On Aug 12, 2011, at 3:40 PM, DJ Delorie wrote: > In this case, "the right thing" is to do what most people expect it to > do.
You have no idea what that is, and you can't have any idea without spending a lot of time actually watching a lot of people use the software. What people report is miserably unreliable as to what the *right thing* really is. I learned how to do optical telescope/instrument control software by watching astronomers work. But the system some students and I cobbled together in Forth back in the early 80's was pretty crude. So, we got a very bright MIT student to rewrite it in C as a thesis project. I told him "We'll pay you to take a trip to Kitt Peak. Go and assist an observer for a few nights, see how it really works. Then you'll know what to write." But he had that coder's arrogance, wouldn't do it. Instead, he went around and interviewed observers and coded up what they *said* was the right thing. The result was gorgeous: lovely display, well organized menus, fast, efficient as a demo. But it proved unusable: too often, the "right thing" in the code wasn't *quite* what was really needed in a specific situation at the telescope. The crude, ugly Forth system proved more resilient: it let the observer compose what was really needed on the spot, so the observers continued to use it (to my dismay, because I didn't want to keep supporting it). John Doty Noqsi Aerospace, Ltd. http://www.noqsi.com/ j...@noqsi.com _______________________________________________ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
