From: Albert_Rich@msn.com   
      
   On Sunday, November 2, 2014 7:07:00 AM UTC-10, clicl...@freenet.de wrote:   
   > Albert Rich schrieb:   
   > >   
   > > It's always good to hear from a "big fan of Rubi"!   
   > >   
   > > Inspired in part by your post on September 7th, improvements made to   
   > > the just released Rubi 4.6 enable it to integrate   
   > > log(2+x)^3*log(3+x)*x^3 wrt x.  The result is comparable in size to   
   > > the Mathematica's and contains 6 polylogarithm functions.   
   > >   
   > > More generally Rubi 4.6 can integrate, at least in principle, any   
   > > expression of the form   
   > >   
   > >     Px F(g+h x)^m (a+b log(c (d (e+f x)^p)^q))   
   > >   
   > > wrt x where Px is a polynomial, m is a non-negative integer and F is   
   > > log, arcsin, arccos, arctan, arccot, or their inverse hyperbolic   
   > > counterparts.   
   > >   
   > > However, like the other systems tested, Rubi 4.6 is unable to   
   > > integrate log(2+x)^3*log(3+x)^2*x^3 wrt x.  I would guess this   
   > > expressions is not integrable in terms of the elementary and   
   > > polylogarithm functions(?).   
   > >   
   >   
   > I would guess so too. Do you find it too difficult to reduce the number   
   > of polylogarithms, or just somehow inadvisable for a rule based system?   
   >   
   > Martin.   
      
   I do not understand what you mean by "reduce the number of polylogarithms" and   
   in what context.   
      
   Albert   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)