@quid ok, I should have been more precise. Yes, the text is not entirely lost. But if tag-wiki for some tag is created, then the tag is removed (by removing all ocurrences) and if tag with the same name is later created again, the new incarnation does have empty tag-wiki.

Even though the text of previous wiki exists somewhere, it it snot automatically linked to recreated tag.

For example, when tag-excerpt for algebra contained information that the tag is deprecated. After the tag was emptied and later somebody created (algebra) tag again, the new (algebra) tag has empty tag-wiki as described here: meta.math.stackexchange.com/questions/473/…

This is precisely the reason why in tag-related discussions on meta when talking about tag-info I often link to the revision history. If you look at the question "What to do with the (subspaces) tag?" only the links to revision history in the last pargraph still show what used to be there:

""In the meantime, the tag-info has been added. The current wording seems to suggest that the tag is for vector subspaces. (I will also add link to the revision history of the tag-excerpt and the same for the tag-wiki.)"

All of this is not that relevant to current discussion. But perhaps it makes clear what my misconception was - I thought that tag-warnings are linked to tags in similar way as tag-info is. I.e., no longer linked to the tag if the tag with the same name is recreated. (This was the reason why I mentioned the example with tag-wikis.)

@MartinSleziak yes, my remark was tongue in cheek. I understood your initially confusion; indeed I shared it to some extent and would not have entertained the idea of creating a tag warning for an non-existing tag.

merging I have two questions which are basically identical. The only difference is that in one case variable is denoted by $i$, in other case it is denoted by $k$. So after changed both question and answers in one of the instances, the questions would be identical and merging would be theoretically possible.

I think that it would be also useful - both of the two questions have useful answer and at least one answer in each case is not provided in the other one.

In such situation, I should first edit one of the question and then flag for merging? Or should I first flag and say in the text of the flag that after they are merged I volunteer to make necessary edits?

The reason I am asking is that both possibilities have disadvantages.

If I edit first and moderators disagree with my suggestion to merge, I wasted some time. And the posters might be surprised that the question/answers have been edited, some of them might even rollback the edit.

If I flag first, then the merged post will have inconsistent notation between question and some of the answers for the period between merging and editing.

Or would it be better to ask the above on meta (so that it is useful for other users who face the same dilemma)?

However, I suspect that something like this is relatively rare. Or am I wrong and there are many cases when the editing+merging combo happens?

@arjafi, or anyone else that might be helpful. I'm kicking myself because I voted to close an algebra question (now -5 votes) as failing in the context arena. Within minutes, I found a well received question that is almost identically worded, asking for the same proof. So I came back to the recent post I voted to close, linked to the found post, describing as a "Real Duplicate."

But I can no longer vote to close as a duplicate question, since I voted to close for lack of context. Is there any way that the reason for closing can be edited to a more appropriate reason for closure? It seems that upon retracting an earlier vote to close, one loses any power to vote to close on that particular question.

Just frustrating, because I held off on "voting to close..off topic, no context)" and found the duplicate question, I would have voted to close as a dupe...and upon my vote to close (given my dupe-hammer), all would have already been cleaned up.

@amWhy Are you talking about question which is already closed, or about a question where some close votes are still missing?

@MartinSleziak Question where some close votes are still needed to close.

This is marked status-completed on meta.SE: Allow another close vote after retraction. Hm, strange...?

@amWhy I have cast duplicate close vote there (with the duplicate target you suggested). So the duplicate should be offered to the users when voting to close.

Thanks. Martin! And also, thanks for the link to the post, above...still trying to "get it"

Oh, I see. This comment explains what is meant by status-completed.

Wow... @Martin You are an incredible resource on matters broad and specific. Thanks.

All I did was that I searched on meta.SE.

(And then read the comments.)

@MartinSleziak I tend to stick to MSE, meta.MSE.... I should start venturing out. Is meta.SE about the entire SE network, or is it where site-wide policies are discussed, challenged, etc?

networkwide

@MartinSleziak Well..."there you go!" (American expression meaning, in this case: "Well, there you go, self, no reason not to venture there!" (meta.SE)

I think I'm going to need to write some "protocol" for myself. Since this has happened before, finding an exact duplicate of a question I've already voted to close for a different reason, I think my first line of (if the question is clear in the first place) will be to do a quick search for an dupe, and close as a dupe, before choosing another well-chosen vote to close, if applicable.

Hi @quid!

hello @amWhy

Interesting thing I learned above. I did not know one could recast at all after retracting a vote.

@quid I did not know that either. But waiting for 14 days to do so, means really being invested in having the question removed!