[R] A Few Suggestions to help out newbies
ggrothendieck at yifan.net
Tue Apr 2 23:01:29 CEST 2002
On 2 Apr 2002 at 12:01, Liaw, Andy wrote:
> Well, as Thomas Lumley said in the other message, because no one who wants
> it bad enough had *contributed* the code. Many things in Open Source
> projects such as R get done because some users wanted them bad enough that
> they wrote the code, and then contributed to the project. Just think about
> it, how many people have the motivation to write and contribute code that
> they don't need themselves, just to please some people who do nothing but
We need to be careful here. Its an advantage to have user feedback and it
should never be viewed as complaints. Without such feedback, how
else can one know what people find easy and difficult and what they view
Providing personal requirements and wish list items and telling others what they
find easy and difficult contributes to development just as much as programming
and design does. The entire process is a feedback loop and the loop is not closed
unless all the links are in place.
Since many people in the R community likely work in academia let me drraw a
parallel in that world. A course which has been given multiple times is generally
better than one given the first time since on subsequent rounds one can focus better
on those items that students had difficulty with in prior rounds. Knowing the material is
different from knowing what material the students find difficult. This feedback from
user to developer/teacher is crucial. The same goes with text books. If you can find
one that has been student tested its often more useful than one that has not.
It has been said that the key difference between Open Source projects and commercial
projects is not just that the commercial packages cost money but
that the commercial developers listen to their users better than Open Source
developers do. (I am not saying that R developers don't listen, I am just
bringing out this view of Open Source vs. Commerical, in general). If commercial developers
don't listen to their users then they don't stay in business very long!
Its worthwhile for Open Source projects to set up some sort of analogous dynamic as well.
In the case of the R, the R Help list is an integral part of R since it provides some of this
Another thing that would be nice would be some sort of organized wish
list / requirements. This already exists to some degree (see the NEWS files
at http://stat.ethz.ch/R-alpha/ ). However, it might be nice to give this aspect
better prominence and provide a mechanism for the large community focused
on R to provide this feedback.
This would be helpful since anyone who wished to contribute would then have
a framework for understanding what is needed.
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !) To: r-help-request at stat.math.ethz.ch
More information about the R-help