[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Fri, 03 Feb 2006 08:52:07 -0800
Gnus/5.110004 (No Gnus v0.4) Emacs/21.4 (gnu/linux)
Jason Stover <address@hidden> writes:
> On Thu, Feb 02, 2006 at 04:20:20PM -0800, Ben Pfaff wrote:
>> Jason Stover <address@hidden> writes:
>> I seem to recall that when we met at Stanford this summer, it was
>> sort of a "side goal" of yours to write statistical libraries
>> that were useful both inside and outside PSPP. If that's still
>> the case--I do like the idea--then I'd encourage you to take a
>> stab at it.
> That's still the case. I have been taking a stab at it, though
> not as much lately as I would like because of work.
Makes sense--glad to hear it.
>> Is there a reason to create a separate "statlib" for statistical
>> libraries? I don't know of a reason that the existing "lib"
>> wouldn't do. The goal of the division of source code in PSPP
>> between "lib" and "src" is just a hint that code in "lib" is not
>> PSPP-specific, whereas code in "src" is.
> I'm not attached to the name 'statlib' or any other specific name.
> If 'lib' is meant to encompass anything that doesn't depend strictly
> on PSPP, then 'lib' sounds okay.
That's the way I'd like to interpret it. Perhaps we could add a
README in the lib directory that states this, so that it's clear
to people other than me :-)
"Implementation details are beyond the scope of the Java virtual
machine specification. One should not assume that every virtual
machine implementation contains a giant squid."
--"Mr. Bunny's Big Cup o' Java"