Mailman 3 March 2009 - Edu-sig

musings on Javascript etc.
by kirby urner March 11, 2009

March 11, 2009

So still gangbusters for Chicago, though I haven't found a CouchDB to surf yet (stupid joke). PPUG was great last night, a really growing community, poor Produce Row for beers eh? All those paying customers. So in what ways am I pro Alan Kay? A very open minded guy in a lot of ways, worried about 'idiocracy' (I'll expound), and really a task master in terms of gym coach, like Richard Simmons on steroids. On the 'Idiocracy' front, he openly doubted whether 'people today' were smart enough to invent tcp/ip, like if we had to start all over, did we really have what it takes? On the open mindedness front, he was taking another look at JavaScript and really liking what he found, and that attitude communicated clearly enough to where I'd say my attending these Admirers of Javascript @ Cubespace are owing to his opening my eyes awhile back (Shuttleworth Foundation summit, me with the loud fan Toshiba). On the gym coach front, he'd openly diss with this "low pass filter" talk that drove people crazy, really challenging to hear. You could tell he admired Python though, a lot (he made no secret about it). So yeah, Alan the guy is impressive. I knew Arthur would be curious about my impressions, as he was very openly *not* an Alan Kay fan (talking about this very archive -- good reading). That's about the limit of my interactions. Here on edu-sig, I tried to communicate my discovery that he's actually a "slayer" versus Smalltalk, running a Buffy number against this early Frankenstein experiment, got OO out of the starting gate, but it's kind of like one of these horses at this point (from a deep DARPA past, like IDLE someday?): Anyway, thanks Alan for the heads up about Javascript. Probably my main disagreement with Alan's philosophy is I'm less attracted to neat, clean, recursive environments, prefer the more losely coupled heterogenous environment of competing paradigms. It's somewhat the difference between a great painter with a definite look and feel to explore, and an art curator, who enjoys artwork from multiple great painters (as I'm sure we all do). Like, I'm not that motivated to "solve" the problem of GNU/Linux being a world of bewildering complexity. Maybe Alan isn't either, in which case we're on the same side on this one. Note that I didn't say anything about Python, as I admire it too, just as Alan does. I cut my programming teeth on APL, found Python to be in that same groove for me, though not the same paradigm (APL was pre-OO in many ways, with the J language a more contemporary offering in the same lineage). I also thinking namespaces are a honking great idea. Honk honk! Back to PPUG, I've got the usual blog write-up. Basically, Python is doing well in the parsing, machine learning, and semi-structure data departments, though it's lagging vis-a-vis Tokyo Tyrant. Kirby

1 1

Education BoF at PyCon
by Vern Ceder March 10, 2009

March 10, 2009

I just wanted to remind everyone who's going to PyCon that we have a BoF/Open Space page set up on the PyCon wiki at http://us.pycon.org/2009/openspace/EduSig/ Please feel free to add your name to the list of attendees and also to add any suggestions for topics, activities, dinner, etc. I hope to see many of you in Chicago. Cheers, Vern -- This time for sure! -Bullwinkle J. Moose ----------------------------- Vern Ceder, Director of Technology Canterbury School, 3210 Smith Road, Ft Wayne, IN 46804 vceder(a)canterburyschool.org; 260-436-0746; FAX: 260-436-5137

1 0

Re: [Edu-sig] PyWhip - Keep on crackin' it!!
by David MacQuigg March 8, 2009

March 8, 2009

At 08:41 AM 3/8/2009 -0700, kirby urner wrote: >What I'm seeing at your site so far is a very low level workout in >Python basics -- not a criticism in any way, just suggesting you could >apply the same technique to more challenging problems. > >Here's one I like: > >Write a Python generator so you get successive rows of Pascals >triangle, like this: > >o = Pascal() # o for object >>>> next(o) >[1] >>>> next(o) >[1, 1] >>>> next(o) >[1,2,1] >>>> next(o) >[1,3,3,1] >.... >and so on (Pascal's triangle). Note I'm using Python 3.x syntax. > >One of the best solutions is: > >def pascal(): > row = [1] > while True: > yield row > row = [i + j for (i,j) in zip([0]+row, row+[0])] Nice! If you don't mind, I'll use this in our category "math" at level-3. It is more challenging than the level-1 we are focused on now, but the more advanced problems will come soon. The idea is to organize the problems in layers. Layer-1 is problems that don't depend on students knowing anything that isn't in a short help page on a specific topic, like strings. Layer-2 can depend on knowing all the basic techniques in layer-1. Thus, we can now ask string questions that are best solved with a loop. Layer-3 would include the more challenging problems, like the one you are suggesting. -- Dave

1 0

Re: [Edu-sig] PyWhip - Keep on crackin' it!!
by David MacQuigg March 8, 2009

March 8, 2009

At 10:51 PM 3/6/2009 -0400, Andre Roberge wrote: >David MacQuigg wrote: >>I started to do this as a downloadable program, then Athar jumped in and said he could do it just as easily in Google App Engine. He did in one weekend what took several weeks on my own server in an earlier project. >> >>The advantage of a web-based program is you don't have to download or install anything. > >I completely agree. I will definitely have to mention it during my Pycon talk about Crunchy. > >Well done! The credit so far, goes 90% to Athar, and of course, Nick Parlante (JavaBat.com) for inspiring us to even think of it. Now we need community input. We have 6 problems so far. I'll do strings. We still need math, logic, lists, dictionaries, and recursion. Let's make this the most popular Python site on the planet!! At 10:34 PM 3/6/2009 -0800, kirby urner wrote: >David MacQuigg wrote: >>The advantage of a web-based program is you don't have to download or install anything. > >This is true, but I think if you're not going to download Python and >actually run it locally, then you're mostly just kidding yourself if >you think you're committed to learning it. We need both - Python running on our own computers to do quick interactive tests, and PyWhip to see if we got the problem solved, track our progress, etc. PyWhip is not an interpreter, but a test environment. Let's say you are asked to find a particular sequence in a list, and you don't remember Python's indexing syntax. Much easier to try a few quick tests in IDLE, then write the complete solution in PyWhip. I even use Python/IDLE to solve problems in JavaBat!! I find in much easier to work out the solution in Python, then translate to Java. This is especially true for more complex problems, even big projects in Java. I have used Python as an Interface Description Language, and it beats the IDL tools I am aware of. I don't use UML diagrams anymore. I wish Java had an environment like IDLE. The closest I've found is Eclipse, and that is rather cumbersome compared to IDLE. >That being said, having a server running locally (including GAE, which >is designed to work on your laptop as well) is a great idea. If >you're connected to some remote curriculum using this configuration, >well, that's what Subversion is for (like with Django, just ask for an >update and ye shall receive). Subversion is very cool, but it won't keep track of user logins, session state, etc. It's all that "framework" stuff that took so long on my previous project. When you write a program in App Engine, Google provides it all for you, along with Django templates, database support, and a bunch of other stuff you don't have to develop yourself. >I'm more interested in students than teachers. I'd encourage them to >download Python locally, update curriculum source using some checkout >procedure, if that's what's required. How do you handle testing? Seems like automation is essential if you have more than a few students. -- Dave ************************************************************ * * David MacQuigg, PhD email: macquigg at ece.arizona.edu * * * Research Associate phone: USA 520-721-4583 * * * * ECE Department, University of Arizona * * * * 9320 East Mikelyn Lane * * * * http://purl.net/macquigg Tucson, Arizona 85710 * ************************************************************ *

2 1

PyWhip ???
by David MacQuigg March 7, 2009

March 7, 2009

Athat Hameed has finished the prototype of our online Python practice system. See http://py-bat.appspot.com We need to change the name so as not to conflict with Nick Parlante's JavaBat.com, which was the inspiration for this site. Athar and I like PyWhip, but we are open to suggestions. We need teachers, and anyone else who is interested, to contribute problems. Let's start with the categories Logic, Strings, Lists, and Recursion. Don't copy Nick's problems. He's planning his own Python site, with automated translations of his Java problems. We could also use a few more help pages. See my Strings example at http://py-bat.appspot.com/help These are basically tutorials, starting with the simplest examples first, and focusing on just those features of Python we are testing in the exercises. Try to resist including everything in Python. Python.org already has that. Let's put together some good problems for beginners, and inspire Athar to make the website as nice as JavaBat. -- Dave ************************************************************ * * David MacQuigg, PhD email: macquigg at ece.arizona.edu * * * Research Associate phone: USA 520-721-4583 * * * * ECE Department, University of Arizona * * * * 9320 East Mikelyn Lane * * * * http://purl.net/macquigg Tucson, Arizona 85710 * ************************************************************ *

5 6

PyWhip - Keep on crackin' it!!
by David MacQuigg March 6, 2009

March 6, 2009

We've moved the project to http://pywhip.appspot.com and fixed the permissions so anyone can now edit their own problems. At 01:35 PM 3/4/2009 -0800, kirby urner wrote: >This idea of "running Python over the web" doesn't seem necessary to >me, not sure why people think that's so important. I started to do this as a downloadable program, then Athar jumped in and said he could do it just as easily in Google App Engine. He did in one weekend what took several weeks on my own server in an earlier project. The advantage of a web-based program is you don't have to download or install anything. A small annoyance, for sure, but it adds up when we are making updates every day. Also, if we get serious about teaching Python to a class, the ability to check student progress online is critical. There are frameworks for doing automated testing (Web-CAT, etc.) but I like the way JavaBat does it so much better. >Anyway, best wishes on catching up in Arizona. Someday you'll be in >the same league as Oregon! Something to look forward to. I hadn't thought of the project as having any particular location. I guess you could say it is as much in Pakistan right now as in Arizona. U of A has nothing to do with it, and I don't expect it will be used there either, unless maybe my CS2 project gets some support. Then it will be useful as a quick transition from Java to Python, so we can spend most of our time on science and engineering. -- Dave ************************************************************ * * David MacQuigg, PhD email: macquigg at ece.arizona.edu * * * Research Associate phone: USA 520-721-4583 * * * * ECE Department, University of Arizona * * * * 9320 East Mikelyn Lane * * * * http://purl.net/macquigg Tucson, Arizona 85710 * ************************************************************ *

3 2

Re: [Edu-sig] Order of operations confusion
by Laura Creighton March 5, 2009

March 5, 2009

In a message of Thu, 05 Mar 2009 15:19:52 MST, Charles Cossé writes: >On Thu, Mar 5, 2009 at 2:52 PM, Laura Creighton <lac(a)openend.se> wrote: >> In a message of Thu, 05 Mar 2009 12:51:40 MST, Charles Cossé writes: >> >> The interesting question (to me at least) is whether he would be willin >g >> to rewrite a + b + c as a + c + b >> >> and then if he would be willing to rewrite >> a + b as a - (-b) > > > > >Yes, I confirmed that it was definitely due to having been repeatedly >told the simple rules for order of operations; he observed that "there >was something to be multiplied" and believed that he couldn't proceed >until he had carried out that multiplication, which immediately led to >a gridlock situation for him ... and he would have been okay rewriting >like you did there since there was no multiplication. Guess it's just >one of those many things that you don't realize kids think until you >come across it ... and then it's a surprise > >Charles I'm teaching 10 and 11 year olds here in Sweden, but over the summer holidays or on weekends in optional classes. I am always curious as to how the rest of the world teaches things. Can you ask him for a favour for me and see if he would be willing to accept that: 82 + (15 * 10) - 51 is the same as (15 * 10) + 82 - 51 or if the * is really stopping him cold. And thank him very much for me for answering my question, if he is willing. Laura

1 0

Re: [Edu-sig] Order of operations confusion
by Laura Creighton March 5, 2009

March 5, 2009

In a message of Thu, 05 Mar 2009 12:51:40 MST, Charles Cossé writes: >So, the equation we were dealing with was something like: >82+b*c-51=d-13, and the thing that surprised me was that he didn't >think it was legal to take 82-51 before having dealt with the b*c term I suspect that he thinks your notation is sloppy, and you should have written this as 82 + (b * c) - 51 The interesting question (to me at least) is whether he would be willing to rewrite a + b + c as a + c + b and then if he would be willing to rewrite a + b as a - (-b) Can you ask him? How old is he, anyway? Laura

2 1

Order of operations confusion
by Charles Cossé March 5, 2009

March 5, 2009

Hello All, Just thought I'd relay a rather interesting story: Last night I was having my son work-out a simple problem from a new little webapp that I coded up called "MakeItTrue" ( http://www.asymptopia.org/MakeItTrue.html ) which had 3 arbitrary variables embedded into a larger equation, and the objective is to simply choose 3 values which make a true equality of the thing.... So, the equation we were dealing with was something like: 82+b*c-51=d-13, and the thing that surprised me was that he didn't think it was legal to take 82-51 before having dealt with the b*c term ... on account of the "MyDearAuntSally..." or "BODMAS" ... or whatever acronyms they teach in school ... he had no idea that he could do that, merely because it had been driven-home in his head that multiplication must always be done first ... Well, I thought that was rather significant, so thought I'd pass it along to you all. Take care out there, -Charles -- AsymptopiaSoftware|Software@theLimit http://www.asymptopia.org

3 2

riddle way to learn python :)
by Jurgis Pralgauskis March 5, 2009

March 5, 2009

http://www.pythonchallenge.com/

1 0