Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: st: What is the proper way of modifying user-created ado-files. . .


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: What is the proper way of modifying user-created ado-files. . .
Date   Wed, 18 Nov 2009 19:53:57 -0000

Dear Roy: 

You are clearly very angry about this subject for reasons that remain
highly obscure to me personally. 

I don't want to impugn or inhibit your freedom of speech, which someone
else accused me of a short while ago, I believe unfairly. 

But I think you are in error about the purpose of the Statalist FAQ. I
note that you quote this rather indirectly, which hardly fits your
stance on the importance of giving proper credit and referencing. 

The Statalist FAQ exists to advise people how to send postings to
Statalist. Although some of it touches on various other subjects, it is
not designed as a code of ethics for use of Stata software, or indeed
any other activity. It is emphatically not appropriate for the
maintainer of that FAQ to attempt to lay down rules for how people
conduct themselves in using user-written software. 

Naturally, I personally deplore plagiarism of code just as you do. I
just happen to believe that in practice it is not at all easy to
establish plagiarism of code. I touched on some of the reasons in a post
earlier today. But equally I hope very much that people can learn from,
and imitate, tricks and techniques in my code that are relevant to their
problems. Those principles sit together uneasily, but there is a simple
recommendation. Anyone paranoid about being ripped off should keep their
code private. Period. 

Similarly, I strongly support the importance of proper referencing and
acknowledgments in code. But I really don't know why you believe that
this is being undermined. I sense no collective calamity or breakdown of
trust within the Stata user community, as you seem to be implying. 

Beyond that, there is law in various countries that may or may not apply
to situations such as yours. I really can't judge for all sorts of
reasons. There are also personal and public belief systems that affect
many people's behaviour. But the Statalist FAQ is, to repeat, not a code
of ethics, and I won't participate in making it one. 

By the way, only one person is responsible for the Statalist FAQ,
currently myself. Nevertheless I have attempted to be sensitive to a
consensus of opinion in revising the FAQ from time to time. I think that
several people could testify that as they gained experience on the list
they could see the point of the suggestions it contains. Indeed, a few
have morphed from naughty newbie to upright citizen. Besides, although
many people have been firmly requested not to do various things on the
list, no one has ever been thrown off the list for bad behaviour and we
are proud that this is so. 

Please note also that neither the list convenor (Marcello Pagano) nor
myself is elected democratically. Rather, Marcello runs a list and I
document its FAQ and anyone's rights are limited to making their views
known and setting up some other list or forum if they do not like how
Statalist is run. That may seem authoritarian or autocratic, but it is
how the list has been run since inception and I believe that it is
widely appreciated. Equally, there is no sense in which Statalist
oversees the Stata user community. Many users are not even members and
owe it absolutely nothing. 

More importantly, if some people are behaving as badly as you seem to
believe, it is hard to imagine that preparing a document telling them
not to do that is going to be anything but futile. 

It is not my place to say what StataCorp should be saying. But I don't
see that StataCorp has any role in laying down the law on this topic
either. 

I am very unhappy to see you upset about this topic, but if you have
specific grievances, as you clearly believe, then nothing can be
discussed seriously without specific evidence. Equally, I believe that
an email forum is not a suitable place for accusations or insinuations
of personal bad behaviour and that you won't promote your cause
effectively by emails such as those you posted today. For example, I
strongly support Kit Baum's recent comments rejecting your claim that
user-written programs have been getting dumber recently. I can't imagine
that the user community can feel anything but collectively insulted by
such comments, especially new user-programmers attempting to contribute
as best they can. I am sorry that my recent programs, included in such a
remark, have not met your own standards. 

Sincerely, 

Nick 
n.j.cox@durham.ac.uk 

Roy Wada

>> and imply that a suitably large portion of the credit for the program
belongs to
>> the original authors."

I would suggest that several responsible people do some powwow and
change this language. This is a problem waiting to happen again and
again. I would further suggest a Stata FAQ (not Statalist FAQ) be
established regarding copyrights and authorship and the proper way to
give credit and citation.

I would also suggest that people, including myself, start treating the
program and the help file more like an acadmic output if we are so
keen on presenting them as academic outputs. This means the help file
should be expanded to include a small reference section when
appropriate.

As I have mentioned before, giving citation is mandatory for anyone
who borrows substantially from another person's work. There is no if's
or but's about this. I don't make these rules. These are the academic
rules and traditions dating back to the end of middle age when people
involved in research realized the need for systematic citation in
order to make any academic collaboration possible.

In the past, people have proudly declared that their programs are
superior to the original without mentioning where they got it in the
first place. Giving citations will prevent this kind of nonsense.

Imagine yourself in a situtation where your own work has been
exploited in this manner. Only sure way to prevent that from happening
is to refuse to share work, which is not good for anybody, or have a
system in place that protects everyone's work and their reputation
against this type of nonsense.


*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index