|
Notes from Informal Group Meetings |
|
 CSE Home |
Group Home |
About Us |
Search |
Contact Info |
|
______________________________________________________________________ future of the group [suggestions roughly in order, near term to long term] bring Martin, Hal into loop establish research seminar separate from ET reading papers in CS ed. and ed. tech involve undergrads in research ask Hank re specially equipped classroom in new building find context for experiments teach a course? "take over" a course? ask ed. people how they do this maybe help w/ Vibha's 403 course? human subjects different levels of review (minimal, full) exemptions from dept. chair? see also: Steve's ITR ______________________________________________________________________ SIGCSE 2002 Doctoral Consortium submissions [could link submissions here; e-mail yasuhara@cs if interested]
______________________________________________________________________
CSE 101, modelled after ENGR 100?
implementation ideas
experimental course
maybe FIG seminar series
CSE 101 vision
concepts as focus, programming as reinforcement
[discussion postponed in favor of next section]
what we envy about ENGR 100
independent, self-motivated students
task-oriented
discovery-based learning
problem-solving methodology, including design, testing, etc.
different labs focus on different parts of engineering process
physically manifested projects (resource intensive?)
collaborative
projects w/ goals beyond learning, e.g., friendly competition
Richard's thoughts (via Tammy)
won't fly in CSE
too resource-intensive
Weld tried similar course in recent past
consider requesting CSE section in ENGR 100
see also: notes from 303 proposal meeting, 06 Dec 2001
______________________________________________________________________ brainstorming for research/reading seminar 19 Dec 2001 prep for SIGCSE by reading papers to be presented "best of" recent 2-3 of proceedings ed. research methodology (Tammy says ask Bill Winn, Education.) evaluate methodology of "best of" work group journal subscriptions? ______________________________________________________________________ Tue 08 Jan 2002 Attendees were Doug Johnson, Tammy, Nick, Steve, and Justin. * Tammy talks about 590ET Tammy has set up group projects to serve as the focus for 590ET this quarter. I'm not going to note most of this info since it will come up again in 590ET notes. However, here are a couple of points.. Someone (?) pointed out that 417 might not be the best venue for the distributed office hours but rather 142/143 where Hal will be a vocal supporter and we can have some surety of heavy use. And now I can't remember the other point! D'oh! * All discuss EGR100 Mary Cook is looking for a project from us by the beginning of next quarter at the very latest. We're not sure if this will be in addition to the robotics project or perhaps in the sections that don't do the robotics project, but regardless, it should probably hit CS fundamentals. Nick suggested writing game controllers/AIs for some game and suggested a simple mathematical game (pegging nodes in a graph) as one possibility. Steve pointed out that this would work well since students could, for example, use conditionals without knowing loops, etc. by writing rules for a rule-based, reactive agent. One concern Steve had was that knowledge gained from this might not be easily portable for students. Tammy recounted several of the other projects that we had considered: dismantling a computer and the web scavenger hunt are unfortunately the only two I recall right now. D'oh! We agreed that it might make sense to dedicate at least one meeting (next week) and maybe half meetings each week to EGR100 until we have a project ready for Mary. It occurs to me only now that this could be assigned to someone in 590ET as a group project! Would anyone be interested? * All give "state of the quarter" addresses ** Justin Justin is doing quals this quarter by implementing the automated question customization system. He feels that some of this (e.g., numerical substitutions) is fairly straightforward and so is concentrating on larger variations in question such as automated generation of Boolean expressions from CFGs and similar work on code snippets. Something I thought of but forgot to point out in the meeting: a number of tutoring systems have been developed to observe student work/reading and come to conclusions about what topics students understand. Perhaps Justin could use similar logic to enforce that solution of a question requires hitting "Topic A", "Topic B", and "Topic C"? ** Steve Steve is continuing work on the contextualized feedback system this quarter. Ideally, he should be testing it in a classroom at some point in the quarter. He's also been excited by Henry Kautz's thoughts on having the system itself provide feedback in a classroom feedback system: e.g., the system might detect volume levels throughout the class and suggest to the instructor that she speak up when those levels fall below some threshold. ** Tammy Tammy will be testing out the CATs (?) and, if possible, the virtual mylar et al. in PMP this quarter. She is, of course, also managing 590ET again without any complaint. Thanks, Tammy! ** Nick Nick is working on quals courses, which are somewhat out of his area this quarter (architecture and software engineering). Also, he and Hannah are TAing for Henry Kautz in 326 which gives us a good in to another class. ** Doug Doug has no teaching obligations this quarter. He's focussing on "miscellaneous" things he needs to do and prepping for next quarter. He's most excited (wrt our work) about the stuff Tammy and to a lesser extent I am doing for distributed classes (since he taught PMP last quarter). * Miscellaneous ** Fares for SIGCSE have gone back up again. I'm calling a travel agent about these today. ** Doug would like to teach if we have an experimental class we wish to roll out
______________________________________________________________________
15 Jan 2002
ENGR 100 CSE project planning
......................................................................
Steve's pre-meeting notes
As we decided last week, this week will be dedicated to discussing a
project for EGR100. I'm going to take a quick stab at what a "good" CS
focussed project would be:
- Teaches fundamental concepts of computer science
-- algorithmic thinking
-- problem decomposition (functional, modular, object-based, or other)
-- basic programming constructs (control flow, expressions, variables, and
syntax in service of the topics above)
-- model of computation (basic model of a computer, hand-computation,
debugging)
- Group-centered
- Approximately 2 weeks in length, with 3 3 hour meetings per week (is
this right?)
- Exciting and engaging
- Decomposable into relatively small steps
- Low cost (though it can certainly rely on infrastructure)
- Sustainable by the EGR folks
- Teaches easily transferable skills (?)
OK, that's a first cut, anyway. I think we can all agree on most of the
latter points; it's the first and its subtopics that we might have to
fight over :). I also threw in the final point as a possibility.. it's
unclear how important that is.
......................................................................
meeting notes
recall: little/no programming experience; might have one
group w/ students that have some
project ideas
agent control
crypto might be better
+ context for discussing algorithms w/o using computers
- likely to result in ad hoc algorithms, random ciphers
compression/decompression
modularity highlighted in any of the above projects?
animation
image manipulation a la PixelMath (www.mymath.com)
AI game
sorting algorithms, tracing and time analysis of
web page design
web browser plug-ins (probably too technical)
Macromedia Flash? (www.macromedia.com/software/flash/)
simulations, e.g. particle
computer disassembly
Is project intended to replace or follow engine disassembly?
candidate beginning programming languages, environments
Alice (www.alice.org)
Stagecast (www.stagecast.com)
Lego Mindstorms (mindstorms.lego.com)
Toontalk (www.toontalk.com), ltd.-time trial free
Logo-like systems (http://el.www.media.mit.edu/groups/logo-foundation/)
for next meeting...
Alice & Python (Doug, Justin)
other programming environments/languages (Ken)
ENGR 100 project requirements, goals (Tammy)
find out more about engine disassembly project (Tammy)
papers on successful projects (Nick)
______________________________________________________________________ 22 Jan 2002 ENGR 100 CSE project planning meeting w/ Mary Cook project duration: 3-4 weeks to take place of LEGO robot project for a section example project sequence... bridge engine dissection LEGO robots existing projects virtual car: prototype, build, compete Mary likes puzzle/maze idea recall: assumption of limited math, science, programming background CSE project keys programming env't w/ gentle learning curve emphasis on algorithm design, not ad hoc solutions inter-team competition aspect to motivate students What are the deliverables? report? programs? adversarially designed mazes for other teams? algorithm descriptions? next steps outline project sequence, documents examine existing projects check programming environments
______________________________________________________________________
29 Jan 2002
[Ken: Thoughts I added while typing are in square braces.]
......................................................................
ENGR 100 CSE project planning
tasks
ensure StarLogo is good fit for project (Ken?)
prototyping maze, robots, data structures
begin drafting documents (Steve)
What is group size? 4?
motivate maze-solving w/ real-life examples of robot navigation
search & rescue
[Mars rover]
[deep sea probes]
[volcano crater exploration]
project's key concepts
general engineering; evaluating options based on priorities
algorithmic approach to problem-solving
good algorithms produce significant, real results
computers as brittle instruction-following machines
starting point for free development of navigation algorithms
basic ADTs: stack, queue, priority queue of locations
[need to check StarLogo on data structures, interfaces]
sensing
marking
teleport to previously visited location
......................................................................
theories of learning
wide variety of learning styles across different people; not sensible
to attempt to find a single paradigm universal across humankind
however, can understand learning better through examining wide range
of paradigms argued to date
rebuttal: evidence suggesting language acquisition exploits
universally pre-existing learning capabilities
maybe good to foster new learning styles in students, resulting in
more effective future learning, more effective communication with
audiences through understanding their learning style(s)
______________________________________________________________________
12 Feb 2002
ENGR 100 project drafting
to-do list
StarLogo
capabilities: file I/O, saving mazes, inter-turtle communication,
wall sensing, data structures, control flow
start code infrastructure, API
decide on subset of turtle commands
project outline
6-9 2 hour sessions (FYI a total of 12-18 hours)
small milestones, expecting 2-3 to be attained per class meeting
0. warm-up
concepts
computer as sequential instruction-follower
precision (required of syntax)
activities
familiarization w/ env't, turtle commands
manual navigation
1. procedures
concepts
procedures as named sequences of instructions (no params. yet)
activities
save command sequence in procedure
2. leading up to algorithms
concepts
distinction between cmd seq. tailored for particular maze vs.
more general problem-solving approach
repetition?
activities
discuss maze-solving strategies
discuss, implement simple, randomized algorithm (w/o calling it that)
______________________________________________________________________ 19 Feb 2002 ENGR 100 project drafting began fleshing out Doug's mini-milestones list; see project page ______________________________________________________________________ 26 Feb 2002 (just before SIGCSE) ENGR 100 project drafting more fleshing out of milestones; see project page
______________________________________________________________________
05 Mar 2002
decided that we can finish ENGR 100 project by mid-spring
target draft completion by first week, spring quarter
need to contact Mary, confirm delivery, update on plans
plans for next quarter's seminars
ET for practioners
ED for us, the researchers
J. of CS Ed. (individual subscription pending, courtesy of Richard)
more foundational readings in education, learning sciences
explore newly discovered publications
Justin will request Tue 10:30
Distinguished Lecturer possibilities
Mark Guzdial, GA Tech (DC discussant)
Greg Abowd (sp?), Classroom 2000
Margolis & Fisher in area in April?,
Raj Rao contacting on behalf of Diversity Committee?
______________________________________________________________________
12 Mar 2002
Steve's notes
We discussed the feedback study in Martin and Grace's class yesterday.
Nick suggests: preset certain feedback categories not to appear during
lecture (such as "cool topic")
Tammy is curious whether Grace would want to see individual student logs.
This isn't feasible (as we told the students that only aggregate data
would be presented to the instructor), but it might be worth asking about
how she would use individual feedback?
Positive points: the system worked, the students enjoyed using the system,
the instructor enjoyed the system (?), there was a visible effect on the
course of the class.
Tammy wants to know the effect on students *without* the laptops. Steve
recalls that there was a rise in muttering after about two of Grace's
response episodes, but he couldn't hear the specific sentiments voiced.
Justin suggests doing a pure notetaking study: no feedback enabled,
just notes (in notepad or whatever). Doug and Tammy suggest that most
students would rather not take notes on a laptop. [I believe Doug
actually just said that note-taking on laptops is common in PMP
courses, consistent w/ what I've observed in CSE 595 this
quarter. --Ken]
Doug is concerned with management of time for the lecture. What happens if
feedback pushes the instructor further and further from the planned
course? Some response from the group (and Richard in particular) that
perhaps the abstract topics provided by the feedback category are less
conducive to that phenomenon than targetted (and potentially off-topic)
questions.
Nick points out that there's a danger that people who feel they are the
only ones with a question will actually get "evidence" that they are the
only ones when they see that no one else has marked their point. Responses
seem to be anonymity and non-specific feedback might mitigate this.
Richard is concerned about Ken's role as a mediator. Ken and Steve
point out that Grace actually started to attend to the screen herself
as the lecture went along. Furthermore, Steve points out that this
setup (someone mediating the system [was actually at Grace's
suggestion --Ken] [might consider mediation as another way of easing
adoption of system --Ken]
Justin is curious how other students responded to the situation. Did they
feel that stopping and responding to laptop feedback was appropriate? No
response to this; perhaps we can get some of that sense from the P&P
folks? Can we survey some of the students who didn't participate?
Nick is curious whether these "hard stops" that occurred in class might
fall away as the instructor gets practice with it. Ken points out that we
have a variety of usage scenarios.
We discuss the first use of the system in class (the tracking number
explanation request). This is seen as a "perfect usage" of the system.
Several people point out that this will work to: comfort the students,
keep them focused, and avoid students "giving up" because they're lost.
Ken discusses the possible impact of this system on how well they follow
their own slides. He mentions one slide where Grace was about to go on but
some explain marks held her back and she hit a point she might have
otherwise skipped. Justin is unsure that holding to the slides is a good
thing, but Richard points out that if you're going to have slides, perhaps
you should hold to them. It's worth noting that this was one of the
strongest pieces of feedback that Martin received in his pen and paper
feedback from two weeks ago ("stick to the slides" style of feedback).
Doug suggests the ability to annotate verbal discussion (and presumably
any other random content outside of the slides). Ken mentions the MSR work
on anchoring annotations. Richard mentions freeform annotations.
Doug mentions that he (as an instructor) is very fond of the symbols as a
vehicle for feedback.
Doug brings up the dangers of a large wireless network: IM, game playing,
etc. Richard points out that people are bringing these devices in anyway..
may as well co-opt them for educational purposes.
Tammy suggests asking Grace what kind of learning culture she's trying to
create in her classroom.
|
 |
Department of Computer Science & Engineering University of Washington Box 352350 Seattle, WA 98195-2350 (206) 543-1695 voice, (206) 543-2969 FAX [comments to yasuhara@cs.washington.edu] |
