ISFDB talk:Authors that only exist due to reviews
OK, the refresh does include some authors already "fixed" by making the reviewee a variant of the canonical author. This still leaves a bit of a gap between knowing "who we really mean to refer to" and WHY we really mean them. Personally, I think if an author has no titles listed, there should be at least an explanation of why the name exists. BLongley 22:10, 11 March 2009 (UTC)
- OK, when the pub exists here already under correct name and author, I'm experimenting with creating the pseudonym, and creating the variant title, but not leaving a publication that doesn't exist. The variant title has a comment "This title exists to indicate this form of the author name or title exists in a Verified review." It won't actually show up under the pseudonymous author name normally, but if it does the note should be a reasonable explanation of why it exists. The reviews are linked to the canonical title/author as linking to the variants doesn't show up until ISFDB software changes. This seems to leave the actual pub as is, the review exactly as is, and yet provides the explanation for the mismatch. See Walter Hunt for instance, where the review got both title and author wrong. Comments welcome. BLongley 21:31, 12 March 2009 (UTC)
IMO: If it's due to a review that typoed the name - leave a note on the typoed author at least, or "fix" the review to the right author and leave a note about how it was actually spelled in the review. We're already having problems with a lack of variants of reviewers, but those are mostly traceable. But if you find an author with no apparent justification for existence, it's probably down to a review. Add the publication (and LINK it to the review, it's not automatic), add notes, fix reviews, whatever - but at the moment Ghost Authors are a problem for anyone using ISFDB online that can't use the underlying database, there's just no links anywhere. BLongley 22:10, 11 March 2009 (UTC)
"Fixing" by creating a variant of the reviewed author
I don't feel that this actually fixes the problem. It creates pseudonyms (author variants) for works that have never been published under the author's name. For instance, someone "fixed" the non-person author record "Zack Hughes", by creating variants only because two reviews misspelled his name. If you go to Zach Hughes summary page, you see that he has had work published as by "Zack Hughes". I know he hasn't, but any newcomer to the site would make that assumption because we list the name under "Used These Alternate Names". But if you look further down the page, you don't see [as by Zack Hughes] on title record. I personally feel we should fix the problem by a.) creating a title record for a pub that's reviewed but not in the database, or b.) correct the misspelling of the review and make a note of it in the review's title record's note field. The only way to accomplish this is for the software to be changed so that we can see the review record that created the "author that only exists due to reviews", and that ain't gonna happen soon. MHHutchins 04:41, 25 March 2009 (UTC)
- If people aren't too protective about "exactly as stated in the review" then fixing the review is the simplest option. I'd certainly be happy with fixing an unverified review of a verified pub. Not so much a verified review of an unverified pub. Can we take some of this to Rules and Standards discussion and see if there's some consensus on at least some of this? BLongley 21:31, 25 March 2009 (UTC)
- Yes, let's do that. MHHutchins 04:33, 26 March 2009 (UTC)
Feature Request
There's a new request that could make maintaining this sub-project unnecessary, but might also make people too lazy to find out why there are stray reviews. Discuss. BLongley 00:55, 14 October 2010 (UTC)
- I think I'm mostly in favour, but would also want a title equivalent of the "Bibliographic Warnings" message we have for publications that aren't quite right. In the meantime, shall I refresh this list? BLongley 00:55, 14 October 2010 (UTC)