On Sun, Dec 27, 2009 at 9:08 PM, mhampton <[email protected]> wrote:
> I'd like to object to the policy you are proposing.  I would like to
> first emphasize that I completely support the goal of having functions
> include INPUT and OUTPUT blocks.  But the policy you are proposing is
> far too rigid.
>
> As one example, imagine that a Sage user finds a documentation error.
> The user wants to file a bug report.  They are then encouraged to get
> a trac account and post a patch.  Imagine that they do so.  Then a
> reviewer flags the patch as "needs work" because the function in
> question doesn't have INPUT/OUTPUT blocks.  I think this would be very
> discouraging.
>
> Perhaps you meant your policy to only apply to new code, or to
> functions (rather than modules) that have been revised.  I think it is
> worthwhile for you to clarify exactly what you mean.
>
> In general I would like to point out that such policies can make it
> unlikely that new people will become Sage developers.  It is a
> delicate balance between encouraging new effort and maintaining
> quality.  At the moment I think the current policies are quite good,
> but if they became more rigid it would turn new people off.
>
> -

We aren't writing enough INPUT/OUTPUT blocks to describe the input and
output of functions.   I want to encourage this very, very strongly
since I think it will make Sage much more usable, and many other
people have requested it.  I don't care about "policies" with respect
to this particular discussion at this point.    However, if you are
refereeing some code and you see that the INPUT/OUTPUT blocks are
generally lacking, I hope you will feel confident in complaining about
this.  I certainly will.

 -- William

-- 
To post to this group, send an email to [email protected]
To unsubscribe from this group, send an email to 
[email protected]
For more options, visit this group at http://groups.google.com/group/sage-devel
URL: http://www.sagemath.org

Reply via email to