caf@omen.UUCP (Chuck Forsberg WA7KGX) (04/17/87)
I have some wishes for goodies in the new version of that cupsy newsreader, RN. 1. An option to display the article titles (after kill processing) when entering a new newsgroup without having to type "=' as a separate step. 2. In the avove display, show each subject title only once, possible indicate the number of srticles with the same subject. 3. After reading an article, Rn currently searches for a followup or something related. If I select an article with a "/string" command, it would be nice for rn to continue that search automatically. 4. An intelligent Subject grabber that would bypass the spurious "Submission to mod.foo" subject that serves only to enhance the moderators' egos and display something useful such as the "real" Subject line 5. The readnews feature I miss above all is intelligent handling of digests. (Maybe #5 could be handled by starting "less" on the raw article?)
fair@ucbarpa.Berkeley.EDU.UUCP (04/18/87)
In article <522@omen.UUCP> caf@.UUCP (Chuck Forsberg) writes: >5. The readnews feature I miss above all is intelligent handling of digests. The proper way to handle digests is not to have them at all. They break all the netnews readers we have. They also usually have incorrect, or unusable return addresses in them, useful only to the mailer at the site that constructed the digest in the first place. Digests are a form of message batching that the USENET has never needed, because we put our message batching in the netnews transport, invisible to the users. DEATH to digests! Erik E. Fair ucbvax!fair fair@ucbarpa.berkeley.edu Chairman of the USENET Committee to Stamp Out Digests
kre@munnari.UUCP (04/18/87)
In article <18445@ucbvax.BERKELEY.EDU>, fair@ucbarpa.Berkeley.EDU (Erik E. Fair) writes: > The proper way to handle digests is not to have them at all. Nonsense. > They break all the netnews readers we have. That's a problem with the news readers, readnews does a fairly good job of handling them. The others could. > They also usually > have incorrect, or unusable return addresses in them, useful only to > the mailer at the site that constructed the digest in the first place. This is a totally specious argument. The From: address in any usenet article is constructed at the initiating site, and is no more likely to be correct or incorrect at any particular reader's site than the From: address in a digest. I know that Erik isn't seriously proposing that the Path: header in usenet articles is relevant to this. > Digests are a form of message batching that the USENET has never needed, If you view a digest as being solely a transport hack then I agree. (Though "never" is a bit much, usenet didn't always have batching). And indeed, some digests are no more than that .. when Erik got the moderator of the Telecom digest to send the raw data instead of the digested form I didn't mind at all, that digest had no advantages. However, other digests are much more than the sum of their articles, when AI-LIST started being sent to usenet undigested I unsubscribed within a day or two, before then I had at least glanced at each issue. I would hate to see RISKS or a few of the others split, SUN-SPOTS and a bunch of others like that I wouldn't mind at all. kre
chapman@eris.BERKELEY.EDU (Brent Chapman) (04/18/87)
In article <1568@munnari.oz> kre@munnari.oz (Robert Elz) writes: >In article <18445@ucbvax.BERKELEY.EDU>, fair@ucbarpa.Berkeley.EDU >(Erik E. Fair) writes: >> The proper way to handle digests is not to have them at all. >If you view a digest as being solely a transport hack then I agree. >And indeed, some digests are no more than that >However, other digests are much more than the sum of their articles, >when AI-LIST started being sent to usenet undigested I unsubscribed >within a day or two, before then I had at least glanced at each issue. >I would hate to see RISKS or a few of the others split, SUN-SPOTS >and a bunch of others like that I wouldn't mind at all. In general, I tend to agree with Erik on issues relating to news and mail and networking and such, but this time, I must disagree, for about the same reasons as Robert Elz. There are certainly digest that are much more than just a dump of the moderators inbox every week. The example with which I'm most familiar is the RISKS digest; Peter G. Neumann consistently does an excellent job of moderating the discussions, keeping them on track, and killing them when they begin to lose relevance (even if he does have an abhorrent penchant for puns :-). I think part of the problem may be that "digests" are an artifact of the internet; many (most?) of the "digest" newsgroups are actually internet mail digest gatewayed into USENET (much of that gatewaying is done at ucbvax, here at Berkeley, and maintained by Erik, so I'm sure he's intimately aware of all this). Looked at from the standpoint of the internet, digests make a lot of sense. Without them, you end up with lots of little, mostly irrelevant, messages in your mailbox every day. The question is, should these internet digests be "undigested" at whatever point they are gatewayed into the USENET? I'd have to say that's a decision to be made on a digest-by-digest basis; as pointed out earlier, I certainly wouldn't want this to happen to RISKS; others, I'm not so sure about. Brent -- Brent Chapman chapman@mica.berkeley.edu or ucbvax!mica!chapman
mangler@cit-vax.Caltech.Edu (System Mangler) (04/26/87)
The primary impetus for this site joining Usenet was that I was sick of digests. I used to get Unix-wizards by ARPA mail, and that was ok. Most articles I would read once and discard, keeping the good ones for later. I could skip over long articles, and stop reading after any article, without losing my place. When it became a digest, I was presented with 1000 lines to read (without skipping) in one sitting, and if I wanted to come back to any article, I had to go through the whole digest. At 1200 baud. (I read news over breakfast - it sure beats the back of the cereal box). I learned to detest digests so much that I got us onto Usenet, against everybody's recommendations, just to avoid them. In article <1568@munnari.oz>, kre@munnari.oz (Robert Elz) writes: > readnews does a fairly good job of handling them. I use 2.10.3 readnews, and it does not. It has simplistic notions of what starts an article, and is often fooled. There is no reply or followup command. There is no erase (mark unread) command. The save command doesn't properly interpret $NEWSBOX. The only thing it provides is moving around - something any text editor can probably do better. > I would hate to see RISKS or a few of the others split, SUN-SPOTS > and a bunch of others like that I wouldn't mind at all. SUN-SPOTS is the worst. It has a delay measured in weeks, making discussion impossible. (Which is probably the main purpose of digests). Most of each digest is requests for information, with few articles offering any answers, and typically those have already appeared in other newsgroups. This is quality? Sign me up, Eric. Death to digests! Don Speck speck@vlsi.caltech.edu {seismo,rutgers,ames}!cit-vax!speck
robert@jimi.cs.unlv.edu (Robert Cray) (04/27/87)
In article <2481@cit-vax.Caltech.Edu> mangler@cit-vax.Caltech.Edu (System Mangler) writes: >I used to get Unix-wizards by ARPA mail, and that was ok. Most articles >I would read once and discard, keeping the good ones for later. I could >skip over long articles, and stop reading after any article, without >losing my place. When it became a digest, I was presented with 1000 >lines to read (without skipping) in one sitting, and if I wanted to >come back to any article, I had to go through the whole digest. So just use MH and do a burst -- this seems to work fairly well, you can reply to messages just as if they were sent directly to you, you can read messages in an entirely haphazard way, then do a "scan unseen" to see which messages you have not read yet. Also, you can have such things automatically refiled into their own folders before you ever see them, so they dont clutter up your mailbox, and so you can keep track of what was really sent to you, and what was sent to you because you are on a mailing list. --robert -- CSNET: robert%jimi.cs.unlv.edu@relay.cs.net UUCP: {akgua,ihnp4,mirror,psivax,sdcrdcf}!otto!jimi!robert
dean@violet.berkeley.edu.UUCP (04/28/87)
In article <554@jimi.cs.unlv.edu> robert@jimi.UUCP (Robert Cray) writes: >In article <2481@cit-vax.Caltech.Edu> mangler@cit-vax.Caltech.Edu (System Mangler) writes: >> When it became a digest, I was presented with 1000 >>lines to read (without skipping) in one sitting, and if I wanted to >>come back to any article, I had to go through the whole digest. > >So just use MH and do a burst -- this seems to work fairly well... > --robert I suspect (though I don't know) that this tool suffers from the same problem that most news-mungers have: it can't help those of us who read news kept on a remote machine. To save having the news stored on every machine in sight, we keep news on one machine and use network connections to read it (with a special version of rn). Digests are doubly annoying in that case because we a) have a significantly long pause... while a digest article is shot over to our current machine; and b) once it gets there we (I, at least) have to paw through it at 1200 baud. I don't have any religious feelings one way or the other about the worthiness of digests, I just ask that those considering solutions keep in mind those of us who read mail remotely and slowly, and therefore would like a chance to see the headers of individual items before screening them. -Dean (dean@violet.berkeley.edu)
wayne@fmsrl7.UUCP (Michael R. Wayne) (05/01/87)
I would LOVE an option to shut off signature display so I would not
have to look at those 55 line "cute" pictures any more!
--
Michael R. Wayne Voice: (313) 322-3986 | philabs \
Working at (but not employed by) Ford Motor Company | pyramid !fmsrl7!wayne
| ihnp4!mb2c / lwall@sdcrdcf.UUCP (Larry Wall) (05/06/87)
In article <876@fmsrl7.UUCP> wayne@fmsrl7.UUCP (Michael R. Wayne) writes: > I would LOVE an option to shut off signature display so I would not > have to look at those 55 line "cute" pictures any more! Try -EPAGESTOP='^-- ' or something like that. Then you just have to watch for '-- ' at the end of the page and hit 'n'.. This presumes your rn is at patchlevel 4. Larry Wall {allegra,burdvax,cbosgd,hplabs,ihnp4,sdcsvax}!sdcrdcf!lwall
wtm@bunker.UUCP (Bill McGarry) (05/06/87)
In article <876@fmsrl7.UUCP> wayne@fmsrl7.UUCP (Michael R. Wayne) writes: > > I would LOVE an option to shut off signature display so I would not >have to look at those 55 line "cute" pictures any more! This is not exactly what you are looking for but it may help. (This is from the section on environment variables in the "rn" man page.) PAGESTOP If defined, contains a regular expression which matches article lines to be treated as form-feeds. There are at least two things you might want to do with this. To cause page breaks between articles in a digest, you might define it as "^--------". To force a page break before a signature, you could define it as "^-- $". (Then, when you see "--" at the bottom of the page, you can skip the signature if you so desire by typing 'n' instead of space.) To do both, you could use "^--". If you want to break on more than one pattern, you can use "|" to separate the alternatives. There is some overhead involved in matching each line of the article against a regular expression. You might wish to use a baud-rate modifier to enable this feature only at low baud rates. Default: undefined Bill McGarry PATH: {philabs, decvax, fortune, yale}!bunker!wtm
ram@nucsrl.UUCP (Renu Raman) (05/07/87)
How about an option for "newsaholics". Monitor reading frequencies
of groups and articles and either cut off access to that group beyond
a threshold value or warn root as well as reader about the habituation.
Also, I don't if "rn" has this feature, but notes does not. String
search within articles would be an useful feature. many times, while
reading long articles or searching for a topic/subject that was discussed
would be greatly helped by this feature.
-------------------
Renu Raman UUCP:...ihnp4!nucsrl!ram
1410 Chicago Ave., #505 ARPA:ram@eecs.nwu.edu
Evanston IL 60201 AT&T:(312)-869-4276 page@ulowell.cs.ulowell.edu (Bob Page) (05/09/87)
ram@nucsrl.UUCP (Renu Raman) wrote in article <3230001@nucsrl.UUCP>: >Monitor reading frequencies of groups and articles and either cut off access >to that group beyond a threshold value or warn root as well as reader ... Can you say Big Brother? I knew you could. ..Bob -- Bob Page, U of Lowell CS Dept. page@ulowell.{uucp,edu,csnet}