[EGN] Re: Please any one suggest me some project ideas
From: Randy Howard (randyhoward_at_FOOverizonBAR.net)
Date: 02/07/04
- Next message: Willem: "Re: identical executables, different results, continued"
- Previous message: Michael J. Fromberger: "Re: generating random numbers in c"
- In reply to: Edward G. Nilges: "Re: Please any one suggest me some project ideas"
- Next in thread: Nils Petter Vaskinn: "Re: Please any one suggest me some project ideas"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Sat, 07 Feb 2004 01:01:40 GMT
In article <f5dda427.0402061309.2faa92ca@posting.google.com>, spinoza1111
@yahoo.com says...
> > Georgia Tech did something like this to detect cheaters in
> > programming assignments. It's called cheatfinder, and works by
> > doing a more intelligent compare of two programs' structure. It
> > flags programs that match more than a pre-determined level.
>
> I find the idea of such a program reprehensible, since it might be
> used in a false accusation, and if the instructor does his job
> (emphasizing good style), his students' programming style might well
> converge leading to false positives.
And it would be quite easy for an instructor to interview the students
identified by this during office hours and immediately establish
which, if any, of them knew the material and which did not. It's a
filter, not a jury. Also, "hello world" assignments from a first
semester course probably all do look alike, but the more advanced
assignments later in a degree program are unlikely to pass well
designed tests for such commonality. Even if they do, the interview
process would be sufficient to demonstrate the requisite knowledge
of the assignment. It would certainly be more than sufficient to
demonstrate for example that Mr. Nilges cannot read and understand
C code, despite claiming such knowledge on his resume.
> In fact, the whole thing sounds like something from Georgia
He has no progressed to ad hominem attacks against entire states.
Illinois allows him to live there, without exile, which is far
worse of a statement about Illinois than anything one might
dream up to say about Georgia.
> the Bush-inspired assumption is that (1) students will have no joy in
> learning and (2) the whole purpose of education is constant testing
> with the constant expectation of cheating.
Given the abilities of Mr. Nilges demonstrated previously in this
newsgroup, it is completely understandable that he should have a
strong negative reaction towards testing for competence.
> Many programming shops in fact demand that the programmers have a
> uniform style, such that you cannot tell who's the author.
Irrelevant. They don't have to worry about cheating, because such
business situations do not usually have more than one programmer
perform the same work.
> I agree that it's often faster just to use windiff. The problem is in
> code (for example, in 250 stored procedures I recently had to convert)
> where the change is so systematic that windiff just gives up, and
> shows a mass insert and delete.
A pragmatic person might suspect that Mr. Nilges simply is incapable
of using the tool properly, like so many languages he has trouble
with.
-- Randy Howard 2reply remove FOOBAR "I admit that it is probably too much work ... to wade through my prose to get to anything useful." -- Edward G. Nilges
- Next message: Willem: "Re: identical executables, different results, continued"
- Previous message: Michael J. Fromberger: "Re: generating random numbers in c"
- In reply to: Edward G. Nilges: "Re: Please any one suggest me some project ideas"
- Next in thread: Nils Petter Vaskinn: "Re: Please any one suggest me some project ideas"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|
