Richard Williams wrote:
> At 09:49 AM 4/6/2005 -0500, Alan Riley wrote:
> >Stata 8 already was set up to deal with this problem because of the
> >difference in maximum value label length between Intercooled Stata 8
> >(80) and Stata/SE 8 (244). If Stata 8 encounters a label with any
> >mapping greater than its maximum length, it does not load that label:
>
> Ok, suppose a Stata 9 and Stata 8 user were both working on the same file
> and modifying it. Could the Stata 8 user zap the long value labels created
> by the Stata 9 person by saving the file? I know I'd sure be annoyed if I
> spent all this time typing in 32,000 character value labels and somebody
> else accidentally erased them :)
I don't know, but I think so.
Personally I seldom -save- datasets. When I work with a dataset I write
Do-Files to make the necessary changes in the dataset and then perform my
analysis in the same Do-File. There is no need to change the original
dataset, then.
If I were to -save- a dataset, I very seldom (never?) overwrite the original
one. Instead I use a new filename. If your Stata 8 user provides the Stata 9
user with Do-File of his modifications, the Stata 9 user can run this
Do-File on the original dataset and end up with a Stata 9 dataset with long
labels.
Sometimes I accidentelly save a changed dataset under the same name as the
orignial dataset. This happens if I accidently hit "Ctrl S" in Stata, which
causes Stata to execute the command -save, replace-. I have once asked
StataCorp to make -save, replace- impossible, but (unfortunately) with no
success.
Uli
--
[email protected]
+49 (030) 25491-361
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
