Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 08 Oct 1999 12:23:04 -0600
From:      Brett Glass <brett@lariat.org>
To:        "Jason C. Wells" <jcwells@u.washington.edu>
Cc:        chat@FreeBSD.ORG
Subject:   Re: Targeting the server: Not such a good idea?
Message-ID:  <4.2.0.58.19991008111822.0453a530@localhost>
In-Reply-To: <Pine.BSF.4.10.9910082116380.80955-100000@s8-37-26.student. washington.edu>
References:  <4.2.0.58.19991008083634.044de740@localhost>

next in thread | previous in thread | raw e-mail | index | archive | help
At 09:48 PM 10/8/99 +0000, Jason C. Wells wrote:

> >"Targeting the server only is a death wish. Novell tried, and they're being 
>
>Internet?

If I construe your comment above correctly, you mean to say that the rise
of the Internet hurt Novell, not its server-only positioning. But if this
were so, why is NT prospering? NetWare, like NT, supported NFS and TCP/IP
long before the Internet gathered speed as a phenomenon.

> >beaten bloody by NT. Banyan tried -- they even used UNIX -- and never 
> >became popular in the first place even though they were years ahead of 
> >everyone. Microsoft even failed with LAN Manager. FreeBSD will fail at this 
>
>Internet?

See above.

> >too; everyone will go to Linux whether it's better or not.
>
>Linux ain't bad. And if it is the evil some say it is, then it will also
>fall to something better.

Well, as you know, I personally believe that Linux is a problem because,
being licensed under the GPL, it is part of a destructive agenda. But
the fellow with whom I was speaking didn't know much about the GPL, its
motivations, or its effects. He thought that Linux was inferior
to FreeBSD because it was not based on the same time-tested technology
and was less reliable and stable. But he seems to think that the desire
of Linux developers to challenge Microsoft on the desktop is key to its future 
success.

> >"The trouble is that no one wants to have separate training, separate 
> >software, separate configuration, or separate experts for the server. 
> >Companies are tired of paying a CNE 'guru' big bucks to fix NetWare and 
>
>No need for an NT admin. Bullshit. That is the biggest myth of NT. 

I think that what he was saying was that there was no need for a SEPARATE
NT admin. And I can see his point: CNEs in NetWare shops were treated like
gods and gurus and had to have specialized training. Institutions are rankled
by "priesthoods."

>And if
>the workload requires a staff of six, then what diff does it make if one
>person is Unix and 5 do help desk.

Because the "server administrator" must be a separate person who is more
specialized and more expensive. (Specialists cost more.) And if there's 
only one, there's no backup for him or her, so you really must have
more than one.

>Who wants to pay for licenses? Who wants to pay for down time? Downtime in
>manufacturing for example costs even small companies MILLIONS each day.
>The cost of _any_ admin is trivial compared to this.

Not from a company's standpoint. Mandatory benefits and overhead can make
the cost of an additional employee at least double his or her salary. 
NT isn't cheap, but still doesn't cost as much as more bodies. So, NT
seems more economical in both the short and long terms.

> >then keeping a whole separate staff around to support DOS and Windows. And 
> >they want their workstations to act like servers: fast, efficient, and rock 
> >solid. They don't want to see these traits limited to the server! The 
> >operating system that everybody wants will run on everything, maybe with a 
> >few tweaks for what it's doing, and will be reliable, fast, and secure 
>
>If this was true, Unix would bo on the desktop already.

Alas, NT has beaten it there, except in certain market niches. (Stock traders
have been using Sun workstations for years.) And Linux is close behind. This
fellow's point, however, is that FreeBSD becomes irrelevant if it stays a 
server-only OS, because it does not satisfy the burning desire for a "single"
solution (or, actually, a family of very closely related solutions built upon
the same code base).

> >if it weren't so insecure. It took over my organization just because 
> >Microsoft *promised* that it would run on both the server and the desktop. 
>
>And it does run, for a couple days.

True. But companies value standardization of business processes and 
procedures -- as well as hardware and software -- so much that they will 
sacrifice a lot to get it. Including reliability.

>Well there you go. The people this guy works for made the wrong decision
>based on dubious criteria. 

He agrees that they made the wrong decision, but understands their criteria.
And I do too.

>Do these people really think the family van is
>going to work as well on the race track as getting groceries?

No, but they both use many of the same parts. They use very similar engines;
it's the tuning and the peripherals that are different.

> >The OS graveyard is littered with the bones of OSes that wanted to be 
> >server-only. If FreeBSD is going to make itself look like it's only for 
> >servers, the same thing will happen to it."
>
>What about the cost of the software? Was that a factor? 

Linux will solve that for these people. When it's fit for the desktop,
it stands to do very well.

>Was the software
>really any good? What about the internet and it's protocols. Did companies
>adapt to that? 

All of the important Internet protocols are available on NT. Yes, the
implementations are suboptimal, insecure, and often a bit incompatible,
but they're close enough for most customers. I only get the chance
to convert them to UNIX after they've been bitten -- hard -- by
security holes and proprietary lock-ins. And, even then, they are
in denial and need a hard sell.

>Is it possible that perhaps alllllllll of the issues aren't
>really shown here 

What other issues do you think are important?

>so that Brett can beat us up some more about how fucked
>up we all are?

"Beat people up?" (He says, quoting from the famous "Tai Kwan Leep" comedy 
sketch by The Frantics. See the transcription at 

http://saturn.math.uaa.alaska.edu/~royce/Funny/leep.html 

or the MP3 at 

http://linux11.ma.utexas.edu/users/pgoetz/Tai.Kwan.Leap.mp3

to get the reference. The transcription just can't do these
comedic geniuses justice.)

Seriously, my goal isn't to beat anyone up. That benefits no one. But
it's always useful to bring some good points to light.

>Back to gloom and doom, Brett? I know you couldn't stay away from it for
>very long. 

No, but maybe useful information that might lead to a minor course
correction. Why try to quell discussion when the issue seems to be
important to FreeBSD's future?

>Brett, I really think you should jump ship now so that when FreeBSD hits
>the Davey Jones' locker you can be safe.

I don't think that it's rational to jump ship when you see an iceberg
in the distance -- especially when there's plenty of time to steer
clear. You might say that I'm calling from the crow's nest well in
advance.

>Oh, and FreeBSD does the desktop as good as any unix but not as good as
>Mac or Windows. 

Yes, but NOBODY KNOWS IT! And so they're foregoing FreeBSD for other things.
Which is a shame.

--Brett



To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-chat" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4.2.0.58.19991008111822.0453a530>