Received: by zonker.hpl.hp.com (1.37.109.8/15.5+ECS 3.3+HPL1.1) id AA25820; Tue, 27 Sep 1994 12:57:18 -0700 Date: Tue, 27 Sep 1994 12:57:18 -0700 From: Ken Poulton <poulton@zonker.hpl.hp.com-DeleteThis> Return-Path: <poulton@zonker.hpl.hp.com-DeleteThis> Message-Id: <9409271957.AA25820@zonker.hpl.hp.com-DeleteThis> To: windspeed@zonker.hpl.hp.com-DeleteThis Subject: list info
I have just changed the lists to send out a Reply-To: field that
indicates the original sender rather than the mailing list. Berekely
mail(1) still replies to the list with r{eply} but to the sender with
R{eply}. Your mileage may vary. This is certainly good for the data
lists, but may be annoying for wind_talk. Let me know.
Occasional posters often wonder whether their messages to the lists got
lost; by default the listserv system avoids sending you a copy of your
own message. To avoid the problem, I have changed all subscriptions to
acknowledge your mail by sending you a copy of your message along with
everyone else on the list. You can send mail to the list processor to
change back to NOACK for your subscription if you like.
Some people have noted that we're often getting two mailings per hour.
Allow me to explain the reason for this annoyance.
The primary network data source that supplies all the current readings
(don't ask where) has gotten reaaal slow of late. It appears that the
network connection to that whole subnet is very heavily loaded (like,
20-50% of the packets fail on a ping). This causes my data retrieval
script to fail fairly often. It retries, but the result is that data
from this source is often 15-25 minutes after the hour.
I also have a secondary source, but the data there is much sparser and
watered down for use by the public. In particular, the wind direction
is reduced to w, nw, n, etc. I hate using this source for SFO because
there is a big windsurfing difference between 270 and 290.
My compromise is to mail with the secondary data (for SFO only) if the
primary source does not come through by 14 minutes past the hour. In
this case, the secondary-source SFO data shows up at the very end of the
message, with time nn59 and a direction ending in "2". When the
secondary source does come through, we get a second mailing with the
real data.
I'm open to other suggestions; please follow up in wind_talk.
Ken Poulton
poulton@opus.hpl.hp.com-DeleteThis
"The reason God was able to create everything in 6 days was that he
didn't have an installed base."
This archive was generated by hypermail 2b29 : Mon Dec 10 2001 - 02:28:12 PST