Feature:90080 Pseudonym Bibliographies

Feature:90080 Pseudonym Bibliographies - OPEN When one requests Long Works for a pseudonym, e.g., Darrel T. Langart, nothing comes up. However, when you select Alphabetical or Chronological lists, the Long Works published under this pseudonym are correctly displayed. Ahasuerus 13:23, 5 Jun 2006 (CDT)

I’d like to re-report this. Someone just submitted a request to make Janet J. Asimov a pseudonym of Janet Asimov. On the face of it – that seems like a reasonable request. I had been avoiding ISFDB’s pseudonym support though (and that’s why I left the item in the queue) because now Janet J. Asimov bibliographic pages (other than alphabetical) page no longer lists any works and instead you need to look on Janet Asimov page. Marc Kupper 14:44, 26 Dec 2006 (CST)

This is actually by design, so we should probably move the discussion to "requested features". However, I should note that the way I addressed this request was by making Notes for a Memoir into a pseudonymous title first since pseudonym associations are done at the Title level, not Author level. Once I approved my change, the title appeared on "Janet Asimov"'s page with the appropriate "as by Janet J. Asimov" note, but it was still appearing on the Janet J. Asimov page as a "Stray Publication". I then approved the original request to make "Janet J. Asimov" into a pseudonym, which caused two things to happen. First, it made the "Stray Publication" line disappear and second, it added a note about the name being a pseudonym for Janet Asimov at the top. Ahasuerus 14:57, 26 Dec 2006 (CST)

Semi-related is that Janet Asimov’s pages that Notes for a Memoir is not listed as a vt on the alphabetic and chronological pages. Marc Kupper 14:44, 26 Dec 2006 (CST)

The Alphabetical and Chronological pages have never done a good job of handling pseudonyms. It's been an outstanding issue for months and I assume that they will be revamped as part of Al's recent push to redo the display logic. Ahasuerus 14:57, 26 Dec 2006 (CST)

We have to be careful what we wish for here. If we look at a "real" author's page (say Heinlein), we want it from his viewpoint, and we want to see his works listed there, and the works done as a pseudonym (say Anson McDonald) listed as variant titles to the works of a real author. Now if we look at a pseudonym that has become more popular than the "real" author (say, Robin Hobb vs. Megan Lindholm), I can see the desire for a view of just those works done by Robin Hobb. This is the complete reverse of the Heinlein case. In this case, there is a desire to treat the author as a special case, and to only display variant titles, without listing any "real" titles whatsover. That is a completely different type of bibliography, with different types of SQL queries, as it is the bibliography of a pseudonym.

Now let's say we do that. We have one set of rules for when we're a real author, and one set of rules when we're a pseduonym. What then of authors who have served as both (for instance when a book is ghost written on behalf of the author). Heinlein is a perfect example. We think of him as a real author, but in the case of Tomorrow, the Stars the actual editing work was done by Pohl and Merril (see Grumbles from the Grave). In this case, Heinlein is a pseudonym of Pohl and Merril. Now both the real and the pseduonym bibliographic rules have to apply to the same bibliography.

That was the situation yesterday, when we fixed the bug about variant titles showing up twice in the bibliography. In that case, a "real" bibliography was picking up both actual and variant titles and treating them like real titles. The fix is to only load the actual titles for the bibliography, and handpick the variants based on that list. That is, the bibliography algorithm actively suppresses the idea of a variant title being a canonical title. But ALL titles written by a pseudonym are variant titles, so the bibliography algorithm is actively suppressing them. I don't see how we logically can have it both ways.

The real question here is whether or not there is a requirement to show the bibliographies of pseudonyms. At present, the bibliography points to the real author, and displays stray titles which weren't made into variants. Is it a requirement to show more than that? After all, if I look up Christopher Anvil in Tuck, he doesn't present Anvil's bibliography; it says: "(pseud) See CROSBY, H.C." Looking up Anson McDonald in Clute/Nicholls show: "-> Robert A. Heinlein" Alvonruff 15:31, 26 Dec 2006 (CST)

I think that it's perfectly fine to have a single pointer back to the real biblio page for "single person pseudonyms" the way Tuck and Clute/Nicholls do. However, there is some value to being able to take a look at "Will Garth"'s or "Alexander Blade"'s bibliography to see what was published as by "him" and who the culprits were. This is what Day does and I find it useful when looking up an obscure pseudonym. However, I don't see it as a high priority issue, more of a "nice to have item". I would think that other pseudonym-related issues, e.g. the ability to quickly swap a canonical name and one of its pseudonyms (e.g. "C. M. Kornbluth" with "Cyril M. Kornbluth" or "Robin Hobb" and "Megan Lindholm") would be more useful and higher on the list of priorities. More than once I would start entering data for an author only to realize half way through that the majority of his work was published under a different version of his name, so I should really change the canonical name. Which was rather painful to do, but I survived.

I think the recent addition of the "pseudonym editor" feature has closed the most obvious and immediate gaps in our pseudonym support. An editor who knows what he is doing can do pretty much anything he wants pseudonyms-wise now. Now the big challenge (as I see it) is either to document the process of pseudonym association clearly enough to make it usable or tweak the process to make it less confusing. Ahasuerus 16:52, 26 Dec 2006 (CST)