chip@tct.uucp (Chip Salzenberg) (05/03/90)
According to Chris.Holt@newcastle.ac.uk (Chris Holt): >What is going to happen to news groups when the load factor increases by >10 (<5 years, when more universities allow their students access) and >100 (<15 years?, when more non-computing people gain access) and >1000 (<25 years?, when the rest of the world starts linking in)? Long before that time we'll have better newsreading tools widely available. One that's almost out is "trn", which is a threaded version of rn. Written by Wayne Davison, trn shows the discussion tree in graphical form on the upper-right of the article display. It also has a thread selection mode, where specific threads can be read and the rest dropped. In my opinion, newsgroup proliferation is a cosmetic solution to the basic problem of article classification. I'd rather program my newsreader than try to change the Usenet namespace. The latter requires great effort for an unknown benefit, while the former offers immediate feedback and utility. >Question: is it worth providing a forum for discussion of what shape >the net should take? Such a thing is needed. The last attempt failed; the mailing list became an argument magnet, just like news.groups. Who will try again? -- Chip Salzenberg at ComDev/TCT <chip%tct@ateng.com>, <uunet!ateng!tct!chip>
randy@rls.UUCP (Randall L. Smith) (05/03/90)
In article <1990May1.221156.20220@newcastle.ac.uk>, Chris.Holt@newcastle.ac.uk (Chris Holt) writes: > What is going to happen to news groups when the load factor increases by > 10 (<5 years, when more universities allow their students access) and > 100 (<15 years?, when more non-computing people gain access) and > 1000 (<25 years?, when the rest of the world starts linking in)? This "threat" has been around for a long time and it is being handled by subdivision and alteration of additional nets. e.g. Vmsnet, Bitnet, Fido, Clari? :-), as well as commercial nets. Besides, don't you believe horsepower and transmission speeds will increase? :-) I know, you're talking about "How will we ever read it all?". In a phrase, we won't. Me, I'm worried about Alpha Centuari linking in. Yow, their yakkity. > Claim: centralized algorithms are not going to work. A single, > overseeing news.groups cannot cope, so each node will have to have its > own .groups to deal with fissions. The traffic is hardly overwhelming in comp.i386. > Claim: centralized unmoderated groups are not going to work. There When the traffic gets there, we'll know it and splinter. Probably by, experience level, not variations of a theme. The expectations of this centralized newsgroup is that the Intel architecture Unix is merging to the peculiar (System V / BSD) standard at all levels and will stay there. Wherever there is... > Question: is it worth providing a forum for discussion of what shape > the net should take? Clearly people have (perhaps ill-formed) images That is what news.groups is about. This process has taken on the form of a quick and dirty program (apologies to Mark Horton and other pioneers) that continually gets patched from a variety of programmers with a variety of styles. When will we sit back and actively design its future? Who knows? You are welcome to be the messiah that save us from collapse. BTW, the comment re: quick and dirty is the net was never designed or expected to be what is today. Kudos to all involved since it does handle it nicely anyway. > Just a thought. Same here. No flames intended, etc. > "There is a holy mistaken zeal in programming as well as in religion..." Please add, network architecture too. ;-) Cheers! - randy Usenet: randy@rls.uucp Bangpath: ...<backbone>!osu-cis!rls!randy Internet: rls!randy@tut.cis.ohio-state.edu Disclaimer: Oh, am I gonna get flamed for this one... Right Brad?