[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: RE: Potential bugs in user-written commands
Difficult though it may seem, I find myself
agreeing with both Patrick Joly and Lee Sieswerda
in most respects.
First, let me remind members of what is said in
our own FAQ.
"Before posting, consider other ways of finding information:
the authors of user-written ado-files (who usually have email
addresses, but are NOT always members of Statalist)"
That last detail has not been mentioned on in this thread, but
it is often crucial.
Second, I'd rather argue from examples in this case.
(I confine myself to user-written programs. I do think
that it is usually better to refer apparent bugs in
official code direct to Stata Corp. One good reason
for that is not making a fool of yourself. Only the
other day, I referred Stata Corp to what seemed to me
a clear problem and they kindly directed me to the
sections of the manual in which everything was explained.)
* It seems that Patrick's initial post was prompted by
a thread on a program of mine, -regplot-. The thread
was quickly resolved in this case because I was watching
Statalist. It was established fairly quickly that the
problem, reported by two people, was resolved by users
updating their Stata, as in their case -regplot- tripped
a bug now fixed in -graph-. If I had been out of town or
off sick, others would have, I guess, have sensed that
somehow and had a look at the problem and sorted it out
for users. That's what Statalist is for, in part. Also,
the need for updating your Stata may well have been noticed by
others, and served as a timely reminder.
* It is also very helpful to get bug reports direct
from users and I imagine that essentially all user-programmers
benefit from this. A bug is a bug is a bug and every
programmer worthy of the name wants to be told and to
be able to fix it as soon as possible.
* At the other extreme, I know of bugs in user-written
programs which have lain uncorrected for years,
despite reminders. For example, people have moved or otherwise
become inaccessible, or just never get round to fixing them.
In a tail of cases publicising a difficulty on Statalist
may be the only way to resolve a problem.
* For searches and help try: