On 2/13/07, Peter Bowyer <[EMAIL PROTECTED]> wrote:
At 06:46 13/02/2007, kirby urner wrote: >Flash forward: have you seen O'Reilly's 'Head First' series and >its use of icons, sidebars, jokes, diagrams, different type faces, >more icons? Way more "right brained" than traditional CS books, >by a long shot, but just as technical and deep (into Java mostly). >We've asked Tim about a "Head First" about Python, but the >word back, at least then, was we had too small a footprint as >a nation (he has these publisher maps he projects at OSCON) >to merit such a sophisticated and expensive undertaking. We >all still dream of it though. Is there anything to stop us/someone copying this style of book, and producing one for Python? Peter
No nothing, and I think 'Head First' is a trend setter. But of course we of the OSCON tribe would feel glad if the O'Reilly name were on the cover, means we still matter on maps that we care about. A cheap rip off wouldn't feel so good. Not that those are the only two alternatives. Multimedia publishing is no one's monopoly, and many publishers besides O'Reilly already have promising track records with Pythonic titles (though I don't really like the ones that overpad with the very same XML poopka, regardless of language, just to thicken the books, make 'em appear meaty. 'Python in a Nutshell' is at the other end of the spectrum, somewhat thick, but spare, lean and clean. It's not a teaching text though, so much as a reference). Myself, I'm focusing on the anime we want around Python, like a cartoon snake slithering into view showing off __rib__ syntax, going in to split screen views with source code in a stepper, coupled with animal behavior in some 4D++ environment (e.g. Panda 3D). On the one hand, source code, on the other hand, Sims. What we'll be sharing with adults, not just children, under the CP4E banner. To this end, I've voluntarily sunk a lot of my after tax time and energy into startup think tanks, such as Wanderers and the Portland Knowledge Lab (the latter named after a London version I visited last April, stole the idea from (hi Phillip)). Another thread we've shared here on edu-sig is around "rich data structures." I don't know how much need you'd have for 100 nouns and/or verbs in a lookup table, columns being the different human languages, but I'd think there'd be similar wheels you'd not want to have to reinvent, over and over. Do you use any SQL? College professors, banding together on open source principles (already deeply engrained in the culture, as scholarship ethics), oughta be able to divvy the labor and build up a relevant stash of Pythonic data sets pronto, in whatever the science and/or humanity. Just having the planets, with their distances from the Sun, relative radii, in a freely downloadable Python dictionary of tuples say, would be a boon to middle schools across the land. That data could be bundled with such as my orbits.py, to generate VPython animations. http://www.4dsolutions.net/cgi-bin/py2html.cgi?script=/ocn/python/orbits.py A final relevant thread I can think of, is 'suMerian', a mnemonic for how the antiquities have made it still cool to learn so-called "dead languages" (a basis for promotion). Computer science is a relatively new discipline, compared to teaching ancient Greek or Latin or whatever, but has this emerging challenge of either needing to groom future generations of so-called "dead language" programmers, OR try to tell industry why it should pay to have perfectly good source code rewritten, simply because no one teaches M any more (hence the capital M in 'suMerian'). Put another way, there's still a market for FORTRAN coders, even FORTRAN compiler optimizers. I've schmoozed with such people at OSCON, know they exist in the flesh. Kirby
_______________________________________________ Edu-sig mailing list [email protected] http://mail.python.org/mailman/listinfo/edu-sig
