Take Pride in your Work (AKA: The state of some CS TAs at SFU)
by Phil on Mar.11, 2011, under SFU
This started out as a Facebook note, but it has turned into quite the saga. All emails are posted in their entirety and verbatim, except for obfuscating details that could help identify the guilty parties.
Some interesting parts that aren’t specifically mentioned but which will help clarify: This was a group project, but I did it by myself. The project included coding which had to be submitted both electronically and on hardcopy, and a bunch of other material that was only hardcopy.
[Tuesday, March 8th, 2011]
Well, it took 4 years, but I finally found one of the most useless TAs in existence. I got back a 22 page report I submitted (with code) with only two things on it. A checkmark on the penultimate page, and an “88” (grade).
When I asked the TA where I lost marks, he claimed he “emailed us the standard”. Here’s exactly what was emailed to us, verbatim:
Hi all,
The Project 1 has been marked already, according to the following standards:
(1)Basically, the source code should be contained in the submit files.
(2)Furthermore, I prepare some other sample input for Q2 and Q3. If they work, it will be great.
(3)Neat writing, reasonable programming notes, and detailed illustration for wireshark also affect the grade.
BTW: Remember to include your ID in your report and make sure they are correct next time.
Best,
<useless TA’s name>
Note: This was emailed today. I definitely did 1, 2 is odd, since, if he wanted us to process more sample input, how about giving it to us, and 3? Pretty sure those are all done.
This TA seemed totally confused when I assured him in no unprecise terms that “grading” entailed telling me exactly where the fuck I lost points and why. “You want details?” Yeah, I fucking do.
I emailed him requesting such details. Not sure how he’s going to find them, since I have the only copy of the report in my hands (he has the submitted code electronically, though). Judging from his arduous note-taking on the report, I’m sure he keeps detailed logs.
And yes, this is the same course as with the lazy teacher who re-uses all the assignment questions.
Stay tuned…
UPDATES:
[Wednesday March 9th]
Another friend who is in the same situation (got 90%) also emailed the TA and got this back:
Hi, 90 is already a good work. Maybe I could not say where is incorrect for your work. Some groups have higher grade because they do details better. Resend me your code again, or show me your group numbers. I will send you a comment back about it. Best,
<useless TA>
[Wednesday, March 10th]
I received (from the TA):
OK, I will send back a comment about it these days.
I assume that means “one of these days?”
[Thursday, March 10th]
I got tired of waiting, so went to go see the prof and explained the situation. We’ll see what happens.
Later that evening, the TA sent this to the whole class:
In project 1, I just write down the comment if there is a big problem. So, if you need a comment about it, pls email me back with your report and code.
My reply (to the TA only):
Well, I consider any loss of marks to be a big problem if there’s no way for me to tell where I lost them. How am I supposed to then learn what I needed to do better?
As far as code, you have an electronic submission already, so there should be no need to send it again. If you ran my code with sample inputs, as you claim you did in your previous email, then you clearly have a copy somewhere.
As far as the hardcopy of my report, it is in Dr.<Prof’s Name>’s possession, as I didn’t think it was reasonable to have to jump through several email hoops again to find out what my actual grading is after you have had the physical reports for over 2 weeks and the electronic version since well before that (my code was submitted February 13th). Please consult with Dr<Prof’s Name> if need be. I don’t understand why you would need to check it again, as you should have a clear grading metric and a copy of where I lost points somewhere on file (if you followed said metric).
Cheers,
Phil
[Friday, March 11th]
Displaying that he really does not “get it”, Useless TA emailed me:
Well, I have already email you back that I will send you a comment back. This email is to the students who are in group. So, how can I find their code or report without the whole group name and ID?
I believe what we have here is a clear failure to communicate. Here’s my latest reply:
I really have a hard time understanding the problem. Either you have grading records, or you do not. If you do, then you know who was in what group and where they lost points.
If you do not, then how were you planning on telling me where I lost points when you no longer had a copy of my report (unless you photocopied the whole 22 page thing and kept a copy) and the code was only a part of the assignment?
More details as events unfold. If I was in charge of CS at SFU, I’d be embarrassed.
ANOTHER UPDATE:
I could be wrong, but I believe this TA has provided me with proof that he a) is full of shit, and b) never even ran my fucking code.
Hi Phil,
And this is the reasons for marking.
1. If either Q2 or Q3 does not work, it would minus 20 for each.(as standard 1)
2. Add 3 for your reasonable programming notes and 5 for your individual work. (as standard 3)
If you have any question about it, you could have appointment with me.
Thanks,
<useless TA>
Now, I could be wrong, but I don’t see a way that I could get -12 from that. It gets better. In another email:
Hi Phil,
This is the comment about your project 1.
best,
<useless TA>
Attached to that email was a Word document with this:
Q2 and the report for wireshark is OK, but I think the problem of your project 1 is Q3. When I implement your source code, there are warnings and errors as following:
Warning 1 warning C4553: ‘==’ : operator has no effect; did you intend ‘=’? \\freesia\userdata$\feic\desktop\project 1\comment\pba7_2\project1q3.cpp 56
Warning 2 warning C4018: ‘<‘ : signed/unsigned mismatch \\freesia\userdata$\feic\desktop\project 1\comment\pba7_2\project1q3.cpp 73
Error 3 error C2668: ‘pow’ : ambiguous call to overloaded function \\freesia\userdata$\feic\desktop\project 1\comment\pba7_2\project1q3.cpp 82
Warning 4 warning C4018: ‘<‘ : signed/unsigned mismatch \\freesia\userdata$\feic\desktop\project 1\comment\pba7_2\project1q3.cpp 109
Error 5 error C2668: ‘pow’ : ambiguous call to overloaded function \\freesia\userdata$\feic\desktop\project 1\comment\pba7_2\project1q3.cpp 168
A few interesting things some of you may notice.
First, those are mostly compiler warnings. They don’t show up at all with g++ unless you use -Wall.
The two ERRORS (and this is important) are due to him checking on a Windows machine, even though I told this this was written on OSX and also tested on linux.
Let that sink in for a while. Compiler errors. That means the code did not compile, and he could never have even run it.
I believe we call that being a liar fucktard.
CORRECTION: I’m a dumbass. He took 20 off for the “not working”, and added 5 and 3. So he was “correct” in his evaluation. Only one problem…
From an email by the prof when the project was sent to us:
4. There is no particular restriction in terms of the programming language, though
we prefer C/C++/Java (making TA’s life easier). If you use other special languages,
you may need to meet with TA to demo your programs later. For the OS, again, Windows
will make TA’s life easier; if you use Linux/Unix/MacOS, please clearly indicate.
From the notes before the source code in my report:
NOTE: The code was written on OSX and also tested on linux.
Damn, that reading. So hard to figure out.
FINAL UPDATE (I hope):
The TA finally got it running (digging through my submission, turns out I had even included a readme with compilation instructions, platform it was tested on, etc).
In his defense, there actually is a bug in my code, as it fails with his sample input. I’m actually fine with that. My objection was never really my grade (as getting a potential extra 12% for a project worth 7% of the final grade is more than a little irrelevant), but with the fact that it was impossible to get proper feedback in a timely manner.
March 31st, 2011 on 14:47
Being in the same class and dealing with the same TA, I can offer some anecdotes in my dealings with the TA.
For the same assignment, I got full marks despite my program not actually completely working in my own testing (and having misunderstood the assignment). And I’m relatively sure that the TA probably didn’t have a Linux machine with Python3 and the Twisted libraries installed (which I was using for TCP), but the TA might have.
I also emailed to complain about marks on a the midterm where I had answered questions correctly, though not necessarily what the answer key said, and got a reply along the lines of me needing to “Answer the question we asked, not the question we asked” (not a direct quote, I don’t have the original email readily available). I’m sure that’s not what the TA was trying to say, but…
I’ve also found the professor’s sometimes 10+ year old notes to be more than a little dated due to technological progression since they were written.
March 31st, 2011 on 18:25
Yeah, you know what’s coming up? The cloud!
Also, pretty sure the prof still owes me points for the midterm as well.
Phil