License for auto.dta?

[Michael Stepner]

I'm currently corresponding with an open source project that would like to use auto.dta (of -sysuse auto- fame) for a unit test in their code. But before they include it in their codebase, we need to confirm what license that file is released under. Does anyone here on Statalist know?

The only hint I can find is on this Stata Press page:

StataCorp gratefully acknowledges that some proprietary datasets in the reference manuals have been used in our printed documentation with the express permission of the copyright holders. If any copyright holder believes that by making these datasets available to the public, StataCorp is in violation of the letter or spirit of any such agreement, please contact [email protected] and any such materials will be removed from this web page.

If auto.dta isn't available under an open license, that's okay, I can create my own custom dataset for unit testing. But regardless, since auto.dta is the most widely used Stata reference dataset, I think it would be useful to clarify its intellectual property status.

Thanks for your help!

[Nick Cox]

I don't think it's open source.

[Sergiy Radyakin]

StataCorp has provided the following description:

"The automobile dataset auto.dta is copyright 1979 by Consumers Union and is reproduced by permission from Consumer Reports, April 1979."
See: help copyright

R uses another version of the auto dataset. Perhaps it is also suitable for your needs. You can find it with description here: http://lib.stat.cmu.edu/datasets/

Finally, rather then committing the dataset into the repository, consider linking to e.g. http://www.stata-press.com/data/r8/auto.dta These links remained pretty stable for many years.
This may be more appropriate, though not 100% "clean" in terms of autonomy.

Best, Sergiy Radyakin

[Michael Stepner]

Thanks Nick and Sergiy, this is a very comprehensive answer. It's clear that auto.dta is not openly licensed, so we'll go ahead and construct a dataset for unit testing.

Sergiy, your work tracking down that info is much appreciated.