srt@ucla-cs.UUCP (08/13/85)
How about adding a feature to 'rn' that will do some checking to see if the current article has been replied to. Two ways to do this come to mind: (a) Simply check forward through the newsgroup looking for an article that references the current one. (b) Check forward through the newsgroup for an article that references the current one _and_ includes a user-supplied string. The idea is to prevent hundreds of answers to simple queries (a major problem as we all know). //Is he going to pop into his DeLorean again?// -- Scott
lwall@sdcrdcf.UUCP (Larry Wall) (08/15/85)
In article <6562@ucla-cs.ARPA> srt@ucla-cs.UUCP writes: >How about adding a feature to 'rn' that will do some checking to see if >the current article has been replied to. Two ways to do this come to >mind: >... >The idea is to prevent hundreds of answers to simple queries (a major >problem as we all know). > >//Is he going to pop into his DeLorean again?// [Oh, oh. I can't seem to get the silly thing started...] This idea has some merit. It would have more merit if news were transmitted instantaneously. Unfortunately, most duplicate followups occur because of transmission latency, not because people don't read everything that's come in. However, there are a couple of ways to handle the problem, such as it is. First, you can do what I do--use the 'M' command to mark the article temporarily read, then read the rest of the newsgroup, and then either use the 'Y' command to yank back the article, or exit the newsgroup and return, which automatically yanks it back. Or, if you are in subject search mode, and the next article is a different subject, use '-' to get back to the last article of the subject you're interested in. The 'M' approach has the advantage of giving a "cooling-off" period. Often, by the time the article comes around again, I've decided I don't really want to jump into the fray. Alternately, we could probably define a macro, bound to 'f' or 'F', that would list all the unread articles with the same subject, ask for confirmation, and then do the real f or F. The actual macro is left as an exercise for the reader. I prefer to drag my feet on adding more C code, though, when an issue can be resolved by education. And in the case of transmission latency, education is the only near-term solution. There's no software that will suppress duplicate "NEWS FLASH: OLD COKE IS BACK!" articles. Larry Wall {allegra,burdvax,cbosgd,hplabs,ihnp4,sdcsvax}!sdcrdcf!lwall