[UMNiK]

Thanks for the detailed explanation of your workflow! Where you and I differ, it seems, is that I don't really want to have too many actual ebook files for books that I am not currently or imminently reading - what I'd like is to have a way to integrate my antilibrary (made up of empty books in Calibre that are now properly tagged as Goodreads.Want to Read and are on the default Goodreads to-read shelf, some with a native Goodreads note to the reason why) with my regular library (of books that do have files and some of which are read and marked as such both via calibre boolean column and with a Goodreads.Read tag and on goodreads' shelf). As mentioned before, the reasons for doing that are to be able to easily apply any hierarchical tags or reading lists and thus choose books based on multiple overlapping criteria (books of a certain genre with that award/nomination by these authors and so on, i.e. augmenting my personal notes with an external database). In other words, I use my lists to decide whether to get the book files rather than the other way around.

The in-built kindle-goodreads integration requires some adjustments for side-loaded books I do not care to do, nor do I need to track reading progress. All I want is to have a single shared entry for the same book inside Calibre and Goodreads that won't get messed up by any further metadata downloads (happily, the sync plugin and goodreads' exclusive shelves mean that any further operation on that book can be done in Calibre and easily synced to Goodreads). Can I just use the sync from (massive) Goodreads shelf to create the entry and disable identifier download in the Goodreads metadata plugin? Of course, it's great to have the option to manually change editions, but since they will all end up as a KFX file anyways, I'd just like to have a single, most stable id from the outset. My issue is I'm not sure whether adding the book to a shelf on goodreads will get that most stable id, and with no book file, the alternative is manually pasting in the title and author into an empty book and invoking the metadata download from Calibre, then synching the book to Goodreads's shelf, which is obviously a couple more pastes that I'd like (and as compared to the three clicks it takes to add the book to the shelf on Goodreads, then sync from shelf, then bulk metadata download).

UPDATE: After some testing and to be on the safe side, here's what I settled on:
1. Add the empty book to calibre by copy-pasting the title and author from goodreads (obviously, this will also let me see whether I've already shelved the book as ignored and any other notes).
2. Add that empty book to to-read using the Sync plugin and letting it find the best id match (which will also add the correct goodreads-related hierarchical tags).
3. Download book metadata, which should use that best id (at least for Goodreads and Goodreads More Tags (which has to be manually edited at the moment)).
4. Optionally follow the created link from Calibre sidebar to Goodreads (if closed the tab already) to write in any future notes.

This ensures that the best goodreads id is maintained for any further operations on the book. Ideally, there would be a userscript to make sure any Goodreads book opened in the browser opens to its "best edition", thus making it possible to add that to the shelf and sync from that to Calibre, but alas, no such thing exists AFAIK.