Mailman 3 September 2007 - Edu-sig

Python for Halloween
by kirby urner 21 Oct '07

21 Oct '07

Something about using 'ghost' in place of 'self' in a Shell class (allusion to anime as well). Maybe I did something like this last year (having deja vu again). But I'm realizing edu-sig is no longer a preferred venue for my curriculum writing. I'd rather huddle with MIT, Bruce Sherwood, Alan Kay and play *against* Arthur's team. Arthur, you make too many enemies too quickly and too easily. Even more, I'm into my Portland Knowledge Lab (PKL) business (gotta pay that rent today) and want to make lots of Python and Synergetics teaching cartoons for HDTV (my Saturday Morning meme), with lots of brand name sponsors. But here is not the place to brainstorm in safety. Too many prying eyes and self-appointed kibbitzers. So I'm actually resigning this time. Don't expect me back any time soon. Probably won't be tracking either, at least not with any frequency. For me, it's a tilt towards private lists and closed source development (not exclusively -- I'm still a big supporter of a strong public sector). I'm learning why open source is a bad idea sometimes. I've talked to Stallman about this, though he probably doesn't remember -- I was in his audience in Portland and asked about situations when "they" simply *don't deserve* your best code. But I think more it's a matter of not wanting to show unfinished work (a big no no in Fuller's self discipline, according to 'Critical Path' I think it was). Kirby

2 3

News from the K-corner
by kirby urner 25 Sep '07

25 Sep '07

Contract renewed with SA: to teach Python; they liked my Lithuania slides, clear even to non-geeks. Making new agreement with LEP High, also to teach Python, but more as a special case example of what open source entrepreneurship is all about. LEP pix: http://www.flickr.com/photos/17157315@N00/sets/72157602156242583/ My focus will be: how open standards, open specs, don't keep us from competing or compensating, in games to offer better products. The difference: a more level playing field, as there's no unfair protection against the competition having access to the same powerful and effective development tools, as these are by now already paid for in full and handed off to our children, their children at large, per GNU, GPL, OLPC etc. On another day job front, I completed a so-called BarCamp Alpha in Centralia, a really picturesque slice of North America and its many intersecting traditions: http://www.flickr.com/photos/17157315@N00/sets/72157602124435584/ The purpose of this BarCamp was to look at open source in the area of clinical research (both outcomes and quality assurance). Centralia is approximately half way between Seattle and Portland, and so is good neutral ground in which to bury hatchets and form new collaborations (relations between these two metropolitan regions have been rivalrous at times, including these days, what with Portland, not Seattle, being "an open source capital" and all: http://www.csmonitor.com/2005/1128/p03s02-ussc.html More in my blog re BarCamp Alpha: http://controlroom.blogspot.com/2007/09/barcamp-alpha.html On the Fuller School front (a thread in this archive, especially when Arthur and I were arguing (i.e. approximately all the time ( http://pygeo.sourceforge.net/ ))), there's some fresh propaganda in Good Magazine, which I appreciate: http://www.goodmagazine.com/section/Guide/good_guide_r_buckminster_fuller (You might still find some hard copies on the stands -- bright orange with AK-47 x-rayed on the cover). Kirby 4D Solutions Portland, Oregon

1 0

Python course at Foothill College
by Elaine 24 Sep '07

24 Sep '07

If you would like to learn Python, Foothill College in Los Altos Hills California is offering a course starting this Wednesday evening, 26 Sept. The course is designed for students who are already familiar with some type of programming. Here is the course description: CIS 68K "INTRODUCTION TO PYTHON PROGRAMMING" 5 Units This course will introduce students to the Python language and environment. Four hours lecture, four hours terminal time. Advisory: CIS 15A or 27A, and CIS 68A. 2182 CIS -068K-01 LEC6:00PM- 9:50 Wednesdays - HAIGHT Middlefield Campus, Room I5. Course fee, $4. If you would like to sign up for the class, it would be very helpful if you would register beforehand by going to: http://www.foothill.fhda.edu/reg/index.php If you have questions, you can contact the instructor at: haightElaine(a)foothill.edu ____________________________________________________________________________________ Be a better Heartthrob. Get better relationship answers from someone who knows. Yahoo! Answers - Check it out. http://answers.yahoo.com/dir/?link=list&sid=396545433

2 1

Py3k Trivia :-)
by Guido van Rossum 21 Sep '07

21 Sep '07

Would you believe there's a Curious George episode named "Curious George Vs The Turbo Python 3000"? """ George isn't tall enough to ride the greatest rollercoaster of all time, The Turbo Python 3000. He uses licorice whips to measure his height and determines that he is 7-whips tall, one short of the 8-whip minimum! """ http://pbskids.org/curiousgeorge/parentsteachers/program/ep_desc_3.html -- --Guido van Rossum (home page: http://www.python.org/~guido/)

1 0

Python 3.0a: investigating changes
by kirby urner 20 Sep '07

20 Sep '07

So I'm in a Portland coffee shop, iTunes on the speakers, daughter reading a novel. I remember Guido saying something about 'next' changing a little. What I'm finding tonight is the old some_generator.next() syntax is breaking (by design I assume), while next(some_generator) works great. What's getting triggered under the hood is __next__, which of course you can trigger directly. >>> thegen = (i for i in range(10)) >>> thegen <generator object at 0xb7c9430c> >>> type(thegen) <type 'generator'> >>> thegen.next() Traceback (most recent call last): File "<stdin>", line 1, in <module> AttributeError: 'generator' object has no attribute 'next' >>> dir(thegen) ['__class__', '__delattr__', '__doc__', '__eq__', '__format__', '__ge__', '__getattribute__', '__gt__', '__hash__', '__init__', '__iter__', '__le__', '__lt__', '__ne__', '__new__', '__next__', '__reduce__', '__reduce_ex__', '__repr__', '__setattr__', '__str__', 'close', 'gi_frame', 'gi_running', 'send', 'throw'] >>> next(thegen) 0 >>> next(thegen) 1 >>> next(thegen) 2 >>> def fibbo(a=0,b=1): ... while True: ... yield a ... a,b = a+b, a ... >>> somegen = fibbo() >>> type(somegen) <type 'generator'> >>> type.next() Traceback (most recent call last): File "<stdin>", line 1, in <module> AttributeError: type object 'type' has no attribute 'next' >>> next(somegen) 0 >>> next(somegen) 1 >>> next(somegen) 1 >>> somegen.__next__() 2 >>> somegen.__next__() 3 >>> somegen.__next__() 5 >>> somegen.__next__() 8 >>> somegen.__next__() 13 >>> somegen.__next__() 21 >>> and so on. Kirby

1 0

Lore question (re BarCamp) and MVC link
by kirby urner 19 Sep '07

19 Sep '07

I'm planning this little meetup for open sorcerers and doctors (as in medical) and am modeling it after BarCamp. If any of you on edu-sig see shallow bugs or other patches you want to make, here's from some actual recent correspondence sent to a doc friend of mine. Take pot shots at will... """ My model is Portland Barcamp but without the cubes (McMenamins tables and booths instead -- same idea). Dunno if we have wireless -- not having a major blow to the analogy. Barcamp doesn't mean a Camp for Wild Drinking (one meaning of "Bar"). Rather, in geek world, we use FUBAR (a military acronym) as a source of Foo and Bar. These then become the nonsense syllables we so often need to stand for "whatever". FooCamp was taken by O'Reilly (the publisher and guy, Tim). BarCamp materialized soon thereafter, or so goes the lore. """ So what if anything does this have to do with Python? I think of Python mainly as a candidate controller in this very generic picture here in my blog: http://controlroom.blogspot.com/2007/09/more-curriculum-writing-re-mvc.html Probably missing Pirate Festival this upcoming weekend, given it's also Race for the Cure, plus "working vacation" Friday (hotel in the boonies). Pirate Festival last year: http://mybizmo.blogspot.com/2006/09/yar.html Kirby

1 1

More about unicode...
by kirby urner 17 Sep '07

17 Sep '07

The kinds of lesson plans I'm collecting involve storing proper names in their native / home languages, with round trips twixt browser front ends and SQL back ends, Python a controller. It's especially reassuring, when you're in some foreign country on a hospital bed, in the care of strangers, when you see your own name is at least spelled correctly in the right character set on the monitors, along with all those colorful graphs and whatever (beep beep....). Means maybe they're used to multiple cultures, are worldly wise and well funded? Savvy hospitals probably attract top doctors? Maybe this is like a Federation Starship or something? "Kirk?..." "I think he needs more meds doctors..." [ scene goes dark ]. "Bones?..."

1 0

More regarding snake imagery...
by kirby urner 13 Sep '07

13 Sep '07

I'll tie this to education but regarding Pythons and constrictors in general, they have this tendency to eat rodents. So from a marketing perspective, what's "the rodent" in Python World (the computer language)? I think a good answer is: whatever knowledge domain. We wrap around new namespaces quickly and squeeze out all available info in a mercifully short time, then digest what we've learned, and in some sense we "become the mouse" i.e. are now ready to solve problems in ways that fit the needs of that domain. "Python stays relevant" might be the slogan here. The tie to education is unavoidably OLPC-related but I'll risk a quick quote from a recent post to Synergeo (Ken G. Brown listowner): """ Just saying the word "Python" is enough to scare the bejeezus out of the ignorant, as snake imagery tends to be off-putting in some cultures (attractive in others). """ and from the same post (# 35677): """ The last thing we need are lots of pith helmeted MIT "bwana types", traipsing through the Jungle with entourages carrying crates of XOs on their heads, like in some kind of B-grade Tarzan movie. """ In sum, I think it's up to locals familiar with various "locales" (what they call namespaces in J **) to figure how best to handle the snake imagery. We can just hope they do something charming, offering what others -- or we ourselves -- have done as a guide: http://mybizmo.blogspot.com/2007/07/few-photos.html http://controlroom.blogspot.com/2007/04/knotted-snake.html http://mybizmo.blogspot.com/2006/09/sculpture-with-found-objects.html Kirby ** Locale A namespace in J. The locale in which a public name is defined is an attribute of the name. A locale is identified by a locale name which is a character string not containing an underscore. http://www.jsoftware.com/help/jforc/glossary.htm

2 2

Rich Data Streams
by kirby urner 05 Sep '07

05 Sep '07

The idea of rich data structures was to take advantage of the Internet by providing students with more meaningful precoded data through which to plow. Examples: * all the bones in the body as a tree structure (with some fancy networking among skull bones) * GIS info (I posted cities.xml from Winterhaven) * the eight planets of our solar system, plus Pluto and Iris (the latter owned by CBS (smile)). The idea of a rich data stream anticipates Python 3's strong powers around data bitting (new bit type), which'll make OO dissection of IPv6 packets (legacy IPv4 packets) a cinch. Kids'll groove in their ability to really dissect what's going on over the wire (sniffed from the ether or whatever). Rather than encourage too much interloping on neighbors' comm channels (a favorite hacker pass time I realize), we need canned data streams of a simplified "cave painting" nature, with known solutions. Like it'll be a goal of the packetsniffer.py package to zip through these canned packets at high (enough) speed, and to spit back reports on whether they're tcp or udp, used by which flavored protocols and so on (you need ideas about protocol if you hope to reconstruct the payload, for all but the simplest of containers). Students will grow to understand tcp/ip by reading packetsniffer's deliberately *non*-obfuscated source code. What a relief to see it all in Python, vs. staring at low-level C structs all the time. Most would die of boredom before getting there. Python keeps it less boring. I'm aware that the Python Community hasn't waited for Python 3 before beginning work in this genre. Python has some of the more sophisticated networking tools around, including flagship peer to peer bittorrent in wx. Python is *truly* an effective language, which is why certain unnamed schools of low standards prefer students to concentrate on maybe Visual Basic or Java. The former is toothless enough (minus the ActiveX add-ons), and the latter is difficult enough, that relatively few in those two camps are as likely to experience the free flowing and comprehensible style we Pythoneers simply take for granted. Tough luck for them, and best wishes on some next iteration.** Kirby ** of course this rhetoric is full of false dichotomies, I realize. Many of us who code in Python likewise enjoy Java and VB.NET sometimes, or C# or whatever. I came to Python via Java from xBase (which I still use), and before that APL (which got me spoiled on always wanting REPL). So to heap scorn on BASIC, which I'm not doing, is really ahistorical anyway. There wasn't any Python way back then, especially not on the ROM chip when you booted the PC. However, I *do* still oppose this misguided "Python: the new BASIC" meme. I think that's stupid, as it overlooks Python's way stronger potential owing to its much higher level of design sophistication. "Python ain't yer grandfather's BASIC" is the slogan I'd prefer. Too bad JavaScript is so poorly named, as it'd be a better runner up to Python (because of the DOM) than most 2nd banana languages.

6 14

Another Pythonic meme ("color depth")
by kirby urner 03 Sep '07

03 Sep '07

This shift in Python 3 (out in alpha) to a unicode basis, I'm comparing to the move from 16 and 256 colors to millions of colors. The analogy is quite correct, as what it takes to map more colors are more bits, to give a bigger mapping. Same with the space of human languages. I sense a lingering imperialism in some circles that's forlornly clingling to the obsolete notion that all source code of strategic value will stay in the English language or at least in the Roman character set (romanji as the Japanese call it). But I think Python, along with Perl, is doing the right thing in trailblazing a future look and feel. Romanji keywords, OK (that's just part of the syntax in this case), but Lithuanian variable, function and class names, are likewise no problem, at least in 3rd party modules (the vast majority of Python code is 3rd party). We're not there yet, but we're moving in the right direction. I was at one of those Hilton hotel style parties in Bellvue recently (near Microsoft HQS), where some the the guests spoke "beltway" (a curious blend of DC based mumbos jumbo), was chatting with a Voice of America PR person and she's definitely in our camp on this one. Unicode is the way to go. www.voa.gov. Just as English Al Jazeera is taking the time and trouble to reach North Americans in their native vernacular (whether TV execs choose to block distribution is another matter -- Google's YouTube netcasting is a good workaround), so is Voice of America (VOA) netcasting in Urdu. Python's new sophistication around Unicode bodes well for i18n in our gnu math curriculum as well (we make much use of these agiles, as they make the whole field *so* much more accessible -- helps level the playing field). Of course Ubuntu and other Linux distros, Plone, etc. already have a strong track record when it comes to breaking away from "one size fits all" mentality. Kirby

1 0