Ang. Re: [Gretl-devel] Bug in help file

[Riccardo (Jack) Lucchetti]

On Thu, 13 Sep 2007, Sven Schreiber wrote:

> andreas.rosenblad at ltv.se schrieb:
>> Why not just change the name of the gamma function from "gamma" to
>> "gammafunction" or "gammaf" or something like that?
>>
>
> Because it would break existing code.

I agree with a few extra remarks.

The gamma function is not terribly useful, by itself. Its logarithm is. 
One possibility would be to remove the gamma() function altogether and 
have users use exp(lngamma(x)) if they need to. Of course this would be 
quite expensive computationally, but in my experience it is extremely rare 
to use the gamma function instead of its logarithm in iterative 
computations. Similar considerations hold for the beta function.

The backward compatibility issue could be mitigated by officially 
deprecating the use of the gamma function, for instance producing a 
warning each time it is used, with an eye to removing it in a year's time 
or so.

As for the more general point about namespaces: a possible source of 
confusion could come from using functions as function parameters. In the 
future, it's not unlikely that we may add a "function" type to the 
possible parameter type that may be fed into user-written functions. 
For instance, numerical integration. Imagine you have something like

function numint(function f, scalar a, scalar b)
   scalar dx = 0.01*(b-a)
   scalar ret = 0
   scalar x0 = a + 0.5*dx
   loop 100 --quiet
     ret += f(x0) * dx
     x0 += dx
   end loop
   return scalar ret
end function

Thanks to Allin's foresight, the fact that user-defined functions in gretl 
are "strongly typed" does help quite a lot here, but still I imagine that 
the code to handle a case like this would be quite complex.


Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche

r.lucchetti at univpm.it
http://www.econ.univpm.it/lucchetti