MathGroup Archive 2001

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

Search the Archive

Re: Placeholders in matrix notation

  • To: mathgroup at smc.vnet.net
  • Subject: [mg29956] Re: Placeholders in matrix notation
  • From: dooglefish at aol.communicate (D.F.)
  • Date: Thu, 19 Jul 2001 03:57:27 -0400 (EDT)
  • References: <9j39rr$ikv$1@smc.vnet.net>
  • Sender: owner-wri-mathgroup at wolfram.com

>Subject: [mg29956] Re:  Placeholders in matrix notation
>From: "Hugh Philipp" hph at com.dtu.dk 
To: mathgroup at smc.vnet.net
>Date: 7/18/01 2:20 AM Eastern Daylight Time
>Message-id: <9j39rr$ikv$1 at smc.vnet.net>
>
>I've had this problem before and I have used at least a dozen clumsy work
>arounds for it.
>
>Here is one:
>
>ff ={D[#[[1]],x],0}&;
>
>ff@{f[x],g[x]}
>
>Of course, I don't really like this method because you have to do the matrix
>operations explicitly in the function - designating the elements.
>
>
>The other way of doing it goes something like this:
>
>Unprotect[Times]
>
>d1 f_:=D[f,x]
>
>Protect[Times]
>
>{d1,0,d1}.{f[x],g[x],h[x]}
>
>I don't know if the gurus here think this is O.K. - I'm sure there will be
>comments.
>
>Notice to that:
>
>d1 d1 f[x]
>
>evaluates to f''[x]
>
>...a nice 'feature'. - but BE CAREFUL becaue d1 d1 evaluates to 0.  So -
>there might be some other hidden 'features' with this approach.
>
>For example, the vector:  {d1 d1,d1} evaluates to {0,d1}
>
>Hope this helps,
>
>Hugh.


Yes, it works very well. Thank you.


\!\(\*
  RowBox[{\(Unprotect[Times]\), "\n", \(d1\ f_ := D[f, x]\), 
    "\n", \(d2\ f_ := D[f, y]\), "\n", \(d3\ f_ := D[f, x, y]\), 
    "\n", \(Protect[Times]\), "\n", 
    RowBox[{"MatrixForm", "[", 
      RowBox[{
        RowBox[{"(", GridBox[{
              {"d1", "0"},
              {"0", "d2"},
              {"0", "d3"}
              }], ")"}], ".", 
        RowBox[{"(", GridBox[{
              {\(f1[x, y]\), \(f2[x, y]\), \(f3[x, y]\)},
              {\(g1[x, y]\), \(g2[x, y]\), \(g3[x, y]\)}
              }], ")"}]}], "]"}]}]\)


-Doug


  • Prev by Date: Re: Thickness Isn't Thickness
  • Next by Date: Re: Thickness Isn't Thickness
  • Previous by thread: Re: Re: Placeholders in matrix notation
  • Next by thread: A question on machine numbers