respectively, but those are recording dates as opposed to release dates).

Are any of the online databases viewed as more authoritative than others? 78discography is a tremendous site, but if all the dates listed there are recording dates, not release dates, how best to determine release dates? So for something like, say, Victor 16273 (Arthur Pryor's Band / Sousa's Band) with recording dates of 9/14/08 and 10/21/08 respectively, is it reasonable to assume a 1908 (or maybe 1909) release date too? Are there better resources for release dates than Discogs and 45worlds, or is it pretty much a triangulation crapshoot using these three sites?

And one more sort of related question: While it's intuitive that the higher the catalog number the later the release, that's not always the case, is it? Another very simple example: RCA Victor 20-2005 (a Tommy Dorsey disc) is listed in Discogs as having a 1948 release. Meanwhile, 20-2410 (a Glenn Miller disc) is listed in Discogs as having a 1947 release.

Thanks, gang

User avatar
gramophone-georg
Victor VI
Posts: 3984
Joined: Mon Jan 20, 2014 11:55 pm
Personal Text: Northwest Of Normal
Location: Eugene/ Springfield Oregon USA

Re: Conflicting information in public databases

Post by gramophone-georg »

Problem is you're dealing with recording dates v. release dates and I think the online databases get confused a lot.

The other problem you're running into is that sometimes there were multiple takes of a record made and issued. Sometimes records were re- recorded after an initial release for better quality. Then, in the postwar era, they changed the width of the grooves and recording curve, and dubbed lots of earlier recordings into the new format. Then there were reissues and later issues of original pressings such as Sinatra's early Columbias with Harry James. These didn't sell well in 1939-40 but were reissued after Sinatra hit it big with Dorsey and went solo with the same catalog numbers and matrices- only the label now put Sinatra's name first, instead of as "vocal chorus by".

It's all confusing, even for someone like me that has been collecting for almost 50 years.

My go to authority is Brian Rust's "Jazz Records" and "American Dance Band Discography". While there are errors there as well, it's quite an exhaustive study compiled using record company records and personnel lists as well as extensive interviews of surviving leaders and musicians of the era.

Rust's main limitation is he stops at mid 1942.
"He who dies with the most shellac wins"- some nutty record geek

I got PTSD from Peter F's avatar

User avatar
PeterF
Victor IV
Posts: 1913
Joined: Sun Jul 08, 2012 1:06 pm

Re: Conflicting information in public databases

Post by PeterF »

Not a limitation. An advantage! (Wink)

Post Reply