*BSD News Article 41514


Return to BSD News archive

Xref: sserve comp.os.386bsd.misc:5052 comp.os.linux.misc:33967 comp.os.os2.advocacy:75857
Path: sserve!newshost.anu.edu.au!munnari.oz.au!constellation!bubba.ucc.okstate.edu!news.ksu.ksu.edu!vixen.cso.uiuc.edu!howland.reston.ans.net!math.ohio-state.edu!caen!usenet.cis.ufl.edu!anshar.shadow.net!anshar.shadow.net!cdlevin
From: cdlevin <cdlevin@shadow.net>
Newsgroups: comp.os.386bsd.misc,comp.os.linux.misc,comp.os.os2.advocacy
Subject: Re: Linux thoroughly insulted by Infoworld!
Date: Mon, 23 Jan 1995 19:51:36 -0500
Organization: Shadow Information Services, Inc.
Lines: 93
Message-ID: <Pine.SUN.3.91.950123192551.1788B-100000@anshar.shadow.net>
References: <950116203411@lambada> <1995Jan18.214037.6088@cs.cornell.edu> <3fk4hi$iu8@solaris.cc.vt.edu> <D2pK11.EJx@madge1.demon.co.uk> <3fpk32$80i@unix.sri.com> <3fqh6a$sl@manuel.anu.edu.au>
NNTP-Posting-Host: anshar.shadow.net
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
In-Reply-To: <3fqh6a$sl@manuel.anu.edu.au> 



  ------------------------------------------------------------------------
 | Curtis D. Levin  KD4ZKW     1st year / Computer Engineering Technology |
 | cdlevin@shadow.net                 /  Grantham College of Engineering  |
  -----------------* MS-DOS6 * OS/2 Warp * Linux *------------------------

On 21 Jan 1995, Nathan Hand wrote:

 : >
> : >Of course it won't actually do you any good, because first they'll try to
> : >say it's your fault, then they'll misuderstand the problem, and say that
> : >it's fixed by a release which fixes some entirely different bug.  If you
> : >finally get them to accept the problem, they'll say that it'll be fixed in
> : >the next release

     Very true about all products involving computers. Must be a law or 
something.
 > 
> : Hmm, this sounds like about 75% of the answers folks get to questions posted
> : about Linux.  All of the above happen as much if not more in Linux than
> : they do in commerical releases, but with commercial releases you have a much
> : better chance of stability at release time.  If you don't count patchlevels
> : as releases, then Linux is 'released' as often as commercial software.  If you
> : cound patchlevels as releases then Linux has many unusable/buggy releases
> : which would cause most commercial vendors to go bankrupt.
> 
    Which is great if you know the questions. Most people first starting out
   with computers on any level require a great deal of Q&a time, and that's
   just to get the basics down. Most people who've switched to Linux either
   had prior experience with unix type software, or are surrounded by people
   who are knowlegable enough to know what to tell them. It's impossible to
   get a computer to do anything without the correct input. If you don't know
   any command syntax at all, you're basically screwed. 

> : Free software is great, but there is more room for *supported* software in
> : the commercial world than for free software.  When my paycheck is depending on
> : a pice of hardware/software, I want to *know* that it works, and commercial
> : vendors provide me with guaranteed support that NONE of the free software folks
> : will give me.

    Funny, but most pros have a tendency to trust the software that they 
themselves write. A lot of people mistrust commercial packages for 
sensitive material(IMHO). Besides, if the software is written and compiled
on your machine, then there's probably less of a chance that it'll go 
buggy. The real point is that the software, if it's any good, will have
all the support you need, either in book form, or built into the program.
So much for calling tech support, if it's in the manual.
 > 
> Are you arguing that *supported* software like MS-Windows *works*? Commercial
> vendors suffer from the limitation of too little input. They have small teams
> (perhaps under 100) working 9-5 with only the very basic experience. Contrib-
> utors to Linux amount to thousands of highly-learned professors, scientists,
> real-world workers, students and home hobbyists (all of them highly-learned).
>
   Incredibly true, and it reflects in the software, and in the OS. You
   need to be skilled to use linux. I'm just learning it myself, but if
   I was a skilled C programmer, then it would probably be easy.
 
> I dont find the commercial products nearly as stable as the "unsupported"
> Linux. It was because of massive instability (Windows, 'nuff said), slow
> performance (Windows, 'nuff said) and massive resource requirements
> (Windows, 'nuff said) that I switched. I imagine that many Linux users
> will be the same.
> 
   Not under windows, that much I'll give you. But, if you turn off the
  animation and multimedia in OS/2 warp, you might see a difference in 
the  apps that will run under windows. Windows is still faster on a 386
then anything else, but you are right, it's not as stable.

> Yes you often get the "its your fault" and "rftm" answers on linux groups.
> Often not without cause: Im amazed how often the same question will be asked,
> or a complete novice will ask an obviously un-informed question like "how do
> I run my Windows programs under Linux". Hell, Ive asked some real clunkers too.
> I dont take the (sometimes rudes) replies to heart though.  Just wipe away the
> tears and start learning.
> 
    I'm still waiting for the generic OS that will run all types of 
programs in a 32 bit environ. For us 386 users, that could be os/2,
but not really. There's too many programs that won't run right. Installing
any kind of dos program compatible drivers is a nightmare. I'm still 
hacking away at Linux, too, but it would be wonderful to wake up one day
and find an os that does everything everyone wants,and runs all software 
plug and play. Linux has potential. 386's and 32 bit systems are 
relatively new, and the one thing I like about it all is that the 
developer(s) seems to have the right idea. 
     Maybe someday in the not too distant future, we'll see an os that 
can do that, or a shell that can do that with any os. Linux has read down 
pat, and write down for dos, and execution of files actually loaded onto
a kernel shell is reality today. There's no reason why the system can't be
programmed to do a plug and play shell, it's just that there's not 
millions of dollars riding on it in the marketplace, at least, not today 
there isn't. Hmmmmm. Have a good day.