[Edu-sig] poking some dying logs...

[kirby urner]

On Sun, Aug 2, 2009 at 11:06 AM, Edward Cherlin<echerlin at gmail.com> wrote:
> On Fri, Jul 31, 2009 at 4:36 PM, kirby urner<kirby.urner at gmail.com> wrote:
>> I'd like to make another plug for including this title on the edu-sig home page:
>>
>> http://www.skylit.com/mathandpython.html
>>
>> Ian thought it was too much a hybrid of CS and math, not an elegant
>> amalgamation, though I don't have has remarks in front of me at the
>> moment.  Steve was gonna get back to us.  Andre thought he might work
>> it onto the page...
>
> I like the concept but not the execution. The student doesn't find out
> what properties of various data structures and mathematical objects
> are fundamental. There is too much of the old style of telling
> students what to learn, and neither explaining why nor allowing
> students to discover. I find it annoying that the book gives complex
> number examples, but shies away from actually using complex
> arithmetic. Far more CS could be introduced at the level of the math
> being used.
>
> The book uses Python, but none of the very capable math software
> available beyond graphing calculators. I prefer Ken Iverson's
> approach, in which he taught how to write programs to do algebraic
> manipulations and symbolic differentiation.
>
> Does anybody know these authors? Can we engage them in a process to
> improve what they have done?
>
>                          Maria Litvin
> Phillips Academy, Andover, Massachusetts
>                          Gary Litvin
>                    Skylight Software, Inc.
>

Thanks for you feedback Edward.

I have an all day meeting Aug 7 focusing on computer science teaching
issues, hosted by the same lobbying group that showed me this book in
the first place, have the complete text in PDF.

My own response was the book wants to be taken seriously according to
various "math text" criteria associated with 1900s curriculum writing,
somewhat strict and formal, no silly pictures, very not like the new
O'Reilly books.  An old kind of pedagogy and, as such, a "specimen".

However, there's useful stuff here and I'd like my students to go
through it.  I think the strategy might be to come to said Litvins'
text with higher level Python skills that the book demands, so that
even if the approach seems somewhat austere and intimidating, the
Python skills themselves aren't all that high, something we get beyond
in the Sandes book for example.

So I imagine a teacher who is properly respectful of mathematical
content, but doesn't discourage students from making snarky remarks
about the level of Python, which everyone in this class supersedes, is
far better at than this book requires.  The "math" may be high level,
but the Python is middle-grade, yellow belt, even white belt.

I'm thinking teachers themselves need to be empowered to mix a good
mix for their kids, are the alchemists with final authority, and
putting all our eggs in any one textbook, or two, or three, or twenty,
is not a good idea.  The stuff I've published and made available for
free at my web site is both higher level Python and higher level math,
one might argue, and really anyone technically literate is in a
position to publish a portfolio of such writings.  We aren't waiting
for the "authoritative set of textbooks", though we welcome more PDFs
coming online, can always use more.

That being said, many teachers *are* waiting for some "next big
textbook" to come down the pipe.  I'm not saying every school is in
"launch mode" in 2009.  But a lot of us are.  We have all the tools we
need, all the dots connected, plenty of teaching talent chomping at
the bit, not waiting for anything, think starting the race is long
overdue (oh wait, the gun fired years ago, we've been running all this
time.... sometimes I forget I'm a race horse, pretty fast one, duh).

Kirby

>
>> That was all months ago by now, so it make sense to raise the issue
>> again, as the title does break new ground in some ways, has claim to
>> being a math teaching book, yet uses a computer language (one most of
>> us know).
>>
>> 'Concrete Mathematics' and 'The Art of Computer Programming' are both
>> math books of course, amenable to a "through programming" approach.
>> Jsoftware folks implemented the former in J, whereas the latter is in
>> MMX already.
>>
>> Another hot button issue in Portland these days is whether families
>> have the right to demand a PDF version of any assigned textbook,
>> versus a hardcopy edition.  We have lots of tree huggers around here,
>> worried about "green" and unsustainability.  To quote one of my
>> colleagues (from her blog):
>>
>> "We need the text book companies to print thousands of copies of new
>> textbooks every year, not so the authors can make money, though they
>> make a little, but so the companies can make money... Do some central
>> planning, and if the government can't do that without going through
>> corporations, then it is time to [do it ourselves]".
>>
>> Anyway, just wanted to re-raise that as well.
>>
>> I mostly do my computer / technical reading on Safari, have no problem
>> with recycling already printed books, have no problem with small press
>> runs.  But I can see where truck loads of spanking new 400 page math
>> books, hot off the press, none containing any computer programming to
>> speak of, let alone Mites, Sytes or Kites (honeycomb stuff, important
>> to gnu-bees), would provoke a crisis in conscience for our more
>> ethical.
>>
>> This is the kind of thing 15 year olds talk about.  They're suspicious
>> of adults who can't follow their logic (about saving trees),
>> undermines adult authority to not have a response.  So do we all favor
>> an "opt out" option for hard copy textbooks?  Say aye?  Say nay?
>>
>> Kirby
>> _______________________________________________
>> Edu-sig mailing list
>> Edu-sig at python.org
>> http://mail.python.org/mailman/listinfo/edu-sig
>>
>
>
>
> --
> Silent Thunder (默雷/धर्ममेघशब्दगर्ज/دھرممیگھشبدگر ج) is my name
> And Children are my nation.
> The Cosmos is my dwelling place, The Truth my destination.
> http://earthtreasury.org/worknet (Edward Mokurai Cherlin)
>