> > For example, compare: > > 3 :'10+y'&.^. 2 3 5 > > 10&+&.^. 2 3 5 > > For the explicit version, I think it's clearer as: > > ^ 10 + ^. 2 3 5 > > and from this you get the explicit verb (3 : '^ 10 + ^. y') .
This version loses the insight that there is a "dual": 10&+ under logarithms. ----- Original Message ----- From: 0j1 <[EMAIL PROTECTED]> Date: Friday, September 12, 2008 9:37 Subject: Re: [Jgeneral] Verb Sequences - u@:v and [EMAIL PROTECTED] To: [email protected] > > Raul Miller rauldmiller at gmail.com wrote > > Fri Sep 12 22:18:58 HKT 2008 > > On Fri, Sep 12, 2008 at 10:07 AM, 0j1 <j at > asteroid.demon.co.uk> wrote: > >> It's harder to write tacit code than explicit and harder to > debug it > >> when it doesn't work. > > > > In my opinion, this depends on the length of the code. > (And on your > > background.) > > > > Also, in my opinion, the advantage of tacit code (other than speed) > > is that some manipulations are easier with tacit code than explicit > > code. > > Agreed. > > > For example, compare: > > 3 :'10+y'&.^. 2 3 5 > > 10&+&.^. 2 3 5 > > For the explicit version, I think it's clearer as: > > ^ 10 + ^. 2 3 5 > > and from this you get the explicit verb (3 : '^ 10 + ^. y') . > > > Of course, indirect manipulations are sometimes superior > > 1x10 * 2 3 5 > > > > -- > > Raul ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
