Xref: utzoo news.software.b:6272 news.software.nn:1345 Path: utzoo!utgpu!watserv1!watmath!att!att!linac!uwm.edu!caen!sol.ctr.columbia.edu!ira.uka.de!smurf!urlichs From: urlichs@smurf.sub.org (Matthias Urlichs) Newsgroups: news.software.b,news.software.nn Subject: Re: REPLY: Why use NN when there is TRN??? Message-ID: Date: 27 Nov 90 21:43:09 GMT References: <1990Nov23.160854.13868@vlsi.polymtl.ca> <14@grumbly.UUCP> <2867@gazette.bcm.tmc.edu> Organization: University of Karlsruhe, FRG Lines: 27 In news.software.b, article <2867@gazette.bcm.tmc.edu>, sob@tmc.edu (Stan Barber) writes: < < The biggest problem with nn and trn is the fact that they (currently) have < incompatible thread databases. In fact, it would be a pretty neat idea < to build some kind of standard interface for doing this kind of thing < so that it could be ported to more news readers or architecture other < than Unix! We might be able to move some of this proposed standard < interface back into NNTP. However, Unix NNTP has enough problems trying to < cope with the differences in the two BACK-end (CNEWS and BNEWS) to be < mutilated into supporting two different threading architectures. < It might help to have at least a somewhat-standard way to get (part of?) a thread file/database from the server. Something like... C> DBASE TRN news.software.b [section] [range] [whatever] S> 3xy 3628 bytes follow S> [ sends 3628 bytes of binary data ] S> 2yz Got that? As it is now, trn users must have a local database (unless they can use NFS -- I dunno about NN), and each host which runs trn ends up reading every article. Not good. -- Matthias Urlichs -- urlichs@smurf.sub.org -- urlichs@smurf.ira.uka.de /(o\ Humboldtstrasse 7 - 7500 Karlsruhe 1 - FRG -- +49+721+621127(0700-2330) \o)/