Breaking author links via up-merging

[Keith]

I've noticed recently an increasing number of long-established author profiles being up-merged from 3- and 4-digit profile numbers into the 7-digit range, often with the name having been changed from the common version most likely to be imported from other sources to something that someone probably considers to be "more correct" or "authoritative." A notable example is the former Seneca http://www.goodreads.com/author/show/... which is now Lucius Annaeus Seneca http://www.goodreads.com/author/edits...

Arguably the latter is more technically correct, more disambiguated, and perhaps clearer. However, most seekers after Seneca won't know that this is the full name they seek, in addition to this breaking every previously-existing author link to the old profile, and causing everything subsequently imported to go under a new 7-digit Seneca link which will then have to be remerged.

Personally, when I find authors needing to be merged, I always merge down-stream (unless the profiles are only off by a few positions), to the oldest existing profile, under the presumption that the oldest is likely to be the one with the most existing links. If the name needs to be adjusted (like when the old one is "Surname, Firstname C., Surname" while the new one is the correct "Firstname C. Surname") I do the down-merges first and fix the profile name at the end of the sequence, to avoid this very issue. Is this not the current policy, Best Practice, guideline, whatever?

I'm really hoping this isn't some Librarian off on a crusade, or the author links are about to become largely useless. In the Seneca example, I can't tell who the culprit was or I'd try discussing it with them directly. It might have been Josh Clark (who at least changed the name, and generally seems to do very good work), or Benoît Gateau (who at least tired to keep a link to the old Seneca page) or Joe Ungureanu (who made a couple of profile updates, but doesn't seem to have enough edits to be the overall culprit). However, since the merges themselves do not appear to be tracked in the edit history, it could be someone entirely unlisted there.

If this is as wrong as I think, is there some way we can stop it? I recently managed to move Mark Twain back down to the 4-digit range where he belongs from a 5xxxxxx "Samuel Langhorne Clemens" entry, reconstructing the profile from the Wayback Machine, but usable 4-digit profiles are at a premium and that kind of movement is a giant PITA, as well as a waste of time if someone is only going to edit-war them back into silliness.

[rivka, Former Moderator]

While I agree that it's best not to change author names to "more authoritative" (whatever that means, given the number of possible authorities in some cases) from the ones that are commonly used, I don't worry so much about whether an author has an "old" ID number of "new" one.

Breaking links in annoying, but it is going to happen sometimes, regardless.

[Banjomike]

As long as there are people who are convinced that Goodreads does not already have entries for authors like Shakespeare, Byron, Oscar Wilde, (or Seneca) and others of similar fame, there will always be people who add "new" and "improved" versions of them to make up for the failing of the rest of us.

I don't always merge down-stream, I see which one has the most books, most reviews, best photos etc

[rivka, Former Moderator]

Banjomike wrote: "I don't always merge down-stream, I see which one has the most books, most reviews, best photos etc"

That's what I usually do as well.