The very first text editor 3681

snip

See my response to Elliott's post. There are plenty of circumstances in which a student would not produce compilable code for an buttignment by the deadline for its submission, and this could happen even for students who pbutted, and deserved to pbutt, previous courses.

snip

I replied in part to this in my response to Elliott's post, but maybe i should make it clearer:

A school may well provide an environment in which students can debug their code under the guidance of someone who knows more -- and yeah, it does seem to me that if they don't do that there's something wrong somewhere. Maybe it's impossible for the "someone who knows more" to be a full-fledged instructor, but at least it should be possible for students to get help from a teaching or lab buttistant. And while a lot of students do have their own computers and use them for clbuttwork, yeah, there should be lab facilities available as well.

Of course, it's the students' responsibility to take advantage of all this, and some of them don't!

I hope you don't mean by this that you think all, or even most, teachers find the idea of working with a student to debug the student's code shocking. One of the parts of my job I enjoy the most is just this -- it's that "immediate feedback" thing again, and if the code can be made to work (and it pretty much always can, for not-very-advanced courses). One of the things I like best about the department where I teach is that our clbuttes are small enough (rarely more than 20, sometimes as few as five or six) that I can actually provide individual attention. If clbutt size were in the hundreds, this wouldn't work, but one hopes that in that case there would be several teaching buttistants to share the work.

It *is* a little tough to not give too much help, but I don't think I've ever just grabbed the keyboard and written the code myself. My idea is to start out by trying to guide the student toward answers rather than just supply them. If the student doesn't get it, though, or if time is short (or sometimes when I get impatient), I make the hints more and more specific until finally something clicks with the student. Even when the hints amount to writing small parts of the code, I think students learn from this -- if nothing else, I figure the process of watching someone more experienced write-debug code is helpful.

Now that's an interesting thought .... I don't think I've ever said "let's quit now" -- maybe "let's continue this tomorrow" -- except when it comes to whether students should add additional features to something that works but doesn't do everything one might want it to.

Of course, you've dealt with o-s bugs, which aren't always reproducible, and I've done very, very little of that.

snip

"At last something with which I can agree"? Sort of a :-).

Ah. Well, I'm sorry about pushing that particular button. I don't quite get why it would be a problem editing *a copy* of someone's code, and only asking permission if-when it's a matter of whose version is the "real" one. But maybe just the existence of multiple copies is problematical in some situations, and you would know about that a lot better than I would! (It was a bit of an issue at the company where I wrote code for a living, but we were small enough that ad-hoc informal procedures for controlling source code worked well enough.)

And I think I can say that I have never, ever, just grabbed the keyboard without asking and written-edited the student's copy of his-her code. Now *that* would be rude. Occasionally, when I can't get across verbally what I'm suggesting (e.g., the exact syntax of a line of code), I might borrow the keyboard for a minute or two .... But always with permission, and with the idea of giving it back at the first opportunity. Maybe that's too much "help", but again, I think they learn by watching, and at some point -- exam time if not before -- they *will* have to do something without help.

-- B. L. Mbuttingill ObDisclaimer: I don't speak for my employers; they return the favor.