I was surfing this AM and found this article: computer-science-is-essential-for-everyone. It has some interesting info and links on applying computing to non-science fields. Check it out.

Today's activities included:

• Writing a problem and solution in http://pearson.turingscraft.com/codelab (password access required), and debugging it (problem report required);
• Creating a github repo for my Commarch report on http://source.android.com/source, the Android Open Source Project;
• Learning the value of http://www.ohloh.net;
• Downloading and running git_by_a_bus from github;
• Downloading and running gource from mac ports (maybe I should have done it from github, but port ensured I would get all the many dependencies; I hope they are not duplicated now...);
• and learning that WebWork (WW) does not compile and run the python code at this time, although it does (or can) compile and run Java code using JUnit.

Some questions for Friday's discussion:

1. (Is this really Joe Geigel's question?) How do you go about designing/planning the establishment of an open source project that you envision as an ongoing venture?
2. How do you know where to look for information in a project?
3. How do you learn where to put a (non-code) contribution to a project? How do you find the right person to ask? These are issues with those 'medium' size projects in which there may appear to be several persons from which to choose.

I played WebWork (WW) course developer today.

In the UR's server, https://hosted2.webwork.rochester.edu/webwork2 is a 'course' named steel_experiment that Mike Gage created for me. I copied a Cornell-Pace experimental python course set, and used it to create a library of python problems. The name of this is RITPython in the WW library.

I learned how to create and edit homework problem sets, create and modify students, and fix simple bugs found in the original Cornell-Pace problem set. This was a learn-by-doing-and-exploring activity, filled with many, sometimes circular routes, but I got a good feel for what and how an instructor would have to do.

The bug that Jason and I found yesterday was a bug in the Cornell-Pace content. There was an extra answer line in the problem, and the WW infrastructure caught it by walking off the end of a hash table. The fix to the content was simple because the WW error report identified the line and the error in the infrastructure. (It helped that I downloaded the WW source from github yesterday; though I was not running WW, I had the code and could see why the problem occurred from a combination of the course's problem and the WW source.)

With respect to POSSE, this was a tangent; I did not do anything with the WebWork infrastructure. On the other hand, I did get to use IRC in a real situation -- I had to chat with Jason Aubrey and Mike Gage, who were over at UR, to get answers.

The git stuff was good today, although some may have been too much too soon. There were a lot of more advanced items mixed into the basics.

Mike Gage set me up with an account on WebWork to try out the Python stuff, and Jason Aubrey showed me how to get the python examples loaded. I'm going to use IRChat to work with them as I go through experimenting with applying this to python Wednesday.

I'm also interested in the posse site evolution because of its application to courseware.

How/whether I will get into Grand Central Dispatch on ubuntu seems to be falling by the wayside...

Graybeard thought of the day: Non-linearities of open source activities, tools and techniques can be overwhelming for those accustomed to the older RTFM model.

Chat is useful for short answers from a distant person(s). I think wikis are best for information such as was on the etherpad yesterday. The etherpad seems to me too informal for the detailed information such as chat commands, registering your nickname, etc. Just my 2 cents.

Mark Wambach and I had a nice chat last night. Partial transcript is below. I've used AIM in a work context before, and it was essential with a far-flung team. Today Skype seems more up-to-date.

[edited]
[21:04]  The workshop?
[21:05]  that, and this IRC chat and anything related ...
[21:05]  Well, I used to use mIRC eons ago, so this feels a bit primitive.
[21:06]  i feel the same. we used AIM in 2000 for long distance collaboration
[21:07]  I've used Skype to talk to a cohort in China (both voice and text).  [21:07]  I am hoping to end up with a relation with RedHat at some time after this week.  And, to be somewhat immersed in OVIRT.
[21:08]  Skype is much more modern, synchronous and immediate. personally i feel wikis are better for a bunch of this stuff
[21:08]  What is OVIRT?
[21:08]  OVIRT is a free version of hyper manager that runs on top of KVM, to manager your virtual environment.
...
[21:11]  first there has to be a real need for elec. connect. too bad an on-site seminar can't really use it.
[21:12]  What do you mean about the real need for Elec. connect?  It is somewhat true that until this point, I've felt no personal need.
[21:12]  well, if you are in the same room with a colleague, using an IRChat mechanism is silly.
[21:12]  HA ha, very true
[21:13]  The VM I want to set up is for ubuntu experimentation at the OS level.
[21:13]  Yeah, I heard you talking about OS today, what was that about - was it related to the VMs?
[21:13]  I am hoping that that UR 'camp' will enable us to really use IRC to collaborate at a distance; that would make sense.
[21:14]  Yes, are you going to participate in WebWork?
[21:14]  Yeah. I want to get a VM set up to experiment with some libraries that Apple has open-sourced. it is for enhancing performance (and application) of apps using threads,
[21:15]  I am curious about webwork but concerned about its foundation on perl
[21:16]  No clue why the picked Perl unless because it's ubiquitous.  I only use it periodically.  What would be the problem or problems with perl?
[21:17]  I use and write perl scripts all the time for grading projects. maybe it's ok, but they have defined a dialect.
[21:18]  And, it sounds like they're looking for programmers.
[21:19]  The main reason I'm interested in webwork is its potential for use in CS1 for homework assignments; that would involve programming but in a different application from theirs (math).
[21:21]  The apple software is called "Grand Central Dispatch"...
[21:22]  And, I'm hoping to contribute to OVIRT after I have come up to speed a bit and participated.  I am really going to miss VMWare's Lab Manager (plus it's not really free).

It's June, 2012 -- time to get productively lost into free and open source.
POSSE 2012 at RIT has begun.

I was going to use this workshop to look at open source applications in the CS department's OS course -- something to do with Apple's open sourcing of Grand Central Dispatch. Now, maybe WebWork from UofR might be something to explore for CS1. We shall see as the week unfolds and I follow my nose.

The "Fail Better" quotation

Here's the quotation I was paraphrasing last week:
"All of old. Nothing else ever. Ever tried. Ever failed. No matter. Try again. Fail again. Fail better."
http://www.goodreads.com/author/quotes/1433597.Samuel_Beckett