Text on non-combinations?

Abby, John and I are moving the author disambiguation thing along, but it's slow. We wondered if you guys would want to post disambiguation notices on authors, like Christopher Locke has. It would be a text box available on author pages.

Would that be better than the "master list" thread, or worse?

Better, I think. I suggest that it have a "confirm" option like the links box does though - sometimes an author may seem like he needs disambiguating but really he's just multifaceted.

I like the idea... first, having the data on the author page will make it easier to transfer when disambiguation is up; second, it'll look more professional (especially to new users, or the authors themselves!) when someone goes to the page for the first time- in other words, that we know what's going on; third, it'll save users time investigating authors that have already been identified.

It'd be nice to have a template to fill out, though; make them all look uniform, and easier:

X is actually ___ authors, X the author of _____, and X the author of ____. In the future LibraryThing will be able to split authors with identical names. At present, it cannot.

Of course, there's this question: should this text box break down as many of the books as we can? There are only 4 books to deal with in Christopher Locke's case, butJohn Gardner's text box, for example, would be very long. Maybe better to just put one or two for now?

I agree with nperrin on a confirm option -- I for one have been misled by two multifaceted authors so far.

And I agree with jjlong that it will look more professional and save users time. Some sort of template might help, but it shouldn't get too long and drawn out.

This is a good idea. It gives people an idea of the potential issues with authors.

Good idea. It would also be helpful if you could turn off the check box that automatically combines authors if you search. When uncombining authors, you often have to go back and forth between the alternate names to make sure the uncombined works are going to the correct author. With that "combine" box pre-checked, it is just too easy to recombine them by mistake!

This is better than the master list it is at the source of the problem.
while the template might seem a good idea it will lack flexibility. I suspect people would like to use box for other things such as highlighting errrors.
Hypotheticaly informing us for example that Jane Austen did not write Lorna Doone. dubious data to the contrary.

How big a text space were you thinking of?

I'm surprised there hasn't been a groundswell of support for this idea, Tim!
If disambiguation is only a month or two away, maybe this would be just busy work; but if it'll be Christmas before it's up and running, I'd like to see this now.

If this is done, keep the statement short, otherwise, it will be ignored.

Even if disambiguation were to happen tomorrow, I think having this text on the author page is still a good idea (you would have to modify the wording a bit). It should help prevent user confusion. All the confused combining is not merely a lack of support in LT, it's a sign of confusion in the minds of the users of LT.

Yes, do it, keep it simple, with a form. The form should generate sufficient spaces for the number of authors (2, 3, 4...) with the same name. Hopefully a brief form fit to the exact # of authors with the same name will be read.

I dunno why short is necessarily better. People with libraries are supposed to be able to read, and even to like doing so. :-) And we all know how to scroll.

I think it would be quite helpful to show "Pen names used by this author" (including ones not yet entered on LT) --- a long list in a few cases --- and "Other writers with the same name(s)."

Btw, didn't I see something in Talk a while back about showing the full name of the author, with dates? Taken from MARC records or somewhere? Will that happen?

> 12

People with libraries are supposed to be able to read

You are forgetting the natural human tendency towards sloth. The fact is that when someone is madly combining things, and is confronted with a lengthy bit of next, they are likely to just skip it.

If you really think people with libraries actually do read in such circumstances, you've missed the zillions of author images quarantined because someone didn't read the instructions on the "Add a picture" page or the copyright notice right on the picture they uploaded!

Reading:
Some people will skip whatever you put. And some just want their books shared, and couldn't care less if that means combining Jane Austen with John Grisham. But more info would be helpful to those who are interested in trying to get it right.

Although I agree with most everything y'all have said above, I think we're a bit off track..

The point of disambiguation is not to stop ignorant or sloppy combining (and, after all, if a user adds a book by John Smith, it's going to go on John Smith's page. There is currently no other option.)... disambiguation can and will be used to clear up bad combining, authors with only last names, etc., but that's not the main reason for it.

The best example I know: if you go to imdb.com and type "James Stewart" in the search box, you get a page listing 10 or 12 James Stewarts; James Stewart (I), actor; James Stewart (II), art director, etc.

But what if an imdb user got a page that told him James Stewart was the art director on "Boys Life 2", the screenplay writer of "She Lives by Night", sound man on "Harold & Kumar Go to White Castle", and an actor in movies from the 1930s to 2005? With nothing that even indicated that James Stewart was more than one man?

The user would think:

This is useless to me.

What crappy data. How amateurish.

I can't be bothered with this mess.

or, worse: These people don't know what they're doing.

Yet that is exactly what LT users see when they go to an author page: say, John Green's. They see that he has written current YA novels, 60s books about Bigfoot, an Excel programmer's guide, and a pile of Dover coloring and sticker books, and they must say:

This is useless to me.

What crappy data. How amateurish.

I can't be bothered with this mess.

These people don't know what they're doing.

So I think the biggest advantage to Tim's proposal is that we can show people that we know what we're doing.. Yes, posting to each author page will make disambiguation easier once it's operational, but for now, I think we need this... why are users fretting over how to combine Bibles, or editions of Norton anthologies? Because they want it done right.

Authors are not done right now, and I wish casual users, or bloggers, or reporters, or the merely curious could see that we understand that.

Whew.

Yes please!!! A text field would help, at least for those who bother to read it. But don't restrict it to those authors who have identical names.

I have spent an incredible amount of time trying to separate Charles Morris (1833-1922) who wrote about The San Francisco Calamity by Earthquake and Fire, Teddy Roosevelt, and various Historical Tales, from Charles R Morris who is alive and well, writing about American Catholics and the AARP.

Yesterday I had all of the American Catholics sent over to Charles R, but today several new entries tipped the balance, and they have come back to Charles! It would be SO helpful to have something saying "Charles Morris (1833-1922) wrote --- and ----. Charles R Morris is a living author writing about ---- and ---."

They wouldn't see it on data entry, but they might spot it if they went to the author page, wondering why they see only a few copies of a work they know is popular.

Could the text warning be in color, or is that too complicated?

>16 oregonobsessionz: Yes, I agree with oregonobsessionz. There are also authors who have worked together and who get combined by someone who thinks they may be the same person.

I spent a long time separating Iris Johansen from Kay Hooper. They both wrote (separate) books for a series and someone had combined them, probably thinking that they were the same person. Making things even more ambiguous, Iris Johansens' web page lists all her works plus the works of the others in the series. The other authors are listed on her website but without explanation, making it look like "Kay Hooper" could be a pen name.

Which reminds me, the description should also list pen names.

> 17

I spent a long time separating Iris Johansen from Kay Hooper.

I feel your pain. I spent an inordinate amount of time yesterday cleaning up the mess that was caused by Roger Chartier, Philippe Aries, and George Duby having all edited volumes in the A history of private life series. Arthur Goldhammer and Michelle Perrot got thrown in for good measure. The horror, the horror!