Return to BSD News archive
Newsgroups: comp.os.386bsd.development
Path: sserve!newshost.anu.edu.au!munnari.oz.au!news.Hawaii.Edu!ames!agate!doc.ic.ac.uk!uknet!mcsun!sunic!aun.uninett.no!trane.uninett.no!alf.uib.no!usenet
From: bav@bergen.siodata.no (Bjorn Asle Valde)
Subject: 7bit unclean is/was/will-be considered harmfull
Message-ID: <1993Jun15.201541.16351@alf.uib.no>
Sender: usenet@alf.uib.no (Bergen University Newsaccount)
Reply-To: Bjorn_Asle.Valde@bergen.siodata.no
Organization: University of Bergen, Norway
References: <1vi6ak$hs7@umd5.umd.edu>
Date: Tue, 15 Jun 93 20:15:41 GMT
Lines: 55
In article <1vi6ak$hs7@umd5.umd.edu> mark@elea.umd.edu (Mark Sienkiewicz)
writes:
> In article <1993Jun14.081754.18248@alf.uib.no>
Bjorn_Asle.Valde@bergen.siodata.no writes:
> >
> >Huh...? For political reasons, IMHO, *all* and every driver/prog
> >in any usable un*x like system should be 8bit-clean. If that's
> >a problem, dump the device or rewrite the program that gives
>
> Nobody said "Make the system 7 bit only." You should make the system
> 8 bit clean in such a way that it can still be set into a 7 bit mode.
I agree, I also never implied that anyone said "make it 7-bit only".
> For you to suggest that I should throw away hardware that works
perfectly
> fine just because _you_ don't like it is absurd. It would make as much
> sense for me to suggest that you throw away your keyboard because it
> doesn't use 7 bit even parity.
Did I suggest that - really? :-) Read between the lines now:
Don't make drivers that accepts 7-bit *only*
unless there's a __very__ good reason for it.
Thanks!
The way I read Peter da Silva (<C8MHor.H2L@sugar.NeoSoft.COM>) and
Michael Sanderman (<1993Jun15.051625.29309@Sandelman.OCUnix.on.ca>)
seems to indicate that the whole 7/8bit driver thing is under
control. Fine.
> >you a hard time. All now: it's 1993, 8-bit clean is the minimum
> >requirement. (At least in principle, I'm not to rewrite all
> >the old stuff, but as a matter of policy: yes!)
>
> I have no problem with the idea that programs shouldn't blindly assume
that
> the 8th bit is 0, or available for use, or insignificant. Just like you
> would like that, I would like the system to not ASSUME there is meaning
in
> it.
>
> Mark S.
>
> p.s. Where is our 16 bit character support???? :)
Interesting point! Actually, there are nifty work going
on that makes the point that the *interpretation* must be
seperated from the actual representation (read: charset encoding).
But of course, that has really very little to do with the 7/8bit
driver discussion so I'll just stop. Have a nice summer.
Mvh, -bav
--
Bjorn Asle Valde - SiO Data Bergen, Christiesgt.13, N-5015 Bergen, Norway.
Email: valde@bergen.siodata.no Phones: +47-5-320240 Fax: +47-5-320238