Reliable place to get publisher info

[Vanguard3000]

I've been rebuilding my ebook library in Calibre and so far, I've been using the ISBNs to scrape publisher info from Goodreads, but I'm finding it to be somewhat inconsistent. The librarians at GR may have use any number of options, including but not limited to:

• Publisher (e.g. Simon & Schuster)
• Publisher office (e.g. Simon & Schuster UK)
• Division (e.g. Pocket Books)
• Imprint (e.g. Baen Books)
• Other (e.g. "Simon & Schuster / Star Trek" or just "Star Trek")

Presentation of the publisher name may also be inconsistent:

• Simon & Schuster
• Simon and Schuster
• Simon & Schuster, inc
• Simon & Schuster, Inc.

I've tried scraping from other sources and they all seem inconsistent like this. I don't mind checking them as I go but It's not always clear what's best to put in Calibre's "Publisher" box. The top-level company, or the "deepest" level, like an imprint. And is there a reliable place in the title or copyright page to get this info?

Thanks in advance.

[BetterRed]

Originally Posted by Vanguard3000

. .

And is there a reliable place in the title or copyright page to get this info?

Nope - I use the Manage Categories tool to tidy up 'duplicate' Series, Tags, Publishers etc on an ad-hoc basis, usually at the time I happen to spot an anomaly.

BR

[Karellen]

@BetterRed

I think the OP is asking for a reliable site to check/download accurate data. I don't think he asking "how" to fix it.

For the Publishers box, I add whatever is in the Copyright page.
I don't know of any one stop reliable site for this data. I use a combination of wikipedia, archive.org or the publishers / authors site.
I also check the copyright page of the pbook if I happen to own it.
You might find the same release on archive.org and check the copyright page there.
I don't trust Goodreads as a site.

Not so bad if you only add a book once every week or two, but if you are trying to fix a library of hundreds or thousands of books, it will be quite a bit of work

[BetterRed]

Originally Posted by Karellen

Not so bad if you only add a book once every week or two, but if you are trying to fix a library of hundreds or thousands of books, it will be quite a bit of work

Indeed - but unless you are seriously inflicted with OCD or in the middle of the book supply chain itself, as a consumer, or even a retailer, you probably don't want to differentiate between Hachette UK Limited and Hachette UK and maybe even not with Hachette Australia Pty Ltd, or Hachette Australia.

BR

[Karellen]

Originally Posted by BetterRed

you probably don't want to differentiate between Hachette UK Limited and Hachette UK and maybe even not with Hachette Australia Pty Ltd, or Hachette Australia.

... or Hachette Digital

Yep, definitely agree. I generally round down to the common name - Hachette

[theducks]

I strip the business Legal structure off (Inc, Ltd...)
I also use the Copyright page to locate the IMPRINT when possible

Baen Books, (actually, Baen logo has SF or Fantasy on some, so I do use those when obvious) and not not their distribution SS

And with the Randy Penguin buying up so many imprints, this is also why I prefer Imprint.

[ownedbycats]

The Find Duplicates plugin has an option for metadata variations, which can help to some extent.

Publishers are a mess.

[Sirtel]

I tried to clean up publisher metadata... once. The job was so slow that I just gave up. Publisher names are a real hot mess, yeah. I just leave them alone. Not worth the effort.

[theducks]

Originally Posted by Sirtel

I tried to clean up publisher metadata... once. The job was so slow that I just gave up. Publisher names are a real hot mess, yeah. I just leave them alone. Not worth the effort.

You can clean off the INC or LTD in the Tag browser, that does all uses with one change.

I also replace cases of Authors name (That has no other distinction) as Publisher with 'Author Published'

[ownedbycats]

Originally Posted by theducks

I also replace cases of Authors name (That has no other distinction) as Publisher with 'Author Published'

I have have a template in my general cleanup action chain that checks if they're identical and then blanks them.

[Sirtel]

Originally Posted by theducks

You can clean off the INC or LTD in the Tag browser, that does all uses with one change.

Yes, I know, but that was the least of it, unfortunately. There just are countless variations on a single publisher name. As I said, it was really not worth the effort going through all of them (depends on how many books you have, of course).

[Karellen]

For 505 books in my library, I have 96 publishers.

[Sirtel]

Originally Posted by Karellen

For 505 books in my library, I have 96 publishers.

I have about 7000 publishers for close to 40,000 books. Many of those are actually the same publisher with different versions of their name.

[Vanguard3000]

Ugh, so it sounds like it's basically all manual, which is sort of what I expected. I always ensure my books have copyright info so I'll have that to grab the publisher; and where it's missing, I'll try and create one from an image from OpenLibrary or eBay listings.

Thanks for the suggestions everyone. I'll have to just do my best to represent the publisher as best I can and weed out the worst aberrations.

[Wiggo]

Originally Posted by Vanguard3000

Presentation of the publisher name may also be inconsistent:

• Simon & Schuster
• Simon and Schuster
• Simon & Schuster, inc
• Simon & Schuster, Inc.

I hope I haven't misunderstood anything, but why not use the rules for the metadata source - also works with Regex?