Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!wuarchive!julius.cs.uiuc.edu!apple!vsi1!zorch!xanthian From: xanthian@zorch.SF-Bay.ORG (Kent Paul Dolan) Newsgroups: comp.sources.d Subject: Re: Let's have ONE standard lharc! Message-ID: <1991Jan19.155625.4061@zorch.SF-Bay.ORG> Date: 19 Jan 91 15:56:25 GMT References: <767@grapevine.EBay.Sun.COM> <20609@yunexus.YorkU.CA> <10765@hydra.Helsinki.FI> Organization: SF-Bay Public-Access Unix Lines: 59 koreth@panarthea.EBay.Sun.COM (Steven Grimm) writes: > The lharc program is really nice (though a bit slow) and I'd like to > see it become a more widespread standard. oz@yunexus.yorku.ca (Ozan Yigit) writes: > I thought that the prevailing standard (at least on non-un*x systems) > was zip. Not really widely portable, and, my impression was, tied in with a commercial or shareware product so that it is difficult to port widely, legally. None of the platforms I've used in the last three years had zip file support available. wirzeniu@cs.Helsinki.FI (Lars Wirzenius) writes: > Why not use zoo as a standard, it is already quite portable, though > it's compression isn't all that wonderful. Work on this is supposed to > be on the way, however. Problem is, lharc compression is _much_ better than zoo's, and even though it is slower, you pay for the extra storage all the time, the extra time only occasionally when packing or unpacking the data. The economics of gaining an average 20% better compression is just too compelling to ignore, which is why lharc is spreading so fast. Zoo has _many_ advantages: centralized source control, an excellent interface, identical as possible across platforms, near universal availability, several bells and whistles (comment fields, e.g.), and good speed. None of these protect it's existing hold on archiving across platforms from lharc's economic value, which is why many sites I know are taking the time to repack zoo's as lzh's; just as arc's got repacked as zoo's several years ago. > Anyway, one portable standard would be nice, it's a pain in the *ss to > have to keep up with the latest versions of zillions of archivers. It > is of less importance exactly what this standard is, as long as it's > reasonably effective and free (it's stupid to choose a commercial > standard, since there are several good PD ones). Amen on the commercial standard; not only expensive, but more important, too slowly ported to new platforms and rarely available in source form. However, the rapid rush to lharc is precisely due to its effectiveness; archiving is one field where technical excellence is everything. What is needed for lharc is for a single individual with good net and platform access who will take responsibility for melding a single version with a single interface and a well understood format, and for the authors of all the various versions to voluntarily surrender source code to be built into this single, #ifdefed version. Neither the Unix versions author nor the several authors who have ported the other platform's versions have a lot of net presence, so this could be a tough search. Kent, the man from xanth.