ins_akaa@jhunix.UUCP (Kenneth Adam Arromdee) (12/02/85)
In article <81@calma.UUCP> radzy@calma.UUCP (Tim Radzykewycz) writes: >> 1/2 disk per site * 4000 sites * $8000/disk= >>$16000000 wasted by the net on blank disk space, or $2000 PER SITE. >Either I've forgotten most of my upper-elementary-school level algebra, >or you did something wrong. >I think this makes much of the original posting invalid. If Kenneth >makes this kind of mistakes so easily, I wouldn't trust his verbal >arguments much either. In fact, I think his arguments aren't valid >anyway: OK, I goofed. $4000 per site. However, saying you wouldn't trust a verbal argument of mine, not because you have found anything wrong with the verbal argument, but simply because I once made an arithmetical error, does not itself seem to be a very valid verbal argument. >1. many of the sites which are on USENET wouldn't have >the same disk configurations if they weren't on it. I can only go by what you said, which included no evidence for this. >2. The assumption that the average amount of disk space which is >wasted is 50% of one disk is stupid. (no ":-)" here) Assuming it's evenly distributed between 0.000000<n decimals>1 % and 99.999... %, 50% is the average. It's only an order of magnitude calculation anyway. OK, let's say it's 90%, not 50%. Then the cost to the net is .1 disk/site * 4000 sites * $8000/disk=$3200000 for the net. The point is that calculating the disk space "wasted" by Rich Rosen is meaningless because it's swamped by the "wasted" disk space which would have been there anyway, but now holds Rich Rosen's (or any other user's) postings. >3. Kenneth ignored (or perhaps didn't understand) the idea behind >the original posting, which was to say that people should constrain >themselves to posting only original material. >Tim (radzy) Radzykewycz I was well aware of this; I was just responding to one fallacious argument. -- If you know the alphabet up to 'k', you can teach it up to 'k'. Kenneth Arromdee BITNET: G46I4701 at JHUVM and INS_AKAA at JHUVMS CSNET: ins_akaa@jhunix.CSNET ARPA: ins_akaa%jhunix@hopkins.ARPA UUCP: ...{decvax,ihnp4,allegra}!seismo!umcp-cs!aplvax!aplcen!jhunix!ins_akaa ...allegra!hopkins!jhunix!ins_akaa