home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!portal!cup.portal.com!Rick_Michael_Cortese
- From: Rick_Michael_Cortese@cup.portal.com
- Newsgroups: comp.sys.atari.8bit
- Subject: Re: My ego VS UUE
- Message-ID: <72473@cup.portal.com>
- Date: Sun, 27 Dec 92 10:57:31 PST
- Organization: The Portal System (TM)
- References: <71961@cup.portal.com> <kJ98VB2w165w@cyberden.sf.ca.us>
- <72423@cup.portal.com>
- Lines: 37
-
- RE: Point #1, I'll stand by my 1.5 expansion statement. Matter of fact, some
- of the files I D/L'd were that were in that 1.5 range were done by you. It's
- more then just a problem with the encoding scheme. I believe the UUE uses a
- 46 byte line with a two byte checksum for every line which would pump it
- up, the rest of the expansion I'm not sure about, but it DOES make the file
- 1.5 times the original. IMO: I'm sure the original authors had something in
- mind by adding a checksum to every line, but I really doubt if it serves a
- useful purpose anymore. Since most everything is transfered via Xmodem etc.
- it's more likely a bad file to begin with. A checksum for the entire file
- makes a lot more sense to me & apparently the authors of ARC, ZIP, ...
-
- I'm aware of the bias & plan on the numbers being represented by ASCII
- characters from !-r or !-v, something in that range anyway. No control
- characters, no extended characters. I've already sent a file to myself
- on another BBS (ricortes@kludge.com) & I'll double check that everything
- gets through correctly. It shouldn't matter what character sets another
- computer uses *IF* (note the big if) they just echo the file as is. Who
- cares if computer X thinks a 65 in a file means "strip and go nude", as
- long as it echos the 65, were fine.
-
- Point #3 Speed, I've already got a mutiple precision binary division
- routine that'll go to 255 significant figures for the operand/divisor.
- I'm planning on eventually getting it working as an Action! procedure
- along with maybe some multiplication. It may take me a while to get
- everything neatly done, but as long as I'm willing to code in M/L, you
- shouldn't worry about speed. If your really worried, I could run some
- tests, but even if there is a net loss in speed on the decode end, the
- online time/transmission/echos for the entire Internet system will be
- reduced by 15% (Only for files of course!). Who wouldn't take a slightly
- longer decode time in exchange for a shorter down load time?
-
- BTW: I want to thank Douglas Clement for correcting my terminology on
- number base systems. My excuse is I was trained as a chemist so...
-
- I guest for the doubters, I should do a little demo. Turns out Atari
- Basic can just manage to do the job albeit slowly. I'm sure somebody
- about :-)
-