As maintainer of the FAQ, I don't think so -- although naturally I am
open to any further expressions of opinion on the matter.
Kit's point that Martin's code fails if variable abbreviation is
disallowed is undoubtedly correct, but as yet this issue has not been a
source of much misunderstanding on this list. I don't think it is at all
practical to insist that code givers should never use variable
abbreviation, not least because I wouldn't guarantee to do that myself,
for all that I agree that Kit has a strong argument.
Also, I am reluctant to pad out the FAQ with minor details when there is
already much evidence that many people never read it, or read it
through. That may look like idleness, but no doubt those concerned
really feel that they are too busy or know that they should do it soon,
or some time.
I remain much more exercised by how to get all members to pay attention
to much more frequent sources of confusion, irritation and wasted
bandwidth:
1. Failure to read on-line documentation.
2. Formatted messages and/or attachments.
3. New posts that are replies to old threads.
4. Insufficient editing of previous posts.
5. Failure to search archives.
6. Out of office messages.
7. Poorly phrased questions.
8. Not saying exactly what you typed and what Stata did.
9. Not specifying out-of-date version being used.
10. Not specifying origin of user-written Stata commands.
11. Assuming universal understanding of arcane details.
12. Incomplete literature references, especially name (date).
13. Not specifying operating system when it is important.
14. Assuming everyone else is in the same country and even time zone.
15. Absent or dopey subject lines.
16. "Thanks in advance", which often looks disingenuous.
16. Failure to answer secondary questions.
17. Failure to close threads.
18. Repeated posts.
19. "What should I do next in my project?" Depends on what it is...
20. "Is this correct?" Depends on what you want to do...
21. "It did not work." Meaning precisely...
22. "STATA." Not read any documentation recently?
But all that said, Statalist when it works as intended is free, fast,
friendly (modulo marginal grumping), clear and correct, so who's
grumbling?
Nick
[email protected]
Martin Weiss
============
When Kit lumps something together with Windows, you know he means
business...
Well, your message makes me think that we do need clarification
regarding
this issue on the part of the Statalist FAQ.
Kit Baum
========
Martin said
I have pasted my own code into my do-file editor, hit the "Do" button,
and it runs w/o a hitch. ... Where did yours stop? I have used such
code in replies for months, and there has never been a complaint of
this nature. I do have a hunch, though, that it could have something
to do with your email program.
Your hunch is wrong. The code stops when you refer to variable 'cou',
which is undefined. You are making use of one of Stata's most ill-
designed, positively evil features: automatic variable name
completion. This feature bears a strong resemblance to WinXP default
security settings in terms of its potential for damage. I strongly
suggest that everyone
set varabbrev off, permanent
and not rely on this mechanism. Likewise, I often kvetch to authors of
SSC modules that their examples in help files do not run. In my
opinion no one should have to set varabbrev on in order to run sample
code.
I suspect Martin's code would work fine if I were to set varabbrev on,
but I am unwilling to do that, even momentarily, as I might then
forget to turn it off.
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
