[R] R-help mailing list activity

[Duncan Murdoch]

On 23/01/2016 7:28 AM, Jean-Luc Dupouey wrote:
> Dear members,
>
> Not a technical question:
>
> The number of threads in this mailing list, following a long period of
> increase, has been regularly and strongly decreasing since 2010, passing
> from more than 40K threads to less than 11K threads last year. The trend
> is similar for most of the "ancient" mailing lists of the R-project. I
> cannot imagine the total number of R-related inquiries on the Internet
> decreased. It means that contributors have gone elsewhere. Indeed, in
> the meantime, the number of R posts on stackoverflow passed from 2K to
> 100K between 2009 and 2015. Thus my question: what are the
> specificities, the plus and minus of the R-project mailing lists, in
> comparison with other lists, and especially in comparison with
> stackoverflow? A lot of threads are duplicated on both lists, which
> seems to me a little bit counterproductive.

I don't see duplication as counterproductive -- some people like one 
style, some like the other, both will find answers.

However, I think there is less duplication than you might think in many 
areas.  Mailing lists are preferable when the people who are good at 
answering your questions use the mailing lists; Stackoverflow is 
preferable when the good answers are there.

I generally prefer the mailing lists, though I occasionally participate 
on Stackoverflow.  The reasons I prefer them:

  1. Permanence.  If Stackoverflow shuts down tomorrow, all posts there 
will likely disappear.  There are several locations that archive the 
mailing list posts. I have local copies of a few thousand posts on my 
own laptop.

  2. Familiarity.  I've been using the mailing lists for 20 years, and 
its easier to continue than to change.  If you're more familiar with the 
Stackoverflow process, you'll probably prefer that.

  3. Simplicity.  This may be a repeat of 2, but the Stackoverflow 
distinction between answers and comments, it's gamification (badges, 
special privileges to high scorers, etc.) just seems unnecessarily ornate.

  4. Interaction.  The mailing lists are a series of conversations, 
whereas Stackoverflow is more like Wikipedia, i.e. a joint project to 
which you can contribute.  (Maybe there are conversations on 
Stackoverflow as well, but I'm not a big enough user to know about them.)

If I look at my own recent record, I tend to answer far more questions 
on the mailing lists, but ask more on Stackoverflow.  I think this is 
due to my original point:  the experts in the topics I'm asking about 
are more likely to be there than here.

Duncan Murdoch

P.S. Your statistics are a little misleading:  you counted threads in 
one R mailing list in one year, and cumulative questions in all R topics 
over 7 years in Stackoverflow, so the difference in traffic isn't as 
large as your numbers look at first glance.  However, I think it is true 
that the mailing list traffic declined and Stackoverflow increased over 
that period.