Yup, you're right: accepting 9-1-1 messages via SMS has its own -- fairly well known and easy to anticipate -- set of problems.
Do they outweigh the benefits? Maybe.
Will we find out without live testing? I suspect we won't.
Should the FCC have *mandated* it, in advance of such live testing?
Well, to quote Columbia: "Certainly not!"
On the "delayed response" front, yes, that's a problem, but if you tell people "if you don't get a response within 60 seconds, the message *probably* did not get through, resend exactly" (assuming the dispatch system sends an auto response), you'll fix all of that... except the case where a message is delivered *late*.
*Generally*, SMSs are delivered with the original send time attached, so a dispatcher would be able to see that one was stale. What do you do then? Policy call.
As for pranks, SMSs are delivered with the phone number of the sending phone attached, and so far as I know, there's no reliable way to spoof that: if one is sent via a proxy (which is technically possible), *you can see that it was sent via a proxy*, and downcheck it's reliability.
But all of these issues lead to what you're *really* trying to say, I think, which is this: for a PSAP, since it has to dispatch on *everything* it gets, having a new communications channel of questionable reliability is *not* necessarily a good think; a CBA has to be done on whether it will actually earn out what it will cost on the wide scale.
And you're perfectly correct in implying that we have no data on that.
