Now that [rock ID questions have been banned](https://earthscience.meta.stackexchange.com/questions/1758/ive-had-enough-lets-ban-rock-id-questions), I'd like to offer some suggestions about how to achieve the desired result, which is to stop ES:SE from being a honey trap for uninteresting rock ID questions. 

To accomplish this, it's necessary to remove as many unwanted questions as possible, so that search engines no longer bring them up in their results list. 

Part of the ban was to delete all questions tagged [tag:rock-identification] and burninate the tag.

Not every poor rock ID question was tagged [tag:rock-identification].  There are many tagged [tag:identification-request] and also just [tag:rocks]. So, something will have to be done with those tags (they probably attract more casual rock identifiers than the other tag, anyway).

It was also asked [how to retain interesting questions about rocks](https://earthscience.meta.stackexchange.com/questions/1768/rock-identification-questions-are-now-off-topic-old-questions-will-be-deleted).  This way, we don't accidentally ban whole swaths of geology. (Erm...[tag:geology] might also be a honeytrap, but you can rationalize it as "not specific enough".)

I don't mean saving a lot of rock ID questions, even the ones where OP accepted an answer, even if they followed the guide. 

Instead, users who think a question brings up some interesting idea about geology can take action to preserve them: Retag such questions with tags with a more specific technical meaning:

 - [tag:petrology] - the study of rock composition. Come on, I want to see those polarized micrographs!
 - [tag:stratigraphy] - sedimentary rock layering
 - [tag:paleontology] - some fossil questions could fit in with any other questions on the field.