I recently found out that I am experiencing occasions in which,
having saved a modified version of a program before rerunning it in
Stata, the new version does not get run, but instead version from
perhaps seconds ago is run, giving the same error repeated. I write
and run Stata programs in Textpad, and go back and forth from Stata
to Textpad, modifying code, rerunning it, etc.
After thinking I was losing my mind ("I swore I just corrected that
spelling mistake 2 sec. ago. Why is it still there?") What I think
is going on is that, between whatever write-behind disk caching XP
does (even when one turns it off), and whatever file-caching Stata
might do, and whatever Textpad might do, a file saved in Textpad,
even when saved several times in succession, is not resulting in the
version being accessed by "do myfile.do" being the freshest version.
I find that doining a "save-as" rather than a "save" (but to the same
filename) avoids the problem, but I don't think that this definitely
conficts any one of the several softwares involved.
I'm running v.11, updated. It's my casual impression that I did not
have this problem come up in v. 9 or 10. Anyone else experience
this, or have any ideas for a cure? The problem, I think, could be
either a main effect or interaction of the several software
programs. And yes, I could change to a different OS or different
text editor, but that seems like overkill for this problem. Using
the built-in do file editor, for me, is not a choice until it permits
remapping of the keyboard, contains good search and replace features,
reg. expressions, etc.
FWIW, I've had this problem on more than one Wintel machines, of
quite different vintages.
Mike Lacy, Assoc. Prof.
Soc. Dept., Colo. State. Univ.
Fort Collins CO 80523 USA