My preference would be to use tag wikis to list resources, and meta to explain how to use them and why.
There are many potential resource questions that could be asked: grammars, histories, linguistic studies, pronunciation guides, and more. Many of them closely correspond with tags that we already have, and the tag wiki for each one provides a place where resources can be listed. For example, see pronunciation's wiki:
Resources:
A standard reference for Classical Latin pronunciation is W. Sidney Allen's Vox Latina (available as a PDF)
Something similar could be done for all the tags on our site. For example, the wiki for genitive could list the relevant sections in introductory textbooks as well as grammars. The wiki for translation could list dictionaries, grammars, and perhaps online translation tools (if any are worth mentioning!)
To ensure that people are aware of the existence of these resources, I'd include a note to that effect in the tag excerpts.
Meta posts could complement these resource lists by explaining how they can and should be used in the context of this site, such as in How can I ask a translation or homework question?
The pros I see for this approach would be:
- Resource lists are available
- The lists can be edited by most users, but require approval by others
- Editing the lists happens behind the scenes, so the lists do not distract from the "real" Q&A
- We don't have a precedent issue, where we have to be constantly debating whether or not it's okay to have another resource question about XYZ aspect that isn't covered in an existing one.
- It seems arbitrary to me to say we will only accept some small number of resource request questions, when it's quite possible that some people would be interested in locating the best works on pronunciation, comparative grammar, historical linguistics, Vulgar Latin, Ecclesiastical Latin, translations of Classical texts, etc. If we do this, who's to say that we can't have a question asking for a list of the best English translations of Augustine's Confessions?
- We keep the focus on quality, definitive Q&A
Some cons might be:
- The lists are less visible to users than they would be if they were on the main site
- We can't express our opinions regarding which resources are best by voting
To me the first of these cons is a trade-off that I'm willing to make for all the reasons addressed in the "pro" section. But let me address the second one.
Say we have a question on introductory textbooks. A year from now, we have 4–5 answers that range in votes from –2 to +16. Then, a new introductory textbook comes out, and we add it to the list. It may be the best textbook ever, but it will take months or (more likely) years to get more votes than the other answers on the page.
My point is that as resources are added in future months and years, the voting mechanism will not be particularly helpful. My experience is that excellent new answers take a long time to overcome the votes earned by good old answers, making votes misleading for the purpose of recommending resources.
Summary
I'd prefer that resource lists reside in tag wikis and meta posts, where they can be freely edited without distracting from the Q&A. This avoids the problem of how we draw the line between "What are good introductory textbooks?" and "What are good translations of the Aeneid?" since tag wikis can handle both. It does not use the voting mechanism for recommendations, relying instead on a process of approving and rejecting edits, but I argue that the voting mechanism is not particularly well-suited for this situation anyway.