Sarah writes:
>> Merging is imperfect because when you do an author merge, OL does
>> not change the author field in all the works to the same author
>> record, it just groups author records together. This has caused me
>> problems. It is better to empty out all the author records first by
>> switching all the works to the correct author, and then do the merge.

Karen writes:
> I'm trying to picture this. I can imagine that different works for the
> same author point to different author entries... but it seems that
> when those author entries are merged, that merge should affect all
> works. Nope, I can't get a mental image of this one. I hope someone
> who knows the answer posts!

My experience has been that, when merging authors, OL does indeed
change the author field in all of the merged works, *most of the
time*.  It seems to be buggy occaisionally.  Karen is right that is
*should* affect all works (and my experience is that usually happens).
 From what I've seen, Sarah is right, to the extent that it does not
*always* change the author field in all the works.

I will agree that Sarah's method of preemptively fixing the authors on
works-to-be-merged, and then doing the author merge, does work nicely
to avoid any bugs in the merge process.  I don't usually do it myself
because I'm too lazy :-)

I don't know, but I wonder whether the problem lies in the author
merge itself, or in WorkBot coming along afterward.  Not sure.

One other thing I found is that there are apparently legacy author
fields in some edition records, which cause problems after author
merges.  They can be seen in the JSON data, but don't seem to be
removable through the normal web interface.  I posted this to ol-tech
a few days ago, but nobody responded there.  Should I repost that
here?

- Alan
_______________________________________________
Ol-discuss mailing list
[email protected]
http://mail.archive.org/cgi-bin/mailman/listinfo/ol-discuss
To unsubscribe from this mailing list, send email to 
[email protected]

Reply via email to