MathGroup Archive 1998

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

Search the Archive

Re: A "singular" equation

  • To: mathgroup at smc.vnet.net
  • Subject: [mg14815] Re: A "singular" equation
  • From: "W. K. Bertram" <wkb at ansto.gov.au>
  • Date: Wed, 18 Nov 1998 01:29:19 -0500
  • Organization: Australian Nuclear Science and Technology Organisation
  • References: <72bf2q$k3l@smc.vnet.net> <72e2ng$oi0@smc.vnet.net> <72l670$g1k$2@dragonfly.wolfram.com>
  • Sender: owner-wri-mathgroup at wolfram.com

Matthias Weber wrote:

> In article <72e2ng$oi0 at smc.vnet.net>, "W. K. Bertram" <wkb at ansto.gov.au>
> wrote:
>
> > David Withoff wrote:
> >
> > > This worked ok for me.  Check with Wolfram Research tech support. Maybe
> > > you have a corrupted/old/etc copy of Mathematica.
> > >
> > > % math
> > > Mathematica 3.0 for Solaris
> > > Copyright 1988-97 Wolfram Research, Inc.
> > >  -- Terminal graphics initialized --
> > >
> > > In[1]:= DSolve[(1-x^2)y'[x]^2==(1-y[x]^2),y[x],x]
> > >
> > >                                  2          2              2      2
> > >                   x - Sqrt[-1 + x ] + x C[1]  + Sqrt[-1 + x ] C[1]
> > > Out[1]= {{y[x] -> -------------------------------------------------}}
> > >                                        2 C[1]
> > >
> > >
> >
> > Dear David,
> >
> > I have exactly the same problem on my system as the previous writer. I'm
> > running Mathematica version 3.0.0 on a 200Mhz Pentium Pro under Windows
> > 95. When I try to execute the above command I get a "This program has
> > performed an illegal operation ..." message and the notebook freezes
> > up. Is this a feature of this particular verion of  Mathematica? If so,
> > maybe Wolfram might consider offering its affected customers an upgrade
> > to a more recent version  at a much better rate than the A$700 - A$800
> > price that was quoted to me for an upgrade from version 3.0.0 to 3.0.1.
> >
> > Cheers,
> >    Bill Bertram
>
> I agree with Bill. However, Mathematica 3.0.0 has quite a lot bugs like
> this (just browse through the newsgroup archives), and it if Wolfram
> would be willing to fix them all for free, this would keep them busy
> for years without payment...

 Of course bugs are  inevitable in complicated software however, but
when these are reported and fixed up I think  the client is entitled to
receive these fixes without extra payment  (Remember the (in)famous
Pentium bug and INTEL's initial refusal to provide replacements to all
but a few selected users).

As far as cost to Wolfram is concerned, other companies seem to have
overcome this problem quite well.  Companies such as Borland with its
various compilers, Lahey with its Fortran compilers, and many others,
fix up reported bugs in their software by periodically providing their
clients with  patches which can be down loaded free from their web
sites. I can't see why such an arrangement wouldn't work for Wolfram.

Cheers,
    Bill




  • Prev by Date: Re: Re: using Upset for defining positive real values (Re: Can I get ComplexExpand to really work?)
  • Next by Date: Networking notebooks
  • Previous by thread: Re: A "singular" equation
  • Next by thread: ODEs and phase portraits