[FRPythoneers] Doctest

Uche Ogbuji uche.ogbuji at fourthought.com
Mon Jul 30 16:59:08 MDT 2001


Evelyn Mitchell wrote:
> 
> On Mon, Jul 30, 2001 at 04:22:47PM -0600, Uche Ogbuji wrote:
> > I'd also like to keep away from formula.  "ad hominem" is largely in the
> > eye of the beholder.  I think the above is a pretty straightforward
> > response to
> >
> > "Thanks for the insight into how your shop operates ... yet another
> > example of:  there is always time to do it over, but never time to do it
> > right."
> >
> > Which as I read it is a bald insult.
> >
> > This would have been clear if you had quoted the applicable context
> > rather than my remark alone.
> >
> 
> I'm sorry your feelings have been hurt on several occasions during
> this thread.

Well, I should note that I'm not exactly feeling victimized or
anything.  I've been in discussion of operating systems, programming
languages, etc., and I've trodden all this path before.  If I read an
insult, I generally respond sharply in kind, but I'll happily buy the
bloke a drink the next time we meet.


> A more complete statement of the goals and processes that FourThought
> uses in its testing would go a long way towards helping all of us understand
> what you are doing.

Fair enough that my lack of time to give long answers hasn't helped
communication hereabouts, but I thought Mike was doing a pretty good job
of it, so I left it short on my own part.

Unfortunately, I still don't have a great deal of time for exposition,
but I'll just note that Fourthought writes software in a variety of
contexts.  We do write system-critical (if not life critical) software
for clients, and those clients have a QA department with whom we work
closely and we don't institute the test policies ourselves.

However, the relevant context I brought up in this thread was our
open-source software: 4Suite.  We administer this with a balance between
scarcity of resources and the desire for quality, as many freeware
projects are administered.  We have a very large test suite which
includes elements of white box and black box testing, with the emphasis
on black box (as I think is somewhat clear from Mike's explainations). 
We add tests cases usually as we add features and find bugs (I think we
have a very good rate  for OSS of suppressing repeat bugs).  It's not
mission-critical QA, and we've never claimed it to be.

I should note that it's quite odd to have banking, ATC and such software
suggested as a standard in a list where we're all supposedly
afficionados of Python.

By the very admissions of PythonLabs, Python's core development has the
same realistic approach to QA as 4Suite development, and in fact, their
process is quite similar to ours.

> And, granting your correspondants the benefit of the doubt as to their
> intentions in writing you (that is, not taking their statements as 'bald
> insults' but rather as opportunties to clarify misunderstandings of
> your position) will also help to keep the discussion above the level
> of personal insults.

Do you really mean to suggest that the comment to which I responded was
not intended as insult?

Perhaps it wasn't, but even after reflection, it would surprise me if it
weren't.


-- 
Uche Ogbuji                               Principal Consultant
uche.ogbuji at fourthought.com               +1 303 583 9900 x 101
Fourthought, Inc.                         http://Fourthought.com 
4735 East Walnut St, Boulder, CO 80301-2537, USA
XML strategy, XML tools (http://4Suite.org), knowledge management



More information about the FRPythoneers mailing list