[R] R in the NY Times

[Marc Schwartz]

on 01/07/2009 09:47 PM Gabor Grothendieck wrote:
> On Wed, Jan 7, 2009 at 10:26 PM, Dirk Eddelbuettel <edd at debian.org> wrote:
>> On 7 January 2009 at 18:24, Gabor Grothendieck wrote:
>> | By running the code below we see that the:
>> | - sum of the three seems to be rising at a constant rate
>> | - S is declining
>> | - SAS and R are rising
>> | - R is rising the fastest through its completed its phase
>> | of highest growth which ended around 2004
>>
>> I wonder whether we need to account for traffic on all the additional r-sig-*
>> mailing lists ?
>>
>> Of the handful that I follow, some seem to have taken traffic from r-help.
>> This could account for (at least parts of) the apparent traffic growth
>> slowdown since 2004 as many of these added lists appeared only in the last
>> few years.
>>
> 
> Good observation.  It would be interesting to combine the data from all
> the lists to see what the effect is.

Agreed.

You can use the basic framework of the R-Help code that I posted
yesterday to do that.

The key gotcha is that some of the list archives have the posts stored
on a per calendar quarter basis, not monthly. At least one has a mix.
This seems to be somewhat dependent upon list volume, though that is not
a consistent factor.

Thus, you would have to review each archive individually and adjust the
archive URL's in the code accordingly.

You would also see the impact on the subsequent aggregation of the data,
since the monthly time series based analyses (as opposed to yearly) will
have to be adjusted, given the differing granularity of the data.

HTH,

Marc