Date: Fri, 22 May 1998 20:37:13 -0700 (PDT) From: Mark Diekhans <markd@Grizzly.COM> To: freebsd-advocacy@FreeBSD.ORG Subject: Re: FreeBSD A Solution For Business Message-ID: <199805230337.UAA02883@osprey.grizzly.com> In-Reply-To: <01bd85e0$2dccb1c0$f820aace@eliot.pacbell.net> (jackv@earthling.net) References: <01bd85e0$2dccb1c0$f820aace@eliot.pacbell.net>
next in thread | previous in thread | raw e-mail | index | archive | help
>From: "Jack Velte" <jackv@earthling.net>
>you'all are still missing the point. the pointy haired boss isn't going to
>give
>a whetstone if FreeBSD is faster if it doesn't run his favorite Ap that he
>bought.
>
>the competition is not linux or HP/UX, it's MS windows.
I must humbly disagree. FreeBSD is not an alternative desktop system for a
business environment. No Windozes application user is going to give up Word
for LaTeX or Excel for sc. Nor are they going to figure out what it takes to
keep a Unix system up and running. They can handle rebooting several times a
day; its just a red button, they can't handle ps, sysctl and sendmail.cf.
WINE emulation? Why bother? Its not going to buy them anything. I have
pretty much come to the conclusion from using other Windozes on Unix emulators
that it just isn't worth it. If I ever need to use Windozes routinely, I will
just buy another computer, its far cheaper than screwing around with emulation
environments.
Where FreeBSD is an alternative to Microsoft is as a server. NT is not gaining
acceptance as a business critical server. Its ok for small departmental
servers, but it just doesn't have the robustness for something that must stay
up. Of course, Microsoft wants to change this, but its not there yet.
This begs the question, what type of server? what niche to go after?
o Departmental server (file, print, application). Needs to be able to easily
integrate with the Microsoft clients. Out of my area of expertise.
o Central corporate server. The big weakness here is the lack of a database
solution. One of the big three, not MySQL or Postgres. This is a long shot;
these ports are very costly, no vendor is going to do it without strong
demand. The demand will not be there unless the database and applicaions
are there. Even if it where, this is a risk-sensitive environment. Plus
having a Sybase port is not going to do you any good for the oracle or
informix shops. They invest a lot in a database and a database vendor,
more than just money, its in-house expertise and applications. There is
paranoia about these kind of servers anyway, they are not going to take
risk's. The cost of the OS and hardware is nothing compared to the
central DB server being down. Not much hope here.
o Intranet/internet application server. Now this is interesting (but then
again, I get paid to build such things, so I am biased). FreeBSD already
has a reputation as a internet server; expand on it. These systems are
less critical, so the `no body ever got fired for buying IBM' mentality
is not as strong. They are often replicated, so there is some cost
incentive. Some ideas on what is needed here:
o Java. A solid, fast Java VM and development environment. Server
side Java is becoming very popular for intranet/internet applications.
These applications are custom and portable, so FreeBSD as an environment
is not a hinderance. I do such work on a daily basis on FreeBSD and
am using it as a test environment (a solid JVM would help).
o Database connectivity. For Java, JDBC should cover this. For non-java
development, client libraries are needed. Once of the big-3 (sybase,
I believe), has libraries for linux freely available via ftp.
o A port of Netscape enterprise server would be a plus. Apache is
good, but there is a perceived need for a threaded server for
scalability. Maybe more preception than reality.
o Clustering. Built-in IP clustering would also help in promoting
FreeBSD as a solution (although it can certainly plug into existing
routers that do this).
On promoting FreeBSD. Buying Jordan a couple of $1000 suits and a high limit
gold card and sending him out to wine and dine executives is not the answer.
Promoting it to the techincal people within companies has a good chance of
getting somewhere. The IT shop engineers who already know Unix; get them
hooked and they can promote up. These people can accomplish a lot, they have
more power than you realize. They can sell by example:
IT manager: "Hows it going?"
IT programmer: "The Java/WWW based sales force automation application demo
is working."
M: "Wow, that was fast!"
P: "Our only Sun development system was overloaded and the admin's couldn't
seem to get Java working, so I installed FreeBSD on my desktop."
M: "Well, we need to demo this ASAP to the sales manager, can you set it up
on a server?"
P: "No problem, I will leave it running on my desktop, here is the URL."
M: "Excellent!" (rushes out the cubical).
...
M: "The sales manager loved it! He wants to demo it to his VP and get
the senior sales people evaulating it. This is great, it will really
making us look good, but the AIX box that is slated to be the
server hasn't even got all of the signatures on it, its going to
take a couple of months to get it."
P: "No problem, we have a Pentium system in the machine room that isn't being
used for much, lets install FreeBSD on it. He can be showing the VP
in a couple of hours!"
M: "I don't know about this free Unix. Getting the VP to like this is
important, he signs out checks. We can't have it crashing or it will
be back to 3270 based apps! Can we use NT?"
P: "I thought you didn't want it to crash..."
You get the picture, the AIX box gets delayed and the actual app goes
on line with FreeBSD.
"It's easier to get forgiveness than permission."
-- Grace Murray Hopper
All IMHO,
Mark
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-advocacy" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199805230337.UAA02883>