Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!execu!sequoia!rpp386!jfh
From: jfh@rpp386.cactus.org (John F. Haugh II)
Newsgroups: comp.sys.ibm.pc.rt
Subject: Re: RS/6000 survey - will post summary
Message-ID: <18287@rpp386.cactus.org>
Date: 8 May 90 12:19:55 GMT
References: <11111@bsu-cs.bsu.edu> <50279@lanl.gov> <1990May4.222715.23648@cs.utk.edu> <10036@stiatl.UUCP> <1990May6.212646.20580@cs.utk.edu> <1698@ks.UUCP>
Reply-To: jfh@rpp386.cactus.org (John F. Haugh II)
Organization: Lone Star Cafe and BBS Service
Lines: 67

In article <1698@ks.UUCP> drake@ibmarc.uucp (PUT YOUR NAME HERE) writes:
>1.  If you want to see old terse messages, issue:
>
>    export LANG=C
>
>    The new design allows old hands such as yourself to be happy,
>    while making a system that real people can use as well.  This is
>    bad?  

this is far from universally true.  if you look in /usr/lpp/msg/C
you won't find many messages, and many of the developers removed
the standard messages from the source code while changing the code
to support NLS.

so you have a mix of old-style "terse" messages, and newer NLS-ified
messages from the message catalogs.  many of the commands, however,
have exactly one set of messages - the built in messages have the
same text as the ones in the catalogs.

>2.  Who decides when an "old terse message is sufficient"?  Were they
>    sufficient to secretaries?  Or just to hackers?  

most secretaries run vertical applications or integrated packages
such as uniplex or q-office or some other application on top of
raw unix.  i don't think making unix more friendly is going to help
the unix illiterate.  how many ibm'ers still rely heavily on profs
and don't both with vm or cms directly?

>3.  The AIX 3 enhancements in this area were primarily dictated by
>    (a) usability and (b) national language support requirements.
>    The new system is much more flexible in that it can be marketed in
>    languages other than English ... a legal requirement in a growing
>    number of countries.

absolute agreement here.  and anyone looking to start a new business
in the s/6000 market should consider third party message translation
as an oppurtunity.  if someone can come up with a tool with undoes
a "gencat" command, you could un-gencat the catalogs and come up with
new text messages.

>But then you're a native English speaker, used to hacker-ese, and
>are quite happy with software whose limitations are therefore not apparent to
>you.

yes, this is a big win, and the architecture is well done.  but this
is the first approximation and there are some problems.

but every language needs a verbose message set and a terse message
set, not just english.  i predict there will be abundant oppurtunities
in other countries for unix hackers to create concise message sets.
there is value in being concise, you know ...

>      Nor, apparently, does your installation want C2 security in it's
>operating system, which neither UCB nor AT&T ship today.  

yes, you're quite correct.  but sun ships B2 and AT&T ships B1.  ibm
ships security as well, but 3.1 isn't rated just yet.  i read the
AT&T System V/MLS evaluation and wasn't impressed.  i've yet to see
Sun's [ and in fact, didn't see them listed in the catalog of final
evaluation reports - are they final yet? ]

anyone who is interested in security will have a level in mind -
just being B2 or B3 does not imply one is better than all C2 or B1
systems.  it just means the system is B2 or B3.
-- 
John F. Haugh II                             UUCP: ...!cs.utexas.edu!rpp386!jfh
Ma Bell: (512) 832-8832                           Domain: jfh@rpp386.cactus.org