I haven't seen the demo of IM v4 and how survey data is input into it. So this post may be premature.
Before I start, I want to say that I already know that dealing with cropped photos increases error since it throws off principle point. Adding constraints helps reduce errors in these type of projects. But it is a common scenario in photogrammetry to work from old prints. It just comes down to dealing with what you have available to you as best you can. I just wanted to let you know, that I know before you mention it 
If "Partial" survey data can be inputed into v4, please disregard the following:
Now, by "Partial" survey coordinate data, allow me to provide the following scenario to better explain what I mean:
Let's say someone has slightly cropped photos of an object (like prints which is, as I mentioned, a common scenario). Maybe they have no means of obtaining survey data but they do have a partial blueprint of that object. From this incomplete blueprint data which may reveal an X and Y coordinate for a particular point. But it does not reveal a Z coordinate for that same point, then it would seem to me that at least having X and Y constraints on that coordinate would be better than none at all.
I can't imagine the math involved there to reduce the error of Z. But if I'm correct, and correct me if I'm wrong, a planar constraint sort of does the same thing, so I don't think it's too impossible (just hard) to implement.
Anyhow, I hope you consider this idea in the future. That is If the forth coming IM v4 does not have this suggested fuction.
Thank you!
