?

Log in

No account? Create an account

Previous Entry | Next Entry

Team Projects

prolog has been working on a team programming course that reminded me a little of a CIS 540/541 (Software Engineering) project here. It inspired me to issue the following invitation to all your programmers and CS/IT students:
Please tell me about your most interesting, pleasant, or stressful team programming experience.

Here is a rundown of team programming at Kansas State University's Department of Computing and Information Sciences:

I am the instructor for one course (CIS 690 - Implementation of High-Performance Data Mining Systems) where team projects (2-3 members per team) are allowed. I appoint a team leader (or the students elect one) and that person is responsible for overseeing load distribution and accounting for whether everyone has done his or her part. Now, most of the time, I've had grad students in the course, and sometimes they (both grad and undergrad) have "covered" for the slackers, but generally it's worked out all right.


Another team project course is our undergraduate software engineering capstone course, CIS 540. About 130 students take this course each fall (Computer Science and some tracks of Computer Engineering). Here I've heard stories to chill your blood, and others to warm the cockles of your heart. The standard team size is 5, and I understand that the team leader is appointed by the instructor rather than elected by the team members. The main thing that makes this course challenging is that luck of the draw frequently puts multiple hard workers, virtuoso coders, slackers, less competent (e.g., math or programming-challenged) students, or (on occasion) brilliant students on the same team.


Our department head has students in his parallel programming course report on the effort split between students (e.g., 10%/90%), and weights the grades accordingly. Theoretically, then, it is possible for two team members to turn in a successful project but for one to get an 'A' (4.0) and the other an 'F' (0.0). I've never heard of this happening, but apparently, different letter grades are common even when students' scores on other things in the course are similar.


To follow up: in addition to your stories (or in lieu thereof), please tell me a little about your instructor's policies, or your own if you are an instructor:


  • How many students are there in the course? On your team?

  • Does the instructor hold you accountable for individual performance and contribution? How or how not?

  • Has there been any disgruntlement? If so, how if at all, have you made this known to the slacking team members? To the ones who are picking up the slack (even if you are one of the slackers - post anonymously if you like)? To the instructor?

  • What response did (or do) you expect in each case?

  • How would you run the course and hold students accountable if you were the instructor?




Edit, 09:25 CDT Fri 07 May 2004: looks as if prolog is done with the project. Good job!



The End of Mellyn: Take a bow - well played, Friends. I'm going to miss this show despite myself.

Tangential observation: jellybeanzulu once said that on the web, you're only ever a few clicks away from 600-year old Swedish. Well, today I learned also that you are only ever a few clicks away from the Lakhota word for "pink". Hrm. Go fogure.

--
Banazîr

Comments

( 8 comments — Leave a comment )
masteralida
May. 7th, 2004 06:11 pm (UTC)
Do I even want to know why you were looking for/stumbled across the Lakhota word for pink?
banazir
May. 7th, 2004 06:16 pm (UTC)
Surrender the plink
Do I even want to know why you were looking for/stumbled across the Lakhota word for pink?
Stumbled across, to be sure.
Was looking for a Dances with Wolves quote, and went to a Lakhota language learning site when I couldn't find the original.
phawkwood and darana can prolly tell you which quote I am thinking of.

--
Banazir
zengeneral
May. 7th, 2004 08:40 pm (UTC)
yeah, rant time part 1 of 2
Well, cis540 project was fine. We did a robotic simulator and a Monte Carlo Localization “brain”. The neat algorithm was put the robot in a grid world given just a map and let it localize; it was good.

I was the team leader and assigned various tasks. I wrote an unfair portion of the server code primarily because it interested me, and I knew precisely what needed to be done when and where. I had two lackeys that did my bidding, and I just gave neat assignments/problems to the others.

So, fun fun, project for fall was finished; well, given that our project had two phases a fall and spring. dag gave us a choice to switch to a new project or continue the old one. Now in cis541, the new project was localization with a GPS; the continuation was move the brain to the real robot; the grand design of the brain would have worked greatly since no assumptions were made except that the robot has sonar, so the choice was something new or finish fast and goof off the rest of the semester.

Well, I wish we chose to finish fast and goof off…. the GPS unit we have sucks. A potential error of 15 meters; our sonar has precision of 3 meters. The unit does not have a compass, the update in the velocity is really slow and not correct. Our algorithms from the previous semester were no good, so we had to start over. Well, algorithm after algorithm didn’t pan out and so we are stuck with the sole fact that the GPS sucks (I thought the algorithms were good, but each one revealed how bad the GPS is). Our goal was navigation around campus, well… with an error of 15 meters; I am not taking the risk of it thinking the road is safe.

In theory, given the first part of this semester combined with last semesters = original spring project. Bad choice? perhaps, but more is learned in failure than artificial success… Now, I go back to coding the last component.
zengeneral
May. 7th, 2004 08:57 pm (UTC)
rant time part 2 of 2
Has there been any disgruntlement? If so, how if at all, have you made this known to the slacking team members? To the ones who are picking up the slack (even if you are one of the slackers - post anonymously if you like)? To the instructor?
I honestly think that no one really cares about the project anymore; if I was a manager, then the project would have been cancelled due to poor hardware support. Being able to navigate a 1 meter span with a 15 meter error is very difficult!


How would you run the course and hold students accountable if you were the instructor?
I would try to interview students or have a TA interview the students on their code, potentially point out issues (like correctness, scalability, reuse).

I would make several changes to the course:
1. smaller teams
2. more interesting and doable projects ( robots suck! )
3. less BS (the next time I have to make a gantt chart in minutes, I am going to snap.)
4. no time logs
5. more discussion about software architecture and planning for the future (like the real reasons for object oriented design as opposed to modular design).
6. no lab... or a much better lab
prolog
May. 8th, 2004 11:23 am (UTC)
Thanks. May I never have to do another large-group project again.
banazir
May. 9th, 2004 01:27 am (UTC)
Amen?
Thanks. May I never have to do another large-group project again.

I'd say ainsi soit-il, mais, I don't know if this is likely.

--
Banazir
prolog
May. 9th, 2004 10:52 am (UTC)
Well, here's to hoping. I have one more class to take for credit, and I think I know what it is. The prof's projects (I've had him several times before) has always been "one, two people maximum."
banazir
May. 11th, 2004 07:46 pm (UTC)
Good idea
I think I might do that this summer with CIS 690 (Data Mining Systems).

--
Banazir
( 8 comments — Leave a comment )

Latest Month

December 2008
S M T W T F S
 123456
78910111213
14151617181920
21222324252627
28293031   

KSU Genetic and Evolutionary Computation (GEC) Lab

Teunciness

Breakfast

Science, Technology, Engineering, Math (STEM) Communities

Fresh Pages

Tags

Powered by LiveJournal.com
Designed by Naoto Kishi