bobm@rtech.UUCP (Bob Mcqueer) (07/26/87)
Every few weeks I get evidence that vn is out there at sites other than ours
happily rooting through articles. Once I got curious and scanned our spool
directories for followups which seemed to have been posted from vn (you see,
there's a REASON for having different wording on the introductory line), and
found articles from about 50 sites over a 1 week period, which comprised
about 2% of the total followups posted. Not a huge amount, but significant
I think, especially as the program seems to be favored by the "passive"
or "casual" audience that reads & never posts.
I've modified vn a good bit internally with the intention of allowing
it to talk to such things as NNTP servers, databases & so on. I would
like to post this as a new version, which I am "soaking" here a rtech
right now. In addition to formalizing the interface for talking to
servers, and incorporating a couple bug fixes, a few minor enhancements
will show at the interface:
1 Saving articles has some enhancements for editing the old savefile
name, creating individual directories for newsgroups, overwriting
instead of appending, etc. It also works a bit faster (get rid of
that silly business of forking a cat)
A few other strings such as pattern strings are saved for
editing the old copy, rather than prompting for a new one each
time, also.
2 All shell variables used which don't begin with "VN" (such as PS1)
may be overridden by using VN<name> instead. I'm using this to
define VNEDITOR as a script which invokes ispell after vi, and not
having it conflict with the EDITOR variable for ksh.
3 Commands for Top / Bottom / Middle of page
4 Takes care of updating .newsrc with a "read" number reflecting
removed articles. Will prevent accumulating those newsgroup
scanning messages on newsgroups with nothing in them - speeds
things up a bit, too. Will help if your administrators do like
ours and simply remove articles from the spool directories when
space gets tight.
I also allowed it to be compiled to collect some logs, a feature which I
hope is used only for tuning purposes, and is not put to obnoxious uses.
Given the current division of the source newsgroups, I'm once again wondering
where to post the thing. comp.sources.unix? comp.sources.misc? If there's
a consensus, I'll follow it. I'll post the new version in a couple weeks.
I'm following with another article which describes the server interface
routines to tie vn to an arbitrary news server, for the interested parties.
The next order of business would be to provide NNTP'ed versions of the
server interface routines. If somebody wants to help with that, I'm
interested.
--
Bob McQueer
{amdahl, sun, mtxinu, hoptoad, cpsc6a}!rtech!bobmbobm@rtech.UUCP (Bob Mcqueer) (12/21/87)
Some time ago I said I would post an updated vn. About eight weeks ago I
bounced it off to comp.sources.unix. Since I haven't seen it appear there,
I've posted it to alt.sources. Note that I am in no way intending to denigrate
the moderation of comp.sources.unix. Perhaps there's an enormous backlog.
Perhaps there is some failure of e-mail between rtech and the uunet machine.
It's wonderful that somebody is willing to do the job of moderating the
group, and trying to build all the various & sundry stuff that comes across.
If vn eventually gets out there, so much the better. But, I thought it ought
to get out, so it went out to alt.sources for the time being.
Anyhow, some notes. The 8/87 version of vn will seem pretty much the same as
the old version, externally. There are a few fixes and tweaks you will
notice, the most visible being the saving of your old input strings, allowing
you to edit them in contexts where it makes sense, such as search strings
and save file names. There's also some logging features, and some more
flexibility in how you can configure mail to be handled. I hope the logging
features are only used for system tuning, not for "big-brotherism".
The code has been reorganized to allow the interface to be used against
different news servers, as I have discussed before. There is a "server.doc"
file describing the routines needed to provide an alternative server interface
layer.
The source files in this new posting should simply replace your old set -
there are some new files, and some old ones are obsolete. Most of this is a
result of restructuring things to have a formal layer that can be replaced
for a different server.
NNTP. I have made a version now which talks NNTP, and I'm soaking it here.
I will probably eventually post this as a set of diffs against what I'm
posting as 8/87 "standard" version. I would be willing to have some of
you who have written to me regarding NNTP also be trial sites, since I think
our configuration for our NNTP server may be a trifle unorthodox.
------------
{amdahl, sun, mtxinu, hoptoad, cpsc6a}!rtech!bobmdavid@ms.uky.edu (David Herron -- Resident E-mail Hack) (12/22/87)
In article <1488@rtech.UUCP> bobm@rtech.UUCP (Bob Mcqueer) writes: >Some time ago I said I would post an updated vn. About eight weeks ago I >bounced it off to comp.sources.unix. Since I haven't seen it appear there, >I've posted it to alt.sources. Note that I am in no way intending to denigrate >the moderation of comp.sources.unix. Perhaps there's an enormous backlog. >Perhaps there is some failure of e-mail between rtech and the uunet machine. For Your'all's Information ... I've got something in the queue as well which Rich recieved about 1 1/2 months ago. He ack'd it at the time and sometime later said to me that he'd been busy and so forth. That was a couple of weeks back when the Yale StarChart programs came through. This has happened with Rich many times in the past. He's just busy and when things cool down for him at work he'll start posting again. -- <---- David Herron -- The E-Mail guy <david@ms.uky.edu> <---- or: {rutgers,uunet,cbosgd}!ukma!david, david@UKMA.BITNET <---- <---- Winter health warning: Remember, don't eat the yellow snow!