Rejection!
by john on Dec.15, 2008, under Head Tracking, PyCon, Python, wiimote
Together with my brother I submitted a proposal to PyCon to run a tutorial on wiiMote headtracking under Python. Given that they are running two days of tutorials this year I thought we had a pretty good shot. I guess we were wrong. After seeing on a mailing list that acceptance letters had been sent out over a week prior I wrote and asked for a formal rejection if such a thing was coming. A day later I got it.
I’d like to have gotten some feedback, but I can come up with some on my own. The first is that the topic isn’t very Python specific. It happens to use Python, but the same thing could be accomplished in any number of languages. Of course, in some sense this applies to almost any tutorial. You could write a website in any number of languages, but I’ll guess that there are a dozen tutorials dealing with web specific stuff.
The second reason is that this requires some hardware, a $40 wiiMote and some cheapie LEDs. I think this is a legitimate concern, but really this isn’t much of an expense for those interested in it and I’d have supplied the LEDs and associated components.
On the other hand, I think that the course was pretty compelling. I’ve included most of my proposal below. I was really looking forward to running this and seeing what people came up with. We’d have had the infrastructure ready for people to do all sorts of things in either 2D or 3D. Not just headtracking but maybe handtracking or something else.
Because of this I’m considering finding some open space during an evening at PyCon and running some version of this if people at interested. Given the number of people that came to see the headtracking demos we ran at night I’d guess that there would be some significant interest. Of course there was no good way to communicate this to the tutorial selection people and since there was no interaction I have no idea if this was even a concern.
So here’s the proposal. If you’ll be at PyCon and you’re interested in this, let me know.
Tutorial Format
===============
This tutorial will be comprised of a short overview of how wiiMote headtracking works followed by a hands on, interactive session. Tutorial participants will use presenter supplied hardware to build a basic headtracking rig, install necessary software on their machines to access a wiiMote via Python, and build an application that uses headtracking for interaction.
Libraries and sample applications will be provided.
Recording
=========
We give permission to record and publish our PyCon tutorial for free distribution.
Intended Audience
=================
This tutorial is appropriate for intermediate to advanced level programmers with an interest in computer visualization, game development, and human-computer interaction.
Maximum Number of Students
==========================
We could handle up to 60 students.
Prerequisites
=============
Students should be comfortable with Python and have some experience in a Python based graphics library such as PyGame or Pyglet.
Attendee Requirements
=====================
Attendees should bring a Linux, Mac, or Windows laptop equipped with bluetooth. Participants should bring a Nintendo wiiMote. If desired, participants could arrange on their own to pair up and share a laptop and wiiMote.
Presenters will supply basic equipment used to construct IR emitters for headtracking.
Presenter Requirements
======================
We will require a room with a projector and some space for participants to move around in. Two projectors and two screens would be helpful but not required.
Notes for Reviewers
===================
John presented on tracking lasers with webcams at PyCon 2008. At the end he threw in a headtracking demo which got a big reaction from the crowd. This tutorial will not only explain the hardware, software, and theory of headtracking but allow participants to apply it in a program of their own making.
Promotional Summary
===================
wiiMote headtracking demos are a YouTube sensation and the technology is making its way from demos to production games and scientific visualization. It allows users a unique, hands free way of interacting with computers. Learn the theory behind wiiMote headtracking and implement it in a program of your own. Participants will be encouraged but not required to form small groups to work together to create their own demo.
Detailed Tutorial Outline
=========================
0 – 30 minutes
————–
Presentation on:
* how the wiiMote works
* what the infrared emitters do
* headtracking theory
* our Python headtracking library
* how to use the library
* quick demo of navigating a 3D environment using headtracking
31 – 70 minutes
—————
Installing software on user laptops, getting them connected to wiiMotes, and verifiying installation
71 – 90 minutes
—————
Assembly of IR emitters, running a demo program to verify functionality.
91 – 120 minutes
—————-
Break! Refreshments and help stragglers with issues.
121 – 130 minutes
—————–
Brainstorm project ideas as a group.
131 – 190 minutes
—————–
Small groups form around demo ideas and code them up.
191 – 210 minutes
—————–
Small groups present their demos.
Note presenters will at this point arrange times for breakout sessions with groups during the rest of the conference if desired. We will also plan on a game night similar to last year in which people are given a chance to play with headtracking and laser games. Those that are interested can show their demos.
December 15th, 2008 on 10:25 pm
I would be extremely interested in this, but I am not yet sure I will be at PyCon. (Thesis defense rapidly approaching!) If it could be applied to broader applications — say, if it could be made to work with an interactor like TVTK or Coin3D — this would be perfect for a wide variety of scientific visualization needs. (Please feel free to contact me at the email address I provided; I was unable to locate your contact information.)
December 16th, 2008 on 4:56 am
Unfortunately almost all of the tutorial proposals were ‘compelling’, but there were simply too many.
I guess the ones accepted (I wasn’t on that committee) were mainly based on feedback from PyCon attendees as to what subjects they would *like* to see tutorials on.
Personally I would *love* to see hardware hacking tutorials – but I wonder if tying the talk too closely to a single device (the wiimote which is probably seen as being a proprietary device) hurt your chances a bit?
Anyway, you should *definitely* go for an Open Space and if you bring a ‘bag of bits’ could even turn it into an evening event. We are trying to have more of a focus on Open Space this year as many people feel they can be more effective than talks (I’m not convinced yet but willing to be convinced!).
December 16th, 2008 on 11:37 am
Michael Foord,
I wonder how such feedback gets transmitted back to decision makers. My talk last year was very popular and I was told my lots of people that it was the best PyCon talk they’d ever seen. Two people said it was the best talk they’d seen at any conference. I realize that I’m tooting my own horn here and that perhaps people were just being effusive to be kind. But this subject is so out of left field that of course people aren’t going to state on a survey that they’d want to see it, but if it is presented it will generate lots of interest. I really doubt that the organizers are aware of how popular this stuff is or how well the talk went over. So on one hand I can see how it is that they wouldn’t know, but on the other it is their job to know.
The difficulty of the Open Space versus a tutorial is that it is much harder to promote and those that might be interested (and willing to bring/buy a wiiMote) won’t hear about it until the conference begins. I guess I’ll have to try to do a lightening talk to promote it, which isn’t ideal (especially given the difficulty of getting a slot) but I don’t see a better option.
January 22nd, 2009 on 2:40 pm
I am interested in using the wiimote as a head tracker. I am only interested in using the output (elevation and azimuth or pitch and yaw) for non-VR related work. Is there a simple way to get the wimmote output without the additional VR software? Will a bluetooth application do this for me? Could the output be used in Matlab?
Thanks
January 22nd, 2009 on 10:25 pm
Colin,
I’ve put up a presentation that describes simple wiiMote headtracking and even includes some sample code. Hooking up to bluetooth with depend on your platform
January 23rd, 2009 on 8:32 am
Sorry – didn’t see your reply until now.
The feedback comes from the feedback forms at the end of the conference.
January 23rd, 2009 on 8:33 am
“Given the number of people that came to see the headtracking demos we ran at night I’d guess that there would be some significant interest. Of course there was no good way to communicate this to the tutorial selection people”
I included information along these lines when I submitted my tutorial proposal. Worth noting for next time.
January 23rd, 2009 on 3:39 pm
I submitted similar information with my proposal, but I assume that it just looked like I was showing off and probably hurt my chances.