Vitalie Spinu

> the problem with the missing request ids, which is just as painful as the one with the output. Had no idea that we had those. What is the relevant...

> This is caused by the *msg* not being bound in the background thread. I see. This makes sense and is similar to the `*out*` binding problem indeed. If my...

> When running an non-print related eval the results are also "lost": Well, `eval` is the printing middleware in the sense that the `value` returned is the [printed value](https://github.com/clojure/tools.nrepl/blob/master/src/main/clojure/clojure/tools/nrepl/middleware/interruptible_eval.clj#L105-L115) >...

> I don't see why (has-child (tag :tr) (tag :input)) doesn't do what you want in that last example. My last example had one `has-child` missing. I guess my main...

```diff @@ -136,7 +136,11 @@ modify.default

Isn't modify.default called only on recursives anyhow? So special treatment of lists doesn't make much sense here. BTW, modify doesn't work on environments, might be good to add on this...

Ok, this is a bit more complex than I though. The only way to assign NULL to a list-like is to use `[ ..]

> Environments are collections but not vectors, so I don't think we should make it work. Dare to elaborate? Why would type play a role for a generic modifier? If...

I don't care much about numeric indexing, but may I ask why allowing numeric indexing in nested assignment is a problem? I do see a few use cases when extending...

@lionel- Would it be possible to speed up the resolution on this one? Longer it takes, more difficult it would be to modify the behavior. I believe this is a...