[R] Newsgroup - another try?

[Martin Maechler]

>>>>> "ThomasP" == Pesl Thomas <thomas.pesl at arcs.ac.at> writes:

    ThomasP> Is there any reason why the r-help-mailinglist should not be
    ThomasP> converted to a newsgroup?

    ThomasP> These were the advantages of a newsgroup I mentioned earlier:

        ThomasP>   ....

    ThomasP> -) the discussion is faster (I experience that the R-mailinglist
    ThomasP>    has a lag of about 1 to 2 hours (not for everyone!!!).

        ThomasP>   ....

    ThomasP> Now I find that my lag is increasing (up to 4 hours). 

Almost every point has been answered/discussed already.

Just a remark from the mailing list maintainer on the (non)-speed of R-help:

Yes, the delays are longer than they should; if you subscribed late (in the
history of R) and as first person from your "e-mail domain",
you will have longer delays than others.

It is our server (and its setup of "sendmail") which is not faster, but at 
least it's one where I have lots of control.  
Setting up and configuring the basic mail service is not my business though.   

I have tried to use an alternative to sendmail ("bulkmail") for a mailing
list `explosion', but that failed for obscure reasons that I didn't want to
follow up back then.  The situation *will* improve hopefully within the
next few months for another reason: The mail server has currently too many
duties (including CRAN.ch) and hopefully will soon be retired from one of
its major tasks, and hence be speedier in mail handling.
BTW: We have currently 822 adresses subscribed to R-help and 568 to R-announce
     (which is gatewayed into R-help). If I add R-devel and the *-digest
     lists and uniqify the subscribed addresses, we currently have 1338 unique
     subscriber addresses (up from 1119 as of Feb.2, when I started
     collecting the combined statistics).

As to your delays, Thomas, here are the facts :

  > ## Data from Aug.18 -- Sept.4, 2001 [13:45]
  > ## On Server:
  > ## 1)
  > ##   grep -i 'pesl.*delay' /var/log/syslog > syslog-pesl
  > ##  zgrep -i 'pesl.*delay' /var/log/SYSLOG/syslog.2001.Aug.25.gz > syslog-pesl-1
  > ## 2) Using Emacs on these two:
  > ##  i. goto beginning   (query-replace-regexp ".*, delay=" "" nil)
  > ##  ii.goto beginning   (query-replace-regexp ",.*" "" nil)
  > ## Result of both files concatenated in to this one:
  > str(del.pesl <- scan("/u/maechler/Mail/MISC/syslog-pesl-delays", what = ""))
  Read 151 items
   chr [1:151] "00:25:52" "02:44:42" "03:03:42" "03:08:49" "02:38:14" ...
  > str(ll.d <- strsplit(del.pesl, ":"))#- all sublists of length 3
  List of 151
   $ : chr [1:3] "00" "25" "52"
   $ : chr [1:3] "02" "44" "42"
   $ : chr [1:3] "03" "03" "42"
   $ : chr [1:3] "03" "08" "49"
   ..........
   $ : chr [1:3] "02" "47" "01"
   $ : chr [1:3] "02" "33" "59"
   $ : chr [1:3] "02" "28" "42"
  > str(secs <- sapply(ll.d, function(ch3) c(as.integer(ch3) %*% c(3600, 60, 1))))
   num [1:151]  1552  9882 11022 11329  9494 ...
  > dSecs <- structure(secs, units = "secs", class = "difftime")

    ## too bad, we don't have a nice  summary.difftime()  yet ...

  > summary(dSecs) / 3600
     Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
    0.179   2.200   2.336   2.275   2.558   3.583 

so, it's 2:20 that you have to suffer on the `average' which seems
well bearable. As you point out yourself: If you post yourself, you get
answers by direct e-mail very quickly. For everything else, the people who
are on the train for a longer time will have the opportunity to answer
before the newcomers -- which I think is not bad.

By the way, there are quite a few subscribers to "r-help-digest" which is
the digest version of R-help arriving (at most) 6 times a week only.  They
know they lag behind and still sometimes react very helpfully.

Regards,
Martin Maechler <maechler at stat.math.ethz.ch>	http://stat.ethz.ch/~maechler/
Seminar fuer Statistik, ETH-Zentrum  LEO D10	Leonhardstr. 27
ETH (Federal Inst. Technology)	8092 Zurich	SWITZERLAND
phone: x-41-1-632-3408		fax: ...-1228			<><
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._