Sorry if I appeared to complicate things. That wasn't my intent. I replied in a different thread context. I don't know of a way to have a both "seamless" (meaning everybody uses their current favorite editor, eg, NotePad, vi or whatever in exactly the same that they use it now) and high performance (meaning sub-second update) editing of a file visible from both a Unix box and a Windows box.

However, a check-it-out, modify-it, check-it-in model would for sure work. I guess the issue is how to make something like that work as easily as possible... The "get the data into memory and open that data in an editor" is not hard - lots of these things can "read the other guys format". The hard part is "SAVE"..maybe I want to save the data in HIS format, not my native format. That means resolving conflicts with others who might be trying to do the same thing (ie updating the same file) which is the most difficult part and oh, by the way, some format changes along the way.

I guess I just stuck my head into a mess. But these problems have solutions.