rcd@opus.UUCP (Dick Dunn) (10/12/84)
There's one saying we're going to be seeing a lot on the net in the future.
It's:
*** REPLACE THIS LINE WITH YOUR MESSAGE ***
I'm not sure why the newest postnews finds this necessary. I can damn well
find where I want to put the message without any help, but it's easy to let
the funny *** line slip off the screen while you're editing and forget to
delete it.
A second over-helpfulness is sucking in the entire base article prefixed
with "> " for a followup. What we're getting as a result is followups
which contain long, largely irrelevant base articles--even to the signature
lines. (It would have been easy enough to have a separate program to
generate the "> "ed base article--at least that way the default would be
minimal relevant material instead of maximal irrelevant material, plus it's
easier to implement:-)
The business of tucking a copy of your posting into author.copy is another
feature I could do without. If I need a copy I can always save it from the
editor, or I can save it later when I encounter the posting in reading
news. Partly this was annoying because it took some poking around to find
out how to disable it--and one thing I did NOT need was yet another
environment variable.
--
Dick Dunn {hao,ucbvax,allegra}!nbires!rcd (303)444-5710 x3086
...Relax...don't worry...have a homebrew.chuqui@nsc.UUCP (Zonker T. Chuqui) (10/13/84)
> There's one saying we're going to be seeing a lot on the net in the future. > It's: > > *** REPLACE THIS LINE WITH YOUR MESSAGE *** > > I'm not sure why the newest postnews finds this necessary. A good point-- I've started noticing these things on the net. Perhaps a bit on the overzealous side??? > A second over-helpfulness is sucking in the entire base article prefixed > with "> " for a followup. I've fixed this in my version, and the fixes have gone to poor, overworked, rick for inclusion in the next postings from him. what I did was set things up so that by default you didn't get the message but so you could tell readnews and vnews if you DID want it. I also did the same for replies (did you notice how you ALWAYS got the message for followups and NEVER for replies??? Talk about standardization... *grin*) > The business of tucking a copy of your posting into author.copy is another > feature I could do without. Also fixed, originally by Mark at Fortune, and I've hacked it further. My version (also at poor, overworked rick's house) doesn't save the message (following the true unix standard of doing nothing and saying nothing unless asked) unless you set an environment variable. I haven't figured out how to deal with YA environment variable due to lack of time, but I'll probably get frustrated enough one of these days... chuq (2.10.2 is wonderful... Everyone should have a copy!) -- From the Department of Bistromatics: Chuq Von Rospach {cbosgd,decwrl,fortune,hplabs,ihnp4,seismo}!nsc!chuqui nsc!chuqui@decwrl.ARPA How about 'reason for living?'
karsh@geowhiz.UUCP (Bruce Karsh) (10/14/84)
> > A second over-helpfulness is sucking in the entire base article prefixed > with "> " for a followup. I vote that this feature be retained. It helps keep replies in context. I think that users should be aware that they don't have to keep the entire "> " prefixed article in their follow ups. It is sufficient just to keep enough lines that it is clear what they are following up. > > The business of tucking a copy of your posting into author.copy is another > feature I could do without. I vote that this feature be retained. If you don't like it, set the environment variable NEWSARCHIVE to /dev/null. I don't see why it is necessary to complain about features that are so easily disabled. I keep all of my postings in author-copy. Bruce Karsh
rcd@opus.UUCP (Dick Dunn) (10/16/84)
> > A second over-helpfulness is sucking in the entire base article prefixed > > with "> " for a followup. > > I vote that this feature be retained. It helps keep replies in > context. I think that users should be aware that they don't have > to keep the entire "> " prefixed article in their follow ups. I think that, in an ideal world, picking up the article with "> " is a good idea. My concern is that things are not "all for the best in this the best of all possible worlds" and that it's going to give us much longer articles. (Without the base article cited in the followup, there's still the "References:" line which lets you go back to the parent article. Different people need different amounts of context.) Also, I didn't realize that the convention which makes the article's pathname available (in the shell variable A) was not widely known/used. > > The business of tucking a copy of your posting into author.copy is another > > feature I could do without. > > I vote that this feature be retained. If you don't like it, set the > environment variable NEWSARCHIVE to /dev/null. I don't see why it is > necessary to complain about features that are so easily disabled. No feature is free, and even if any individual feature is quite cheap, the sum of them can place a substantial burden on understanding, documentation, and maintenance. (I admit that I overreact to creeping featurism.) In the case of author_copy, what happened was that I couldn't find the documentation for this new version of news on our system. I tried making author_copy an unwritable file--that kills postnews before it posts the article. Next I found someone who knew where the source code was; we looked and found the use of NEWSARCHIVE in the code and I dutifully trudged off and set NEWSARCHIVE=/dev/null, exported it, and re-executed .profile. This worked but still gave me a message and needlessly "wrote" a copy of the article. Turns out that the correct way to disable the feature (found by later experimentation on a hunch) is to set NEWSARCHIVE to a null string--the message, the copy, and everything goes away. (This seems not to be well known...after my original posting a couple of people asked how I had disabled it.) This is what the feature cost me, and tho I admit to stumbling around, I'm not a complete idiot. It would be more useful if there weren't other easy ways to solve the problem. -- Dick Dunn {hao,ucbvax,allegra}!nbires!rcd (303)444-5710 x3086 ...Lately it occurs to me what a long, strange trip it's been.
lepreau@utah-cs.UUCP (Jay Lepreau) (10/19/84)
If that's the kind of "features" in 2.10.2 I think we'll do without. Yuck.
chip@t4test.UUCP (Chip Rosenthal) (10/20/84)
A little history on the current postnews. I'm not sure who is
responsible for the 2.10.2 distribution of postnews; Rick Adams might
know. However, it is very much based on a version I sent to
net.sources several months ago. My version was a quick hack on the
postnews.c Mark Horton posted, and retains much of his structure,
ideas and features.
As I have seen, three features have been questioned:
1) *** REPLACE THIS LINE WITH YOUR MESSAGE ***
2) Including original articles in a followup with '>' marks.
3) Saving articles with the NEWSARCHIVE feature.
The "REPLACE THIS" line was in Mark's original version. I liked it and
carried it over. One of my design goals was to minimize errors by
novice users, while trying to minimize the junk an experienced user
must trudge through. I hear and understand your complaints. I will
probably remove this from a future release, and use some other method
of making a novice user understand what to do when s/he suddenly finds
her/himself dumped into an editor.
I agree with the gripes behind the second point. That is why I have
added a '-f' flag to surpress the original article in a followup.
Also, a '-m' flag may define what mark you want. Including the
original article in a followup is an important feature. It should
remain there, but a method for getting around it (other than deleting
lines in the editor) should also be available. I don't know how to
incorporate these flags into 'readnews' and 'vnews'. I haven't looked
at it because we use 'rn', and adding them was very straightforward.
(It did require a small change to the source and common.h).
I have found the NEWSARCHIVE save feature to be a great help. This
came originally from my "archive" program (also posted to net.sources a
while back) which saved articles from readnews. This allowed the user
to have an archive directory with all saved articles and copies of
posted articles. I haven't looked at the 2.10.2 postnews, but in the
original undefining NEWSARCHIVE would surpress the saving. I strongly
suggest not turning off the "A copy has been appended to..." message.
Several times I have run postnews from a non-login shell, so .cshrc
hadn't been run. The lack of this message warned me that I didn't have
a copy automatically made for me. (Maybe this is an issue only on
Eunice sites, where it is nice to sometimes open up a shell without
getting an environment defined. But since we are running Eunice I am
obviously going to try to put in features which I find helpful. Of
course I will try to keep them from being a nuisance for those folks
with more normal Unix installations.)
So, where do we go from here? I am working on a new version of
postnews (and some other news stuff). I expect to complete it
shortly. I will keep these suggestions, as well as other ones people
might have, in mind. Once completed, I will send out a request for
test sites. Once verified, I would like to see it distributed.
Comments and suggestions would be appreciated.
--
Chip Rosenthal, Intel/Santa Clara
{ idi|intelca|icalqa|kremvax|qubix|ucscc } ! t4test ! { chip|news }greg@sdcsvax.UUCP (Greg Noel) (10/22/84)
In article <1169@t4test.UUCP> chip@t4test.UUCP (Chip Rosenthal) writes: > That is why I have >added a '-f' flag to surpress the original article in a followup. >Also, a '-m' flag may define what mark you want. >... we use 'rn', and adding them was very straightforward. Huh? I don't understand this. In our version of RN there is an "f" command to post a followup and an "F" command to post a followup that includes the original article. The prefix mark is defined by the -F flag and defaults to ">". This is not a new feature -- we were a Beta test site, and this has been in from the very beginning. Why did you have to add new flags? -- -- Greg Noel, NCR Torrey Pines Greg@sdcsvax.UUCP or Greg@nosc.ARPA
mcdaniel@uiucdcsb.UUCP (10/23/84)
As for followup putting the base article into the reply:
In vi and xed, for example, <command>! can mean "do <command> with
something extra". For example, "w!" in vi means "really write this
file; don't warn me about possible problems". May I suggest that you
use two different commands along the lines of "followup" and
"followup!", which is analogous with these editors and with notes?
"followup!" would mean "followup, include the text". Then when you are
trying to get new users started, you just explain "followup", and
mention that "followup!" exists.
Another way to handle this problem is to ask a question before putting
them into an editor.
------------------
If I am obnoxious or if I flame, please find it in your heart to forgive me.
If I am confused, ignorant, or wrong, please gently enlighten or correct me.
Tim McDaniel
Usenet: ...{pur-ee|ihnp4|convex}!uiucdcs!mcdaniel
Csnet: mcdaniel%uiuc@csnet-relay.arpa (really!)lmaher@uokvax.UUCP (10/24/84)
> /***** uokvax:net.news / t4test!chip / 6:23 pm Oct 22, 1984 */ > As I have seen, three features have been questioned: > 2) Including original articles in a followup with '>' marks. > > I agree with the gripes behind the second point. That is why I have > added a '-f' flag to surpress the original article in a followup. > Also, a '-m' flag may define what mark you want. Including the > original article in a followup is an important feature. It should > remain there, but a method for getting around it (other than deleting > lines in the editor) should also be available. I think this is exactly backwards. The flag should be required in order to include the original article, not to suppress it. It's the inexperienced users that don't know about flags. I've already seen several cases of entire articles being quoted verbatim by inexperienced users. More often than not with a one-line comment as well. Notes has the 'w' command to write a response, and the 'W' command to write a response and include the text, without the '>' marks. A user who can't do a 1,$i/> shouldn't be in the editor anyway. :-) This brings up the entire topic of "how do new users learn to use the system?" Perhaps the netiquette file posted to net.announce.newusers should be readable with a news command, like help. At the *very* least, new users should have the newuser group pointed out to them - many sites don't do this, so why not have the news software check to see if this is their first time ever to read news, and if so direct them to the documentation. Perhaps with a short quiz afterwards. :-) Carl {allegra,ihnp4}!convex!ctvax!uokvax!lmaher