This is for StataCorp, but I have to guess
that it is very highly unlikely.
First off, StataCorp _is_ a business. They
really don't want to discourage users from
upgrading, or give them any reason not to
upgrade.
Second, reading is not the only issue here, even in
principle. This would mean a massive hit
on Stata's code. Suppose you read in a
new dataset -- created by someone else's
Stata 10 - with new date-time variables.
Do you then expect Stata 9 to do the right
thing with its time series commands on
those variables? What would be the right
thing? To act like Stata 10? Or do you expect
Stata 9 to ignore all the things it cannot
handle? In that case, the key detail is that
Stata 10 has a -saveold- command. Any friends
who get Stata 10 before you do can share
datasets in that way.
Also, now that Stata 10
is announced, I doubt that there will be
any more updates for Stata 9, either
to executable or ados. If you report
a new bug or misfeature, StataCorp will
attend to it at the first available
opportunity, meaning early in the life
of Stata 10.
Nick
[email protected]
Richard Williams
> At 01:36 AM 6/4/2007, William Gould, StataCorp LP wrote:
> > 2. Because of the new time variables, Stata's .dta
> dataset format
> > has changed. This happens periodically and, because Stata
> > always reads old-format datasets, most users don't
> care or even
> > notice. It does mean, however, that if you use
> Stat/Transfer,
> > you will want to upgrade to Stat/Transfer 9 [sic].
> Yes, it's
> > confusing: Stat/Transfer 9 goes with Stata 10.
>
> Congratulations on the new release. Somebody suggested once that it
> would be nice if the last version of an old Stata included the
> ability to read files from the new version. I think that would be a
> very nice feature, as it takes a while to get everyone converted to
> the latest version.
*
* 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/
