Goodreads Librarians Group discussion

Could something be done about all the JR. author names?

To see the problem slecet the add book/author link.
go to an author search
type in JR and hit search

YIKES!

I'm slowly working my way through them

I have been correcting obvious author name errors for about a week now. But I was hoping that somthing might be done in an automated fashion for all the JR problems

Unfortunately, we get a lot of messed up author names in the data uploads - try running the same search for '@', '&' or '.', not to mention 'dr' and 'dr.', 'ms', 'ms.', 'mrs', 'mrs.', and so on...

Carolyn wrote: "Unfortunately, we get a lot of messed up author names in the data uploads - try running the same search for '@', '&' or '.', not to mention 'dr' and 'dr.', 'ms', 'ms.', 'mrs', 'mrs.', and so on..."

That's understood. But I was hoping that some automated means of dealing with the worst systemmic errors could be found so as to allow the human librarians to deal with those requiring a bit more cognition.

Automated fixes tend to be more trouble than they're worth. (Like the M.D. one that messed up authors named Michael David who publish under their initials . . .)

I was not aware of that automated fix. Isn't there a way in SQL to specify that only those with M.D. as the last four letters of the author name field be changed? I seem to recall that there was.

Stephen wrote: "I was not aware of that automated fix."

It was gotten rid of a long time back. It was an example of why a thinking human librarian works better for these things than trying to come up with complex scripts.

FYI, M.D. as a title is often not imported as the last few characters.

I agree that thinking human librarians are best, however I was looking for a way to eliminate some of the larger, less complex tasks, so that humans could concentrate on what's important, and what they're better suited to.

As to M.D. not always being the last four letters... I'm aware that there are many, many formats but if the most common errors are eliminated, then the less common ones can be dealt with by humans more quickly. If M.D. is only the last four letters 50% of the time, that's STILL a lot of corrections that would not be needed from human librarians.

Stephen wrote: "If M.D. is only the last four letters 50% of the time"

I doubt that's even close.

Have to agree with rivka on this one - I've seen the MD with and without initials, in front of the first name as well as inserted into the middle of the name - that doesn't even begin to take into account all of the variants on 'Doctor', as well as all of the other professional/educational credential alphabet soup out there.

OK but isn't that sort of missing the point? assume that there are only 1000 instances of m.d. being the last four characters of author names... wouldn't eliminating those, help concentrate the human librarians on the tougher calls?

Still I can sense the reluctance. It was only a suggestion.

Stephen wrote: "wouldn't eliminating those, help concentrate the human librarians on the tougher calls?"

If and only if it didn't cause more problems than it solved AND if it didn't require an excessive amount of developer time. Neither has proven true in the past.

Stephen wrote: "OK but isn't that sort of missing the point? assume that there are only 1000 instances of m.d. being the last four characters of author names... wouldn't eliminating those, help concentrate the hum..."

You can work with too much information (so, MD when it's not needed), but you can't work with lack of information (you don't know if MD is needed or not).