User:ErsatzCulture/DRAFTCanonicalAuthorNamePolicyChange

THESE ARE DRAFT NOTES; ONLY ONCE COMPLETE WILL THEY BE POSTED TO A GENERAL PAGE FOR PROPER CONSIDERATION

But if anyone else sees this and wants to comment whilst it's still a work-in-progress, feel free.

Background

This proposal aims to prevent a repeat of the December 2023, when an author posted on this Wiki to request that their entry/entries in the database be swapped around so that the name they now publish under is the canonical entry, and the name used for their older work to become an alternate name. Whilst this work was ultimately undertaken per the author's request, the process of getting to that point garnered a lot of negative coverage of ISFDB on social media, which has persisted in other places such as the Wikipedia entry.

Current policy (as I understand it)

If an author has published under multiple names, then the most frequently used name is the canonical record, with the other(s) being treated as alternate names.

As-and-when the ratio of titles/publications (?) changes such that an alternate name becomes the most frequently used one, then the relationship between the author records is flipped. The varianting of that author's titles is similarly flipped around. For an author with a large body of work, this can require a large number of edits, although (in my opinion at least) this should not be (and has not been) used as a reason not to undertake the change.

NB: to the best of my knowledge, there are no tools or reports that indicate when such an author record switcharound should be done. (I suspect Brian Aldiss vs Brian W. Aldiss, or Paul McAuley vs Paul J. McAuley may be cases where a switcharound should be done per policy, but has not as yet.)

Proposed policy

The flipping around should be done as-and-when the author indicates that a published name is the one that they identify as. (The "published name" is a qualifier to cover scenarios such as someone taking on a new surname after getting married, but continuing to publish under their unmarried name - in such a case, no change would be made.)

An "indication" may be a direct request from the author, but could also be based on public statements in interviews, Reddit AMAs, etc.

Problems with the current policy

1. Primarily that it can create a large amount of ill-will with the wider public, for no benefit.
2. As a consequence, potential editors may feel reluctant to contribute to ISFDB, which could easily have longer term negative effects as the current editor base cycles out, with insufficient replacements.

Problems with the proposed policy

1. Potentially creates extra load on editors. (Counterargument is that if the author continues to publish work, then as-and-when the "ratio" switches, there will be a larger number of titles and alternate titles to flip around.)
2. Mechanisms for an author to request their canonical author entry to be changed do not exist, beyond posting on this wiki, which is open to all the public. An author may for perfectly understandable reasons not want to discuss this on a public forum.
3. It is unclear whether requests from a third party to make a change should be honoured - how would ISFDB editors/moderators be able to evaluate whether a requester is a valid proxy for the author? (In fact, how would we know that someone is not impersonating an author for malicious reasons?)

Scenarios to consider

Author continues to use multiple names

...TODO...