MathGroup Archive 2013

[Date Index] [Thread Index] [Author Index]

Search the Archive

more problems in Warsaw plan? Re: Work on Basic Mathematica Stephen!

  • To: mathgroup at smc.vnet.net
  • Subject: [mg130928] more problems in Warsaw plan? Re: Work on Basic Mathematica Stephen!
  • From: Richard Fateman <fateman at cs.berkeley.edu>
  • Date: Sun, 26 May 2013 05:05:40 -0400 (EDT)
  • Delivered-to: l-mathgroup@mail-archive0.wolfram.com
  • Delivered-to: l-mathgroup@wolfram.com
  • Delivered-to: mathgroup-outx@smc.vnet.net
  • Delivered-to: mathgroup-newsendx@smc.vnet.net
  • References: <kmngb2$3rv$1@smc.vnet.net> <20130519095011.606CD6A14@smc.vnet.net> <28778561.83848.1369203793332.JavaMail.root@m06> <000101ce56f8$aa8600f0$ff9202d0$@comcast.net> <20130523081057.F226E6AF2@smc.vnet.net> <B715FCFB-8648-4841-865E-23EFDC1ED124@math.umass.edu> <20130524102717.A65106B02@smc.vnet.net> <knq0k1$59u$1@smc.vnet.net>

On 5/25/2013 2:34 AM, Andrzej Kozlowski wrote:
<snip>

  (except perhaps one guy who complained that Mathematica was not
> open source but we quickly stifled him).
Really? Easily?
See
http://www.zdnet.com/open-sources-fortunes-in-poland-could-be-about-to-change-thanks-to-education-7000008611/

>
> An even greater heresy: the course also is going to involve some
> knowledge of Wolfram Alpha: in fact it is entitled "Mathematics with
> Mathematica and Wolfram Alpha". (To RJF - no, we are not going to teach
> anything about any other CAS. If anyone likes to use something other
> than Mathematica they can learn it itself, but won't get any credit for
> it).

Since such a variety of mathematical computations are included in
Mathematica, I would be curious as to what parts of Mathematics are 
selected. Will the course mention how Mathematica is implemented, or
will it be used as a "black box"?

  Actually, our course is going to be a "supervised, self-study and
> Internet based course". In other words, we (the project authors, who are
> getting paid for designing and realising it) will prepare study
> materials (scripts, videos, problems in different disciplines that can
> be solved with the help of Mathematica or Wolfram|Alpha, tests, etc. but
> there won't be any lectures.

Sounds like a good idea. After the material is constructed, you will
presumably be able to drop your involvement and let students interact
with the material without you. And you can go do something else. You
could even be fired!

  There will however be tests and students
> will be able to get credit for passing them - otherwise, of course, it
> would be hard to get them interested.

An interesting problem that comes up with MOOCs  is how to assign credit
for people taking courses on the internet, remotely. Would you expect
students to take exams in person?  Would they have access to a computer?
Would the computer be on the internet?  Would they cheat?
Just asking.

>
> A large part of the grant is for buying a fairly large number of copies
> of Mathematica (in addition to the ones that are already here).
A shame.

>
> Our motives for designing this project are quite selfish (and I don't
> only mean the fact that we are going to be paid for this). There are
> several of us who want to use Mathematica in our courses and it is a
> pain to have to waste time on teaching students the basics of
> Mathematica.

So this course on Mathematica is a required prerequisite for these 
courses?  Or if they need only the basics, why would you need to spend
any time on it in your course?

Soon there will be several new courses available to math
> and science students in which Mathematica will be used. One of them is a
> new course on the differential geometry of curves and surfaces, which
> will be offered in addition to the already existing one which does not
> use any computer software.

  Will students in the two versions take the same exams?  Could you 
compare the results?


  (Students will be able to take both
> sequentially - they are going to be sufficiently different for this to
> make sense).

I would be surprised if there was so much demand for differential 
geometry courses.  How many students are we talking about?  10? 100? 1000?

  There is also a course that I have taught in the past, for
> which there has been a big demand among students of finance - on
> modelling jump processes.


Contrary to the strange ideas RJF got from my
> earlier post, this is not something that statistics packages have been
> doing for years, actually it involves things that are pretty new (so
> much that there are still unresolved problems with some of the
> algorithms) and, in fact, I am proud to admit to having played a certain
> role in this implementation in Mathematica.
>
See
http://cran.r-project.org/web/packages/EMJumpDiffusion/EMJumpDiffusion.pdf

or
http://www.proba.jussieu.fr/pageperso/tankov/tankov_voltchkova.pdf

To see what statistics packages have been doing for years.

If Mathematica is so great with this stuff, why is there a new feature 
of a link to R?

My take on R is that it is more or less the lingua franca of statistics.
I am not a statistician by training, and I have not used R.


> We actually have quite a few people here who although they themselves
> don't use Mathematica or CAS systems at all, have shown a lot of
> interest in having students learn such things.

Can you be more quantitative?  Quite a few = 5? 10? people = faculty? 
staff? graduate students?

  The main interest is, in
> fact, not so much in symbolic computation but in graphics used to
> illustrate various phenomena in topology, complex analysis etc.

This is quite revelatory.  If the big selling point for Mathematica is
graphics, what other graphics packages are available? To me the graphics
is not central to symbolic computation, just a user-friendly 
easy-to-demo component added on.  I think of symbolic computation as 
fundamentally the representation and manipulation of non-numeric 
"formula" data. But some people view graphics = symbolic.

  Although
> one might expect that there should already be lots of such illustrations
> available on the Internet (or even on Wolfram's Demonstrations Site"
> this does not seem to be actually the case.

Google finds 205,000 pages mentioning gnuplot examples. There
are many other programs for scientific visualization, some free.

>
> Of course, all of this is at this time only in its infancy. About a year
> from now I should be able to say something about the success or failure
> of this project. At this time the thing that worries me the most is the
> amount of work I foresee in trying to realise all we have promised in
> the proposal.

Is any part of the proposal available online in English so as to see 
what you actually promised?




  • Prev by Date: Re: How eigenvectors are normalized
  • Next by Date: Re: Through[Divide[f1, f2][x]]
  • Previous by thread: Re: Warsaw Univ. course, was Re: Work on Basic Mathematica Stephen!
  • Next by thread: Warsaw Univ. course, was Re: Work on Basic Mathematica Stephen!