Goodreads Librarians Group discussion

1) Otis is going to have to field this one. I've noticed the same issue with other three-part names (like those with a Jr.)

2) Fixed. (Amazon seems to have a number of odd errors like that one, and it was the source for that book.)

Hi John,

1. I believe (and correct me if I'm wrong), that the correct way to list an author is: "Lastname, Firstname Middlename". However if "Garcia Marquez" is actually a two word lastname, our algorithm to break apart a 3 word name doesn't work for his case. We will have to think of a way to correct that, but I can't think of an easy solution.

2. Thanks Rivka!

Would it be possible--and I'm not a coder, so I have no real idea--to merge it into one? GarciaMarquez isn't perfect, but might work a bit better for finding the author in question.

-Rob

Otis, the author's middle name is José. ;) In most Spanish-speaking countries, surnames work a little bit differently than here. It's sort of like the Russian system, except instead of a patronymic and a family name, it's father's last name then mother's last name. It's not precisely a two word last name, but that's probably the closest American approximation.

Here's the wikiarticle on the subject, if anyone's curious: http://en.wikipedia.org/wiki/Spanish_...

Oh, and http://en.wikipedia.org/wiki/Russian_...

I had a thought about tripartite names. Might it be possible to have separate fields for first and last name? That way Garcia Marquez, Gabriel and Tiptree Jr., James would both be listed correctly. I realize that the importing-from-Amazon thing would make the fields not match. Would it be hard to have it automatically stick the last piece in the "last name" slot, and the rest in the "first name" slot? Then three-names or other irregulars could be corrected by librarians.

Rivka,
You are correct. The Library of Congress lists it as Garcia Marquez, Gabriel. I learned this when a native American wrote Blue Highways. His name was William Least Heat Moon. the official listing was Least Heat Moon, William.

Cheryl aka Eager Reader

I would suggest that, when in doubt, consult the Library of Congress authority records.
http://authorities.loc.gov/

In this case, "Garcia Marquez, Gabriel" is the accepted form.

So, if you search "Marquez, Gabriel Garcia", it will refer you to the correct authority record.

If consistency is what we're after, I'd recommend following LOC standards...

"I would suggest that, when in doubt, consult the Library of Congress authority records. http://authorities.loc.gov/ "

Yes, and wouldn't this solve the identical names problem as well, without the kludge of adding extra spaces? LoC gives each author a unique name, often by adding dates and/or a fuller form of the name.

Smith, John
Smith, John, 1931-1995
Smith, John A. (John Angus), 1936-

Since all this work has already been done, why reinvent the wheel? (I apologize if this has been discussed ad nauseum before--I'm a recent arrival and plead ignorance!)

Ben

Even the LOC has a hard time with it still. Check http://nielsenhayden.com/name.html for a one-paragraph rant (complete with bad poetry!) about the problems book editors and writers Patrick and Teresa Nielsen Hayden have, inlcuding the different ways the LOC handles them. They chose a two word, unhyphenated last name, much like the Mac Whoevers and Le Whomevers and De Whatevers of the world, but one that is not a well known foreignism :-(

A very quick glance suggests that Goodreads has--unlike my local Borders--both correctly collected all of Patrick's books and correctly separated him from the philosopher Patrick Hayden.

Hi all,

Rivka, I think you've hit upon the needed solution. We need to separate the fields, and let librarians make the corrections for exceptions.

The Library of Congress does indeed have a solution to the same-name problem, however there is no clear way for Goodreads to take advantage of it, since our data comes from Amazon, and often all the data we have is the authors name and the title of their book. We do have ideas to improve things a bit with some LOC data, but so far our Librarians have been doing such a good job it just hasn't been a priority.

So, did this separation of last-name fields ever get implemented? For instance, I was about to fiddle with John le Carré's entries to make him "le Carré" as LOC lists him, but I think he still gets stuck in the C's.

Benjamin wrote: "So, did this separation of last-name fields ever get implemented? For instance, I was about to fiddle with John le Carré's entries to make him "le Carré" as LOC lists him, but I think he still gets..."

No, this was never put in place, so right now there is no way to properly sort names like "le Carré". You'll have to leave them under "C" until this feature is available.

Isn't it possible to make a 'sort by' field for author names as it is for books titles as there are so many different situations?
Dutch names could also give some difficulties http://en.wikipedia.org/wiki/Tussenvo...