[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: NCDF functions
From: |
Ben Pfaff |
Subject: |
Re: NCDF functions |
Date: |
Fri, 2 May 2014 09:53:34 -0700 |
On Fri, May 2, 2014 at 9:43 AM, John Darrington
<address@hidden> wrote:
> On Fri, May 02, 2014 at 07:44:52AM -0700, Ben Pfaff wrote:
> On Fri, May 2, 2014 at 7:27 AM, Stephan Weibelzahl <address@hidden>
> wrote:
> > I'm trying to calculate non-central distributions using the function
> >
> > COMPUTE a = NCDF.T(0.2, 3, 1).
> > EXECUTE.
> >
> > as documented in the manual. However, I get an error message that the
> > function has not been implemented. The central distribution function
> works
> > fine. Am I doing something wrong?
>
> No. As the message says, NCDF.T has not been implemented.
>
> I think we have had similar questions before about this error message. I
> wonder
> if we should reword it? It is a typical case of confusion which arises from
> use
> of the passive voice in English.
>
> "... has not been implemented" implies that something should have been
> implemented.
> That raises the question: "by whom?". I think some users have misinterpreted
> this message to mean that they (the user) should have done something.
>
>
> Perhaps we should reword it to something like "This function is not available
> in
> this version of PSPP". I am reluctant to use the word "supported" like so
> many
> proprietary software vendors seem to throw around, because in PSPP (as in most
> proprietary software) absolutely *nothing* is "supported" unless one has taken
> out an explicit support contract.
This sounds like a good idea.
> Also, perhaps we should update the manual. It is incorrect when the manual
> says
> "The following continuous distributions are available", when this one is not.
I never expected it to take so long to get these functions implemented.
I agree that we should update the manual.