From andre.roberge at gmail.com Sat Nov 1 04:57:52 2008 From: andre.roberge at gmail.com (Andre Roberge) Date: Sat, 1 Nov 2008 00:57:52 -0300 Subject: [Edu-sig] ANN: Docpicture 0.2 Message-ID: <7528bcdd0810312057o9528e70jeb19789189df5868@mail.gmail.com> Hello everyone, Docpicture 0.2 has been released. You can download it from http://code.google.com/p/docpicture/ Docpicture's goal is to enable embedding pictures inside Python docstrings using some custom domain specific language (dsl). docpicture includes a plugin architecture enabling users to extend it by adding new dsl parsers. Docpicture behaves as an enhanced help() for Python. It requires a default browser capable of displaying svg images (e.g. Firefox 3) ======================= This version of docpicture can display the following type of images: 1. turtles (!). This *has* to be a selling feature for people on this list! ;-) 2. images downloaded from the web (in any format) 3. Mathematical equations (requires matplotlib) 4. UML sequence diagrams (require a live internet connection) 5. Full matplotlib plots. This is very much a proof-of-concept version. Comments and suggestions are most welcome. Code contributions [for other types of images] even more so. ;-) Andr? Roberge From echerlin at gmail.com Sat Nov 1 07:54:57 2008 From: echerlin at gmail.com (Edward Cherlin) Date: Fri, 31 Oct 2008 23:54:57 -0700 Subject: [Edu-sig] More preaching to the choir... In-Reply-To: References: Message-ID: On Sun, Oct 19, 2008 at 4:41 PM, kirby urner wrote: > > Yes, not that hard, agreed. I think we're in the right headspace, here on > edu-sig, re teaching physics / math and so on with all this well known > content, plus adding more experimental fun around the edges, ala programming > just for the fun of it (ab attitude we highly encourage). > > Given how OO makes "objects" so concrete (per Concrete Mathematics) it's not > so unrealistic to suggest adding some quaternions to the mix, just one more > animal in our zoomorphic kingdom (or queendom, not pretending to know). Well, once you get to that point, you can easily add octonions, spinors, tensors, operators, crystallographic groups, and quantum groups. ^_^ However, I have started a project at EduForge.org that I call Kindergarten Calculus, to approach the issues from the opposite direction. The question is how to demonstrate the fundamental concepts of calculus visually to preschool children with no numbers and no symbols. Limits, tangents, maxima and minima, definite integrals (area under a curve), and other materials. I know how to start, but not how far we can go. I can certainly show the Fundamental Theorem of Calculus and the Mean Value Theorems. There are some simple cases of Calculus of Variations that have physics equivalents. We have to think about what has an algorithmic equivalent that we can program visually. Anyone interested is welcome to join the project. > Given XO is hardware with access to the cloud (by design) it's unnecessary > to map curriculum to it or any other hardware device, as this isn't about > hardware in the first place, but curriculum, and there's so much already out > there, much of it very pre-computer in flavor, yet nevertheless relevant, > Hamilton's brief included, Kepler's and so on. I would like to factor each of those lessons into concepts that can be taught at different ages, and to introduce them in sequence at the appropriate ages. This should produce much deeper understanding than waiting until students are supposed to be able to do it all at once. Thus Galilean gravity in a uniform field can be demonstrated via programming to third graders, while full Newtonian gravity requires calculus. > That being said, I like the idea of equipment especially designed with the > needs of young children in mind. I was gleeful that so many adults found > the XO frustrating to use because of the tiny keys. That's the whole point > -- it's for them, not you. Of course in this sense there's a need for Sugar > friendly apps, and/or new kinds of Sugar. I'm eager to find out more. > > All that's really needed, in terms of equipment, is a decent browser and > Python itself (if teaching Python). In addition to writing stuff for > laptops (multiple platform), I'm aiming at the LCD (or "flatscreen") market, > piping directly to coffee shops (e.g. booths), so an intimate group might > study together without bringing any hardware whatsoever. The same LCD plays > music and does Apple style visualizations i.e. isn't just for "study hall" > activities (these might be small screens, between HD and iPod, XO a source > of ideas). > > So now that we know what the future looks like, I'm wondering how geeks will > spontaneously self-organize to deliver a quality product. I imagine more > Rich Data Structures will be a part of it, as in canned Periodic Table > modules, with XML i/o if not natively XML -- I'm somewhat influenced by > Ruby's anti-XMLism, GIS data re cities (example @ my site), insectavora and > so on. > > However, I'm not expected all the stress of development to fall on open > source developers. Much of this stuff will be kept under wraps and niche > marketed under more restrictive licensing, simply because there's a market > for it and open source isn't the answer to every prayer, even if it is to so > many. Several of us here wear multiple hats in that regard e.g. contribute > to open source yet support clients who don't choose to compete in that arena > -- describes my situation at least. > > I'll be interested to hear more about what develops in those projects I > learned about at the most recent OSCON, as many of those *were* open source, > just not based anywhere close to Portland. As you might imagine, I have a > hard time keeping up even just with what goes on in my home town. > > Kirby Urner > Fine Grind Productions > > > On Sun, Oct 19, 2008 at 3:30 PM, Edward Cherlin wrote: >> >> 2008/10/19 kirby urner : >> > >> > >> > Here's a pointer to some related writings @ Math Forum, which I used to >> > tell >> > Arthur S. (this archive) was more like "center ring" in my circus, i.e. >> > where math teachers meet irrespective of caring about geek subculture, >> > computers etc.: >> > >> > http://mathforum.org/kb/thread.jspa?threadID=1845616&tstart=0 >> > >> > Re planets in orbit, another interesting implementation, already field >> > tested, multiplies complex numbers of unit radius, begetting rotation, >> > then >> > scales appropriately. Of course these orbits are merely circular, so >> > not >> > that realistic, so apply some field distortion if wanting elliptic (not >> > something we tried). >> >> I did a simulation of elliptical orbits in TutSim on the Apple II in >> 1982. It isn't that hard. I know of several different ways of >> programming it, using inverse-square gravity, Kepler's law of equal >> areas, and other equivalent mathematical representations. I set up a >> three-body simulation in which one object was thrown out and the other >> two settled into elliptical orbits around their barycenter, after a >> chaotic early period. There are such simulations in a number of >> languages, including Python and Smalltalk, where students can vary >> parameters and watch the results. Some of this is in the Sugar >> software for the OLPC XO. >> >> A. K. Dewdney's book The Planiverse describes simulations of inverse >> linear gravity in 2D, and others have done inverse cube gravity in 4D. >> You can get weird weather, seasons, and climate effects on planets in >> 4D with two independent axes of rotation, depending in particular on >> whether the rotation periods and the revolution period are close to or >> far from simple integer ratios. >> >> > Here's some code: >> > http://www.4dsolutions.net/ocn/python/orbits.py >> > >> > Kirby >> > 4D -- http://wiki.sugarlabs.org/go/User:Mokurai Give One, Get One, from Nov. 17 http://www.amazon.com/xo http://wiki.laptop.org/go/XO_Giving/International From kirby.urner at gmail.com Sat Nov 1 14:44:46 2008 From: kirby.urner at gmail.com (kirby urner) Date: Sat, 1 Nov 2008 06:44:46 -0700 Subject: [Edu-sig] More preaching to the choir... In-Reply-To: References: Message-ID: On Fri, Oct 31, 2008 at 11:54 PM, Edward Cherlin wrote: > On Sun, Oct 19, 2008 at 4:41 PM, kirby urner wrote: >> >> Yes, not that hard, agreed. I think we're in the right headspace, here on >> edu-sig, re teaching physics / math and so on with all this well known >> content, plus adding more experimental fun around the edges, ala programming >> just for the fun of it (ab attitude we highly encourage). >> >> Given how OO makes "objects" so concrete (per Concrete Mathematics) it's not >> so unrealistic to suggest adding some quaternions to the mix, just one more >> animal in our zoomorphic kingdom (or queendom, not pretending to know). > > Well, once you get to that point, you can easily add octonions, > spinors, tensors, operators, crystallographic groups, and quantum > groups. ^_^ Some could maybe. Quaternions predate Gibbs-Heaviside vectors as we learn 'em in school these days, have strong application in the gaming industry, as faster than rotation matrices in a lot of ways. I've published some work on this already e.g. http://www.4dsolutions.net/ocn/oopalgebra.html http://www.4dsolutions.net/ocn/numeracy1.html > However, I have started a project at EduForge.org that I call > Kindergarten Calculus, to approach the issues from the opposite > direction. The question is how to demonstrate the fundamental concepts > of calculus visually to preschool children with no numbers and no > symbols. Limits, tangents, maxima and minima, definite integrals (area > under a curve), and other materials. I know how to start, but not how > far we can go. I can certainly show the Fundamental Theorem of > Calculus and the Mean Value Theorems. There are some simple cases of > Calculus of Variations that have physics equivalents. We have to think > about what has an algorithmic equivalent that we can program visually. > > Anyone interested is welcome to join the project. > So like Sourceforge then, but for lesson programming. Cool. >> Given XO is hardware with access to the cloud (by design) it's unnecessary >> to map curriculum to it or any other hardware device, as this isn't about >> hardware in the first place, but curriculum, and there's so much already out >> there, much of it very pre-computer in flavor, yet nevertheless relevant, >> Hamilton's brief included, Kepler's and so on. > > I would like to factor each of those lessons into concepts that can be > taught at different ages, and to introduce them in sequence at the > appropriate ages. This should produce much deeper understanding than > waiting until students are supposed to be able to do it all at once. > Thus Galilean gravity in a uniform field can be demonstrated via > programming to third graders, while full Newtonian gravity requires > calculus. Yes, my goal as well, to develop a spiraling sequence of topics. However I tend to de-emphasize calculus as a standalone topic, merge it back into the physics from whence it derives, freeing up more of the math curriculum for the group and number theory topics we need to explain RSA pre-college. This will be a focus of a Pycon 2009 tutorial, if it gets approved. Kirby From kirby.urner at gmail.com Sat Nov 8 01:10:46 2008 From: kirby.urner at gmail.com (kirby urner) Date: Fri, 7 Nov 2008 16:10:46 -0800 Subject: [Edu-sig] Using YouTube Message-ID: Once again, I was so happy to have Python + VPython handy for my geometrical studies, what brought me to Python in the first place, by way of Java, trying to write scene description language for a ray tracer at first (POV-Ray, still using it), then VRML (today x3d), then finally, real time (VPython). Given I'm so focused on such spatial-slash-polyhedral studies, I've never been able to get away with just "flat stuff", consider SVG, Pygame and even Flash mostly flat. Java3D is good for round stuff, or whatever Scott Vorthmann is using to write vZome, the virtual Zome builder and free Java download. Ruby also good at integrating OpenGL, expect lots of good curriculum writing for that language already in the pipeline. I'll be interested to see if VPython comes out with a 3.x edition and how soon, as there's some tendency in the teaching community to start right in with the new syntax, not go with the old "print" only to have to relearn the new one, plus so many new shortcuts and add-ons to master, why waste time with 2.x? (is the attitude). Anyway, here was today's effort, stemming from lengthy investigations going on in Minnesota involving dissections and assemblies of five-fold symmetric polyhedra such as the rhombic triacontahedron, always looking for bridges between four and five fold: http://www.youtube.com/watch?v=nEpnOE9Cw_Y What you've got here is a rhombic triacontahedron "shrink wrapped" around a sphere of unit radius, then blown up by a scale factor to a volume of precisely 7.5, only to intersect with a rhombic dodecahedron likewise shrink wrapped, of volume 6 (all volumes in tetravolumes). The 7.5 volumed shape has radius phi/sqrt(2). This all goes by pretty fast in Akbar font, then goes to the VPython canvas, for some insider views. :) Kirby 4D From macquigg at ece.arizona.edu Sat Nov 8 13:38:51 2008 From: macquigg at ece.arizona.edu (David MacQuigg) Date: Sat, 08 Nov 2008 05:38:51 -0700 Subject: [Edu-sig] Ruby in the Curriculum In-Reply-To: Message-ID: <5.2.1.1.0.20081108052936.031167d0@mail.ece.arizona.edu> At 04:10 PM 11/7/2008 -0800, kirby urner wrote: >Ruby also good at integrating OpenGL, expect lots of good curriculum >writing for that language already in the pipeline. Wouldn't it be better to port the good stuff to Python? Same for Ruby-on-Rails. The more we fork, the longer it will take to get Python where it belongs, ahead of Java in our curriculums, preferably at the Freshmen level or earlier, with Java, C++, Ruby, etc. for those who are going to be full-time programmers. From kirby.urner at gmail.com Sat Nov 8 16:47:55 2008 From: kirby.urner at gmail.com (kirby urner) Date: Sat, 8 Nov 2008 07:47:55 -0800 Subject: [Edu-sig] Ruby in the Curriculum In-Reply-To: <5.2.1.1.0.20081108052936.031167d0@mail.ece.arizona.edu> References: <5.2.1.1.0.20081108052936.031167d0@mail.ece.arizona.edu> Message-ID: On Sat, Nov 8, 2008 at 4:38 AM, David MacQuigg wrote: > At 04:10 PM 11/7/2008 -0800, kirby urner wrote: > >>Ruby also good at integrating OpenGL, expect lots of good curriculum >>writing for that language already in the pipeline. > > Wouldn't it be better to port the good stuff to Python? Same for Ruby-on-Rails. The more we fork, the longer it will take to get Python where it belongs, ahead of Java in our curriculums, preferably at the Freshmen level or earlier, with Java, C++, Ruby, etc. for those who are going to be full-time programmers. > Yeah, I'd say it's up to each school, as the field may have morphed to the point where consensus on any "one right way to go" has disappeared, replaced by more complex demographics. Java is still a good choice but many will tell you it's easier to learn Python first, plus that gives you the Jython option, access to more graphical libraries (I've not explored this much myself yet -- trying to do something highly geospatial in Jython). I also wonder whether a four or two year college is the best way to cut your chops for private sector computing. But does the word "best" have much operational significance here? Different strokes for different folks, is what it comes down to. Sometimes, when hiring, it's that kid out of high school, the Django jockey, looking for some experience before considering more schooling, who stands out in the interviews, other times its that curmudgeony guy with the tweedy jacket and three degrees. I was at Powell's Technical recently, taking snap shots of Python vs. Ruby vs. Perl vs. Lisp in terms of shelf space, came up with 3, 5, 3, 1 respectively. Here're the pictures: http://www.flickr.com/photos/17157315 at N00/3001184136/ (Perl) http://www.flickr.com/photos/17157315 at N00/3001183926/ (Lisp) http://www.flickr.com/photos/17157315 at N00/3000344665/ (Ruby) http://www.flickr.com/photos/17157315 at N00/3000344479/ (Python) Of course Java took up tons of shelves, as did Microsoft Access. http://www.flickr.com/photos/17157315 at N00/3000343821/ Kirby From da.ajoy at gmail.com Tue Nov 11 02:43:49 2008 From: da.ajoy at gmail.com (Daniel Ajoy) Date: Mon, 10 Nov 2008 20:43:49 -0500 Subject: [Edu-sig] Rich Data Streams Message-ID: ask prog: Where can I find lists of data in useful computer formats? ie. wordlist, world cities, populations, etc. http://www.reddit.com/r/programming/comments/7civs/ask_prog_where_can_i_find_lists_of_data_in_useful/ Daniel From kirby.urner at gmail.com Wed Nov 12 16:15:11 2008 From: kirby.urner at gmail.com (kirby urner) Date: Wed, 12 Nov 2008 07:15:11 -0800 Subject: [Edu-sig] Rich Data Streams In-Reply-To: References: Message-ID: Very useful, thank you. I'll be poking around for rich data sets we might use in Python classes. Kirby On Mon, Nov 10, 2008 at 5:43 PM, Daniel Ajoy wrote: > ask prog: Where can I find lists of data in useful computer formats? ie. wordlist, world cities, populations, etc. > > http://www.reddit.com/r/programming/comments/7civs/ask_prog_where_can_i_find_lists_of_data_in_useful/ > > > Daniel From kirby.urner at gmail.com Sun Nov 16 21:20:28 2008 From: kirby.urner at gmail.com (kirby urner) Date: Sun, 16 Nov 2008 12:20:28 -0800 Subject: [Edu-sig] Reinventing the classroom... Message-ID: So once we agree we want some kind of computer lab, the question is, what kind? They already come in many flavors, one of my favorite being the virtual hosts approach, where all software and lessons for the day come preloaded such that you experience a properly equipped distro right out of the starting gate, MySQL, SQLite, povray, VPython, some x3d browser, everything you need. No time waisted with RPMs or apt-gets or any of that. Of course from the teacher's perspective, it can't be that labor intensive to manually reconfigure each workstation nightly, which is why the virtual hosts model provides check box menu driven type interfaces, not saying I've seen but a few, schools tend to roll their own sometimes. So as Mr. Urner, I say "tomorrow I want Python 3.x", then that's what'll be there, perhaps turning off install options for every 3rd party package with no compatible version yet (smart menu!). Having been doing some work in the private sector around multiple pyprocessing clients talking to a host via Pyro, I'm thinking my ideal classroom lets students practice teamwork across workstations, even if these are virtual, i.e. the actual classroom hardware is more terminal than desktop computer, the racks being elsewhere. It's not like every day we need interprocess communication across hosts, but some days we might, and this classroom makes it easy for the teacher to set that up ahead of time (e.g. by getting pyprocessing installed -- not Standard Library, emulates Threading but with process dynamics under the hood). I realize we're not there yet, except in some of these executive training centers, i.e. the "students" I'm talking about are business professionals with six figure incomes, hoping to stay in the game during the next tech cycle, where there's lots of hiring but precious few with the skills. That's why I think ramping up more in high school makes a lot of sense, wouldn't mind shoving a lot of that "old Europe" approach to mathematics to one side to make room, but that's just in our more radical / surreal encampments, not saying to try this at home. Maybe a Quaker school here and there might be into making a go of it. However, if we play our cards right, this kind of TuxLab type executive training lab could become more part and parcel of the ordinary pre-college experience, as well as college, where wired dorms are already the norm and computer labs abound. Vendors are already gearing up for this market, as mentioned earlier on edu-sig, with products that'll temporarily limit access or even disable the keyboard at the teacher's discretion, not saying every course need confer such godlike powers (including powers to snoop, lurk). If it's a Quaker school, likely students will share sysop responsibilities so there's a clear view against abuses of power all round, whereas (on the other hand) in some classes, e.g. martial arts, students willingly consent to temporary restrictions i.e. having keyboard go dead, flickering terminals -- could be a valuable simulation and/or learning experience, given likely field conditions this crew is training to handle, maybe erratic power supplies, frequent brown outs. Anyway, if you've been tracking my blogs you see my focus is more the Portland coffee shop scene these days, not so much classrooms, where I just use whatever PSU has to offer. These storefronts already host artists, sculptures, photographers, double as art galleries in other words, which these days might as well include a few LCDs, i.e. flatscreens, framed TVs. Nam June Paik and friends, Hell's Kitchen and so on, have already paved the way for TV screen as art vista, so it takes very little work to get these exhibited, with public buy in that these aesthetics are OK. LA has many examples, with Portland playing catch up (Costello's on East Broadway a good example of what I'm talking about, but also Back Space downtown). Screens host "screen savers" done using Python in some capacity, maybe just playing off a DVD at this point (or streaming) no live CPU in this picture. http://www.paikstudios.com/ http://www.costellostravelcaffe.com/photos/#id=interior&num=1 http://www.backspace.bz/index.php Kirby From ccosse at gmail.com Mon Nov 17 03:20:51 2008 From: ccosse at gmail.com (=?ISO-8859-1?Q?Charles_Coss=E9?=) Date: Sun, 16 Nov 2008 19:20:51 -0700 Subject: [Edu-sig] Reinventing the classroom... In-Reply-To: References: Message-ID: Hi Kirby, On Sun, Nov 16, 2008 at 1:20 PM, kirby urner wrote: > So once we agree we want some kind of computer lab, the question is, what kind? > This sentence sounds similar to something I was just writing about...though reading on in your post it's not really talking about the same thing. Nonetheless, I paste the paragraph for your possible enjoyment .... [delete at will] It is fairly obvious that computers can revolutionize education, but now that every school has them, how does one actually realize all the latent potential residing in all those computer labs? I don't believe the details were ever very clear, just the general agreement that computers _could_ improve education. There doesn't seem to be anything fundamental about computers which affects some corresponding fundamental aspect of education, for example, so it's not like it was a revolution that was going to happen automatically. Furthermore, each subject can only be "revolutionized" to an extent which depends on software availability. Keep on keepin' on ... Charlie Cosse From kirby.urner at gmail.com Mon Nov 17 05:03:46 2008 From: kirby.urner at gmail.com (kirby urner) Date: Sun, 16 Nov 2008 20:03:46 -0800 Subject: [Edu-sig] Reinventing the classroom... In-Reply-To: References: Message-ID: Hi Charlie, just Googling up your earlier suggestion to do more with audio books, for commuters etc. I agree there's a sense of "so now what" regarding some of those computer labs, i.e. are these for uploading pictures of the neighborhood, comparing with historic archival shots, for studying local infrastructure (sewers, cable TV nets, electrical grids...), for doing object orient math, for making and watching YouTubes...? So many possibilities, so little time. I think we need more think tanks looking into these questions. I was impressed when in England to discover a London Knowledge Lab already set up to explore precisely such questions, returned home eager to start my Portland Knowledge Lab experiment, flattering by imitating the way I saw it, actually renting an office on 8th & SE Main, later trucked it elsewhere. Part of PKL's purpose is to amass a collection of videos suitable for math/computer lab use, something of a PR effort, got some cool stuff out there e.g. Adrian's fine "breathing torus": http://www.youtube.com/watch?v=tcMCou6CrK4 (C++ and POV-ray) -- still growing the archive. Anyway, just reminiscing, plus looking forward to more discussions about computer lab design, best practices, etc., admitting up front I'm getting limited impressions in my current walk of life, e.g. not much interaction with high schoolers these days, though I might have another lab session at Portland State in the spring, a traditional Windows setup, everything preinstalled. One thing I'll say is we should be having students practice more working in pairs, solo coding on significant projects becoming relatively rare thanks to management embracing some hard won lessons from the field. Pair programming does not have to mean sharing a computer however, we can do it with various combinations of free tools (I've been doing Skype some, like with a coder coven in LA, though mostly it's the usual "over the shoulder" approach, one of us driving, though sometimes we project the shared screen, sit around a table): https://wiki.koumbit.net/Remote_Pair_Programming http://andrzejonsoftware.blogspot.com/2008/02/remote-pair-programming.html etc. Kirby On Sun, Nov 16, 2008 at 6:20 PM, Charles Coss? wrote: > Hi Kirby, > > On Sun, Nov 16, 2008 at 1:20 PM, kirby urner wrote: >> So once we agree we want some kind of computer lab, the question is, what kind? >> > > This sentence sounds similar to something I was just writing > about...though reading on in your post it's not really talking about > the same thing. Nonetheless, I paste the paragraph for your possible > enjoyment .... [delete at will] > > It is fairly obvious that computers can revolutionize education, but > now that every school has them, how does one actually realize all the > latent potential residing in all those computer labs? I don't believe > the details were ever very clear, just the general agreement that > computers _could_ improve education. There doesn't seem to be anything > fundamental about computers which affects some corresponding > fundamental aspect of education, for example, so it's not like it was > a revolution that was going to happen automatically. Furthermore, each > subject can only be "revolutionized" to an extent which depends on > software availability. > > > Keep on keepin' on ... > Charlie Cosse > From kirby.urner at gmail.com Fri Nov 21 20:49:50 2008 From: kirby.urner at gmail.com (kirby urner) Date: Fri, 21 Nov 2008 11:49:50 -0800 Subject: [Edu-sig] Reality checks at OSU... Message-ID: So I got a lot of reality checks visiting OSU in Corvallis yesterday, a campus I've visited in other contexts, but this was my first time to more central campus buildings that I can recall, parked off Jefferson near Monroe, went to special collections in Valley Library for a visit to the Doug Strain reading room (more info in blogs). The big flatscreen in the library shows study carrels (floor plan, like CubeSpace), and which have computers in use, vs. which are sitting idle, with icons by operating system: Windows, OSX, OSX-music, OSX-video, i.e three flavors of Apple, one of Microsoft (Windows icon). Zero Penguins. Gotta bring those in on laptops, use 'em with wifi, which is what I'd planned for (lugged my Dell in Razz (my Subaru)). The icon for your study carrel turns red or pinkish if you've stayed over two hours. I couldn't tell if this was a badge of honor, shame or neutral, as in "I study really hard" vs "I hog more than my share" vs "who the hell knows what it means" (more my attitude). Of course there's no way to tell, from the flatscreen, which of these cubes was running any Python. That'd be kind of invasive (of privacy), though sometimes a sysop might poll processes for anonymous summarizing i.e. OSU would be able to register upticks in Python usage if it wanted, without naming names, kind of like O'Reilly does around book titles (i.e. what's hot and what's not) etc. I liked how these students looked really interested in learning stuff, and very believably that's what they were there for (to actually study -- not everyone in a toga, no evidence of kegs (this was the library, after all)). My daughter Tara (freshman in high school) was impressed, and with more than just the library. She got to hold a Nobel prize even, one of two in the exhibit chamber. In the MVC design I'm working on, significant work gets performed by Python processes with no obligation to update a browser directly, i.e. the visualization process is on a whole different machine potentially, and works against SQL tables directly. A good analogy might be a closed source archive: you have access to the reference desk (as a browser) but the nuts and bolts application, controlled by way of a web interface, isn't directly talking to any web framework, even if Django's a front end. To "control" is to submit to a jobs queue in SQL, not to chat directly with aliens (other Pythons). This way of implementing MVC means the core controller maintainers don't need to know jack about AJAX or talking to web clients. Just make sure all the relevant info gets out there in MySQL or what have you (sqlite for prototyping maybe), and let geeks doing visualizers work independently, in other cities even, no sharing of source code necessary, just table schemas. I'm not coding solo in this picture, have svn credentials is all. There's possibly an open source angle in that a lot of code gets wasted making up for the fact that your average XHTML isn't designed to be searched (as a namespace, it inherits a lot of visualization concerns as attributes), whereas some flavors of XML are designed entirely around making different ontologies more accessible. So a stripped down version of the parser could be released that just assumes a customized, search-friendly playground, where people post their political views, other thoughts, with the very deliberate intent to guide scoring and polling at the XML level. These are politicians or pundits, opinion shapers, not shy about expressing themselves. In other news, Pycon is suggesting Steve Holden and I rent a blimp to flash ads about a Flying Circus event: him and me doing an experimental prototype curriculum of tomorrow as a twosome, me the futurist Portlander, hot off the jet, and he the focused interlocutor, keeping me on task and on target, in terms of preaching relevantly to the actually present choir. Given the somewhat mathy content of the proposal (wanna do some RSA, fractals in PIL, vectors in VPython), the idea of the blimp was to attract local high school teachers from around Chicago, maybe shifting to an evening time slot on the weekend so they might come after work (a generous concession). My response has been to defend the content as private sector relevant even though it's in the guise of a futuristic high school curriculum (drawing from personal experience actually teaching this content to high school aged students here in Portland, as chronicled in this very edu-sig archive). The goal is to impart skills around core Python plus some important libraries and strategies for using 'em (same as most tutorials, so not so unusual, no blimp need apply (sounds expensive!)). Plus this duo workshop leader format might demonstrate a good way for project leaders in all walks of life to manage teams of peer coders i.e. set up some semi-Socratic Bob asking Alice or Alice asking Bob, with Eve the lurking learner, gradually picking up clues as to where the team has been, is headed, starting to more ask better questions as the picture becomes clear (unlike in cryptography, Eve is on our side in this picture, just might not speak our language yet, i.e. Python). Kirby From jurgis at akl.lt Mon Nov 24 19:51:52 2008 From: jurgis at akl.lt (Jurgis Pralgauskis) Date: Mon, 24 Nov 2008 20:51:52 +0200 Subject: [Edu-sig] edu-sig page is orphaned? In-Reply-To: References: <34f4097d0810310534k6658289fy9d95094d8089cc63@mail.gmail.com> <34f4097d0810310620g40f69a64y3df7fae932e65b7b@mail.gmail.com> Message-ID: <34f4097d0811241051r279b007ao987bf9a41fffecc2@mail.gmail.com> Hello again, so who could actually update the edu-sig page now? (as I listed some issues in the beggining of thread) On Fri, Oct 31, 2008 at 3:51 PM, kirby urner wrote: > I am largely responsible for the page as it currently exists, in terms > of content, but threw that together a long time ago by now, around the > time the site went from cvs to svn, and before the face lift wherein > the new twin-snake logo was phased in. > > Since that time, I've suggested we work out a new management setup, > but only in my capacity as one more edu-sig subscriber (I've never > been the listowner here, albiet I've been one of the most prolific > posters to date). > > I would also encourage web savvy educators to do their own edu-sig > type pages, having Python as a focus if that's their area of expertise > i.e. one way to promote change at Python.org would be to not wait, > demo your dream web page in another domain, give the community more > templates to choose from, in terms of where to take it next. > > Walking my own talk, I co-sponsor an Oregon Curriculum Network site > with a "cp4e" page ( http://www.4dsolutions.net/ocn/cp4e.html -- not > just about Python, but mostly is). > > Kirby > 4dsolutions.net > > > On Fri, Oct 31, 2008 at 6:20 AM, Jurgis Pralgauskis wrote: >> Hello, >> >> http://www.python.org/community/sigs/current/edu-sig/ >> in Resources PyBiblio link leads to Error 404 >> IDLE "page" looks also strange (though I understand it doesnt have >> direct dependence from edusig page) >> >> Software Carpentry leads to old page (as I know http://swc.scipy.org/ >> is quite newer) >> >> PyGeo link is also error 404 http://pw1.netcom.com/~ajs/ >> >> in the editors section PyScripter is a must (it is the best opensource >> pyeditor I know on windows for intermediate students) >> - PyScripter http://wiki.python.org/moin/PyScripter >> >> - Crunchy http://crunchy.sourceforge.net/ is very nice project >> >> For more fun programming (somewhere close to PyGame): >> >> - PyGlet http://www.pyglet.org/ >> - MakeBot (http://mail.python.org/pipermail/edu-sig/2008-October/008808.html >> http://stratotools.python-hosting.com/browser/trunk/makebot/) >> >> - no mention of Turtle, why?!! >> now should be ;) >> http://bugs.python.org/issue1513695 >> http://www.geocities.com/jjphoogrp/clixtur.html >> and alongside I know a book, thtat uses turtle a lot >> http://www.briggs.net.nz/log/writing/snake-wrangling-for-kids/ >> >> edusig mailinglist addres could be at the top, as this is main spot of >> activity :) >> >> I hope, that smb, that is responsible for edu-sig page, >> could quite easily integrate (some of) these proposals :) >> >> -- >> Jurgis Pralgauskis >> Don't worry, be happy and make things better ;) >> _______________________________________________ >> Edu-sig mailing list >> Edu-sig at python.org >> http://mail.python.org/mailman/listinfo/edu-sig >> > -- Jurgis Pralgauskis tel: 8-616 77613; jabber: jurgis at akl.lt; skype: dz0rdzas; Don't worry, be happy and make things better ;) From echerlin at gmail.com Tue Nov 25 01:29:25 2008 From: echerlin at gmail.com (Edward Cherlin) Date: Mon, 24 Nov 2008 16:29:25 -0800 Subject: [Edu-sig] Reinventing the classroom... In-Reply-To: References: Message-ID: On Sun, Nov 16, 2008 at 12:20 PM, kirby urner wrote: > So once we agree we want some kind of computer lab, the question is, what kind? That's a pretty strong assumption, and one that the 1-1 computing community would strongly take issue with. I haven't noticed much good from computer labs, while I have seen quite remarkable results reported from One Laptop Per Child deployments in Peru, Ethiopia, and elsewhere, where the children not only have the use of a computer throughout the school day, but take them home to work on homework jointly over the wireless network, and teach their parents about computers and the Internet. The computer lab strikes us as exactly analogous to a writing lab in which children would have access to paper and pencil for an hour a week. -- Silent Thunder (??/???????????????/????????????? ?) is my name And Children are my nation. The Cosmos is my dwelling place, The Truth my destination. http://wiki.sugarlabs.org/go/User:Mokurai From kirby.urner at gmail.com Tue Nov 25 04:02:19 2008 From: kirby.urner at gmail.com (kirby urner) Date: Mon, 24 Nov 2008 19:02:19 -0800 Subject: [Edu-sig] Reinventing the classroom... In-Reply-To: References: Message-ID: On Mon, Nov 24, 2008 at 4:29 PM, Edward Cherlin wrote: > On Sun, Nov 16, 2008 at 12:20 PM, kirby urner wrote: >> So once we agree we want some kind of computer lab, the question is, what kind? > > That's a pretty strong assumption, and one that the 1-1 computing > community would strongly take issue with. Yeah, not saying the labs are the best solution in all circumstances, just we'd have a tough time without 'em here in Portland for example, use 'em daily, often with adults hoping to hold on to those juicy high tech jobs they've got. Some of 'em are like near the top of Fox Tower, used to have Python User Group meetings in one of 'em... Not either/or, is what I'd say. Anyway, the tone was more: if, in your specific venue, you discover you want some kind of computer lab, then... and like that. Love my XO, Kirby From kirby.urner at gmail.com Wed Nov 26 00:34:41 2008 From: kirby.urner at gmail.com (kirby urner) Date: Tue, 25 Nov 2008 15:34:41 -0800 Subject: [Edu-sig] Reinventing the classroom... In-Reply-To: References: Message-ID: On Mon, Nov 24, 2008 at 7:02 PM, kirby urner wrote: > On Mon, Nov 24, 2008 at 4:29 PM, Edward Cherlin wrote: >> On Sun, Nov 16, 2008 at 12:20 PM, kirby urner wrote: >>> So once we agree we want some kind of computer lab, the question is, what kind? >> >> That's a pretty strong assumption, and one that the 1-1 computing >> community would strongly take issue with. << SNIP >> > Not either/or, is what I'd say. > > Anyway, the tone was more: if, in your specific venue, you discover > you want some kind of computer lab, then... > and like that. > > Love my XO, > > Kirby > Walking my talk re the XO this morning, took it to a coffee shop, as I sometimes do, got lots of cooing going among customers, who all fall in love with it immediately. Any guy geek with an XO shouldn't complain about not getting dates -- you must be doing something wrong, don't blame MIT, is all I have to say. Better than owning a puppy! http://worldgame.blogspot.com/2008/11/posting-from-my-xo.html Kirby From jurgis.pralgauskis at gmail.com Thu Nov 27 00:08:16 2008 From: jurgis.pralgauskis at gmail.com (Jurgis Pralgauskis) Date: Thu, 27 Nov 2008 01:08:16 +0200 Subject: [Edu-sig] xturtleDemo hack to mark current executing line In-Reply-To: <34f4097d0811261427s335df8ebuc596111737ec1dc3@mail.gmail.com> References: <34f4097d0811261427s335df8ebuc596111737ec1dc3@mail.gmail.com> Message-ID: <34f4097d0811261508u7462366fuab4f6573e3896eca@mail.gmail.com> Hello, I made a small trick, inspired by rur-ple "Code and Learn" environment see methods my_trace and set_line_number (left the names same as in rur-ple cpu.py) I don't know TkInter, so I just put *** in front of the current line there are a few issues I don't know how to resolve: * if the line is outside the textbox, it isn't seen, as the text is each time rewritten to textbox (there should be some scroll to line function, I hope) * I'd like to have step (inside of the loaded module) functionallity (as in rur-ple), but it is still magic for me... you can find the modified file http://files.akl.lt/users/jurgis/python/xturtle-hacks/ and I paste the essence here for archiving purposes: def my_trace(self, frame, event, arg): #same called in rur-ple cpu.py if event == "line": if self.code_src == "file" and self.module.__name__ in str(frame.f_code): time.sleep(0.5) self.set_line_number(frame.f_lineno) # print "my_trace: event, frame:", event, inspect.getframeinfo(frame) return self.my_trace def set_line_number(self, lineno): # could be called highlight_line lines = map(lambda x: " "+x, self.code.split("\n") ) lineno = lineno - 1 # the real one :) lines[lineno] = "*** " + lines[lineno][4:] self.text.delete("1.0", "end") self.text.insert("1.0", "\n".join(lines) ) # in start_demo(...) sys.settrace(self.my_trace) result = self.module.main() sys.settrace(None) # this should be also called in StopIt -- Jurgis Pralgauskis tel: 8-616 77613; jabber: jurgis at akl.lt; skype: dz0rdzas; Don't worry, be happy and make things better ;) From mpaul213 at gmail.com Fri Nov 28 18:26:34 2008 From: mpaul213 at gmail.com (michel paul) Date: Fri, 28 Nov 2008 09:26:34 -0800 Subject: [Edu-sig] True or False? - A translation is a function. Message-ID: <40ea4eb00811280926h653c3a84uec0f242373c90995@mail.gmail.com> We were talking about translations in math class, and I was going to show the students how the translation notation in their text could TRIVIALLY be turned into Python code, and it occurred to me that maybe I should first find out if they understood that translations already WERE already functions or if functions and translations were two different kinds of things. The results were very interesting. Lots of hesitation and disagreement among the students. Some had no problem at all saying that yes, a translation is a function. Others got stuck on the idea of a vertical translation somehow violating the 'vertical line test', so they thought vertical translations would not be functions. That was what I was suspecting - our curriculum creates a really boxed in sound-byte understanding of what functions are all about. The kids get trained to respond "is a function" when they see a 'U' shaped parabola and to respond "is not a function" when they see a 'C' shaped parabola. However - the important issue is not a vertical line test - the important issue is clearly designating our input and output. A horizontally opening parabola is simply a function in the form x = f(y). They also tend to develop the boxed in notion that functions are things that operate on and produce single scalar values. That's the effect the current curriculum has. The idea that a function can take in ordered pairs or lists seems strange to most students who haven't been exposed to programming. I found that a useful discussion to have, and I thought it might be fun to see what the teachers in the math department believed. So, I went around and asked. I just wanted to see what their honest in the moment response would be. Amazing - I got the same responses as from the students. Some had no problem with the idea. Others did get it right, but with much hesitance. And, several did not get it right. The reasoning on the part of the teachers who were either unsure or incorrect was the same as the students'. It really does seem that our curriculum fosters among both students and teachers an understanding of functions as things that operate on and produce single numbers and as graphs that pass the 'vertical line test'. All the more argument for weaving in a computational thread in the curriculum. - Michel P.S. A couple of weeks ago I gave my first presentation at a math conference. I had a small group in my session, but they really seemed to get it. The title of my presentation was *Fractions are Objects, not Unfinished Division Problems*. It was the first time any of them had seen Python. They loved it. -------------- next part -------------- An HTML attachment was scrubbed... URL: From kirby.urner at gmail.com Fri Nov 28 20:30:21 2008 From: kirby.urner at gmail.com (kirby urner) Date: Fri, 28 Nov 2008 11:30:21 -0800 Subject: [Edu-sig] True or False? - A translation is a function. In-Reply-To: <40ea4eb00811280926h653c3a84uec0f242373c90995@mail.gmail.com> References: <40ea4eb00811280926h653c3a84uec0f242373c90995@mail.gmail.com> Message-ID: My preferred way of talking about translation is in conjunction with rotation and scaling, as like the three major things we might do with shapes that are angle-preserving (not just surface angles, but also central angles). My Python polyhedron class will implement these (rotation, translation and scaling) as three primitive functions, with subclasses holding the data for actual polyhedra, following the design pattern of "abstract class" hold the "meta methods" common to all polys. But what's a Polyhedron then? By this time we have a primitive Vector concept, except I keep those originating from the origin i.e. (0,0,0) in 3D XYZ. My Edge class is the generic "between any two points" line segment concept, and so Polyhedra tend to be encoded as faces (going clockwise or counter, as seen from without), from which Edges are distilled, each defined by a pair of Vectors (using whatever addressing scheme). That's all one curriculum segment, whereas another would be the standard trignonmetric A*cos(B*x) + C, other trig functions substituting for cos. This goes with the Fourier piece, talked about in this archive quite a bit in the early days, what with our 'Who is Fourier?' by LEX Institute & company. I really dislike all the focus on parabolas minus any attention to "dish" as a concept. The reason that's called a "focus" is because you can get your heat vectors or radio vectors all concentrating to that point, lots of calcs as to what's a useful threshold, below which we don't meet the relevance test. Modern K-12 mathematics textbooks barely hint at the engineering aspects, being completely watered down to keep out any "impurities" ( = healthy proteins). Most people mired in the current math tracks rarely get outdoors, have no concept of tool use, can't do anything practical with those skills they get tested on. We waste zero energy apologizing for such nonsense, assure parents and students alike that we're not that poor quality, would never retreat to that level, having experienced the disaserterous consequences of overspecialization first hand. Kirby 2008/11/28 michel paul : > We were talking about translations in math class, and I was going to show > the students how the translation notation in their text could TRIVIALLY be > turned into Python code, and it occurred to me that maybe I should first > find out if they understood that translations already WERE already functions > or if functions and translations were two different kinds of things. > > The results were very interesting. Lots of hesitation and disagreement > among the students. Some had no problem at all saying that yes, a > translation is a function. Others got stuck on the idea of a vertical > translation somehow violating the 'vertical line test', so they thought > vertical translations would not be functions. That was what I was > suspecting - our curriculum creates a really boxed in sound-byte > understanding of what functions are all about. The kids get trained to > respond "is a function" when they see a 'U' shaped parabola and to respond > "is not a function" when they see a 'C' shaped parabola. However - the > important issue is not a vertical line test - the important issue is clearly > designating our input and output. A horizontally opening parabola is simply > a function in the form x = f(y). They also tend to develop the boxed in > notion that functions are things that operate on and produce single scalar > values. That's the effect the current curriculum has. The idea that a > function can take in ordered pairs or lists seems strange to most students > who haven't been exposed to programming. > > I found that a useful discussion to have, and I thought it might be fun to > see what the teachers in the math department believed. So, I went around > and asked. I just wanted to see what their honest in the moment response > would be. > > Amazing - I got the same responses as from the students. Some had no > problem with the idea. Others did get it right, but with much hesitance. > And, several did not get it right. The reasoning on the part of the > teachers who were either unsure or incorrect was the same as the students'. > > It really does seem that our curriculum fosters among both students and > teachers an understanding of functions as things that operate on and produce > single numbers and as graphs that pass the 'vertical line test'. All the > more argument for weaving in a computational thread in the curriculum. > > - Michel > > P.S. A couple of weeks ago I gave my first presentation at a math > conference. I had a small group in my session, but they really seemed to > get it. The title of my presentation was Fractions are Objects, not > Unfinished Division Problems. It was the first time any of them had seen > Python. They loved it. > > _______________________________________________ > Edu-sig mailing list > Edu-sig at python.org > http://mail.python.org/mailman/listinfo/edu-sig > >