Re: University License fees are short sighted of Wolfram Research

From: Dave (nospam_at_nowhere.com)
Date: 03/21/05

  • Next message: Paul Abbott: "Re: Factoring 'x^2-a*x-4'polynomial using freeware/shareware CAS" 
    Date: Mon, 21 Mar 2005 23:59:36 +0000

    carlos@colorado.edu wrote:

    >>>It is a relatively small market, and "several million"
    >>>is probably inflated.
    >>
    >>I suspect that too.
    >
    >
    > How many toolboxes does your campus Matlab license cover?

    I can't give you an exact figure, as it varies from toolbox to toolbox,
    but I'd guess 500-1000.

    The Matlab licence allows home usage, but the procurement department
    administering that said

    "Yes it does, but you are still going to have to buy a license if you
    want to use it". So I paid £100 (~$180) of my own money to run a copy at
    home on my Sun - since I wanted it on a Sun, I could not get the CDs or
    license key any other way.

    > As for users, here are rough guesses for my campus - I get
    > these from serving in college-level IT committees that
    > are informed about licenses, etc:
    >
    > Matlab: 1500-2500 students/yr, mostly engineering instruction
    > Mathematica: 300-500 students/yr, mostly applied math &
    > physics instruction
    > Maple: unknown - not used in courses except as matlab toolbox
    > Serious users (research) 200 Matlab, 50 Mathematica
    >
    > Total campus students: 29000. Engineering: 3500. Sciences:
    > (math, physics, chemistry ...): 6000
    >
    > There are 100-150 similar campuses across the US so for the
    > US-total append 2 zeros. Doesnt reach to 1M users though.
    > I agree with Professor Fateman that the market is small, but
    > it is slowly growing.
    >

    Whilst this is slightly off-topic, I'm interested in how other
    universities manage site licenses for software. Where I am the process
    seems to be.

    A) The procurement (the most useless department of all) buys the site
    license. Thankfully the Mathematica license is *not* done by procurement
      any more (they made a real mess of it) but by the computer science
    department.

    But I gather it will go back to procurement's hands (if renewed), as CS
    are making a loss on it, which they don't want - procurement are
    expected to make a loss. (Something like that anyway).

    B) The procurement (or CS in the case of Mathematica) department then
    issues CD's and a "license certificate" (nothing more than a bit of A4
    paper produced internally) on the receipt of a "license fee". A license
    for a whole department will cost more than a license for one user.

    This "license fee" we pay to CS or procurement is set to recover the
    cost of the site license paid to Wolfram Research, or whoever produces
    the software.

    This means only people that use the software pay for it. If you don't
    pay, you don't get issued the disks, passcodes or a bit of A4 paper that
    says you can use it.

    Hence we have two equations.

    cost_per_user=license_fee/number_of_users (1)
    number_of_users = f(cost, how general purpose it is, etc) (2)

    I guess the English Department would not be keen to contribute towards
    the Mathematica site license if they never use it, no more than my
    department would want to contribute to the cost of some software for
    inorganic chemistry, since it is not something we would use.

    However, this leaves a lot of "gray areas" where someone might well use
    some software (like Mathematica) if it is readily available to them, but
    since it is not, they don't use it. Many in my department did use
    Mathematica seriously, but no longer do because of its cost.

    Since we have a Mathematica site license, in theory anyone can use it
    and I'm sure Wolfram Research would like that. But in practice it is
    available to few, as the license fee we pay internally is too high. This
    has to be high to cover the cost paid to Wolfram Research and the fact
    there are few users.

    It's difficult to see a solution to this one.

    PS
    Just to illustrate how useless our procurement department were with
    Mathematica, I found out by accident one day from a Wolfram Employee
    that our site license allowed use at home. Nobody had ever told us this.
    So I asked for a SPARC license. Originally Wolfram Research said a Sun
    was not a home computer, but did this as a one-off, which I appreciated.

    In fact, our current site license is much more restrictive on home
    usage, as the computer needs to be owned and managed by the university.

  • Next message: Paul Abbott: "Re: Factoring 'x^2-a*x-4'polynomial using freeware/shareware CAS"

    Relevant Pages