Re: 4th "Atelier PARI/GP" (january 2014)

[Joerg Arndt <arndt@jjj.de>]

* John Jones <jj@asu.edu> [Aug 08. 2013 19:50]:
> Here are some thoughts on areas to improve pari/gp.
> 
> 
>    - allow implicit multiplication in the gp parser (i.e., accept 3x^2+5x-3)

Please NO!   Let's not join the club of Mathematica sloppiness.

>    - Galois groups of higher degree polynomials (I see it is already a wish
>    list item in the bug tracker, so this is another vote)
>    - computation of resolvents ala magma (give a polynomial, compute its
>    Galois group, then specify the subgroup, or specify a degree to get
>    resolvents for all subgroups with the given index)
>    - Impliment
>    http://www.sciencedirect.com/science/article/pii/S0022314X08001121 for
>    bnfcertify
> 
> John
> 
> [...]

And I chime in with

- call by reference

- vectors/matrices with selectable range(s) of indices

- deprecating certain function (log, sqrt) from working
  element-wise with matrices (bad, bad, bad!).

Best,  jj