biep@klipper.UUCP (J. A. "Biep" Durieux) (01/02/85)
In article <6719@brl-tgr.ARPA>, geoffs@brl-tgr.ARPA (Geoffrey Sauerborn (TANK) <geoffs>) writes: >In article <24@epsilon.UUCP> you write: >>In article <393@klipper.UUCP> biep writes: >>>>> From: schu@drutx.UUCP (SchulteSM) >>>>> Subject: Re: A Modest Proposal >>>>> >>>>> >>>>> >>>>> q >>>>> >>>>> >>>>> w >>>>> . >>>>> dfslkfl; >>>>> >>>>> >>>>> lkfdg >>>>> ;lfdkg;lka;dlgk >>>>> >>>>> >>>>> >>>>> ZZ >>>>> agkfjlkadfgkj >>>>> >>>>> >>>> >>>>Yes, I agree. This is definately a proposal worth considering. >>>> >>> >>>No, no, no!!! >>>I know, it sounds nice, but believe me, it just doesn't work. >>>Someone over here brought this up on our local network, and we've >>>been discussing it over and again, and indeed, it sounds appealing, >>>but when you work out the details you'll find lots of inconsistencies >>>and problems. Don't try to reinvent the wheel, there are already >>>enough other proposals on this net which ask for our attention. >>> >>Well, we've actually done it here, and it works out just fine. Of course, >>we had to make certain minor changes like "DfXlkfl;:" instead of >>"dfslkfl;" as was indicated in the original article. You'll note that >>otherwise a contradiction arises when "ZZ" is implemented. >> >> Ed Sheppard >> Bell Communications Research > > > Alright! I'm sick of this! Why is it that every time someone >puts a good piece into public domain, somebody has to good and change >the source! The next thing to happen it some BO-ZO will try to use >this changed version without taking the time to read the documentation - >and naturally the FOOL starts flaming to the original poster!!!! > > > Geoff Sauerborn You all don't seem to get the point. Indeed, Ed, it is possible to adapt the thing to your local network, when all machines are compa- tible, none is running notesfiles, all are little-endian, and at least some of them are not feeding news to decvax, mcvax or purdue (and per- haps others, I've not yet got time to find out), and some other little things. *But not all sites of USENET do!!!* And what is happening then is that everybody starts making his local patches, and we end up even worse than we started. The whole mistake is due to the starting "q", which supposes a local "edsgr w534cb67835", since otherwise indeed "ZZ" goes wrong in boundary cases. I must say I do not yet completely understand Ed's patch, it occurs to me that this only works because of some local changes to the rot13 algorithm. That can hardly be called "portable", can it? Geoff, I think your argument goes wrong at the word "good". I would say: quit the idea, the gains for sf-movie-lovers do not outweigh the burden for the rest of us. -- Biep. {seismo|decvax|philabs}!mcvax!vu44!botter!klipper!biep I utterly disagree with everything you are saying, but I am prepared to fight to the death for your right to say it. --Voltaire
ken@boring.UUCP (07/24/85)
References:
Sender: ken@mcvax.UUCP (Ken Yap)
Reply-To: ken@mcvax.UUCP (Ken Yap)
Followup-To: net.news
Distribution: net
Organization: Centrum voor Wiskunde en Informatica, Amsterdam
Keywords:
[There are more flames in net.news than net.flames now :-)]
Amidst the net.flame brouhaha, I read in mod.newslists that this week
net.sources.mac accounted for 10% of total net traffic thru seismo,
twice that of the next volumous newsgroup.
Curious as to how much redundancy remained in Mac Binhex postings, I
tried the following script on 4 large articles (actually Rick Jansen's
TeX stuff).
tr -d '\012' < hex | compress | encode | fold -72 > newhex
and obtained reductions between 15 and 30%. (Encode is not uuencode
but a program by Robert Elz (?) to map 8 bit data to 7 bit printable
ASCII.)
The reverse transformation would be:
tr -d '\012' < newhex | decode | uncompress | fold -64 > hex
Questions: It it worth applying this compaction to postings to reduce
net traffic? Would it cause problems? Is there already a new Binhex
format that optimizes packing?
Problems I can see: Everyone would need compress (which is PD),
encode/decode (which if not PD are simple enough to recreate) and fold.
This format would be sensitive to small errors but Binhex already is
anyway. Lines starting with ~ could upset mailers. Some links already
apply compression to net news.
Comments welcome. Flames to /dev/null.
Regards,
Ken
--
UUCP: ..!{seismo,okstate,garfield,decvax,philabs}!mcvax!ken Voice: Ken!
Mail: Centrum voor Wiskunde en Informatica, Kruislaan 413, 1098 SJ, Amsterdam.