Re: st: how to cite others' code?

[Nick Cox <njcoxstata@gmail.com>]

A belated postscript to this, and broadening the issue beyond
Rebecca's question:

Let's take it as understood that code you used published through
journal articles or books should be acknowledged in the same sort of
way as any other journal articles or books.

... and also that acknowledgments are usually welcome (although there
can be reasons on occasion to decline politely to be acknowledged).

The more difficult area stretches from personal or public
communication or dissemination of code (e.g. Statalist, SSC, personal
websites) that falls short of journal or book publication.

An extra rule of thumb of mine, which may or may not be echoed by
others, is to look at whether someone bothered to write a help file.
If they did, then I think program authors usually expect
acknowledgement. Writing the help can easily be as much work as (in
some cases, much more work than) writing the program and usually
signals that the author believes the work has some lasting or wider
value, at least when they did it.

There appears a widespread impression that putting code in the public
domain and making it freely available is a kind of relinquishment of
intellectual rights, in so far as it has been "given away". In many
ways, the opposite is true. However mild or modest the tone, the
implication is "For what it is worth, I did this work".

It is also true that, as I think Feynman said somewhere, the same
problems have the same solutions, so that it very easy, indeed
entirely to be expected, to find that people write very similar code
for the same task, so sometimes originality can be hard to assess or
establish.
Usually, we should just decide not to worry about this.

Nick

On Fri, Feb 25, 2011 at 8:29 PM, Rebecca Pope <rebecca.a.pope@gmail.com> wrote:
> Hello Statalist-ers,
> Is there a standard format or set of guidelines for citing code
> written by others? I am _not_ asking whether one ought to cite pieces
> of code written by others, but what is required to be in that
> citation. If a standard does exist, does it differ depending on
> whether the code comes from Stata Corp, user-written ado files, or was
> posted as an example to the list? I did try to search for this before
> posting, but all I found were some posts relating to people not citing
> others' code. If guidelines are already codified & available, I'm
> afraid they have been obscured in my search results by this previous
> discussion.
>
> As a concrete example, Svend Juul posted a question a few days ago
> about generating new variables in a command when a -varlist- is also
> required. I do something similar in one of my programs, but Nick's
> response (http://www.stata.com/statalist/archive/2011-02/msg00861.html)
> is far more flexible than my solution to the problem. If I were to
> replace my code with what Nick posted, how would that be correctly
> cited?
*
*   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/