Off topic but...

[tafkaMr.E]

...possibly of some interest to someone.

This is one of the replies recieved to a question I asked about the effect of a recession on the Operating System markets on the comp.os.linux.advocacy newsgroup.

BILL DOOR wrote:
>
> Just wondering if any of you guys had any thoughts on the effects that a
> recession in america (and maybe a large portion of the world) would have on
> the OS market.

In many ways, the recession of 1983, the recession of 1987, and
the
recession of 1992 were all closely tied to Open Source. And
Open
Source software played a key role in the recovery.

In 1983, we were recovering from interest rates that had soared
to
nearly 20%, inflation was also insane, and government spending
had
run amok. The Reagan administration tried to cool the economy
off,
and tax simplification had people trying to shift spending, and
save
money. Also in 1983, AT&T was allowed to sell UNIX, but BSD
4.x had
already raised the bar to a height that could only be met with
BSD
software. Open Source became a critical component in systems
used
for telecommunications, power grid control, military, and any
near
real-time mission-critical applications.

Open Source officially started in 1984, with the drafting of
the first
"General Public License", it protected authors and contributors
from
predatory competitors who would try to make proprietary
products out
of software which had been contributed to the public domain in
good
faith. Ironically, it increased the interest of corporations,
who
quickly saw that they could collaborate on key software, while
developing their strategic software "in house".

In 1987, DEC and a number of other minicomputer makers had
"pulled
out all the stops" to sell their proprietary operating systems
(VMS, VRTX, OS/400,...) while UNIX literally sucked the wind
out of their markets. The trigger event was that oil wells
in the persion gulf had been bombed. The fear of oil prices
doubling sent the market into a dive, the failure of
proprietary
X.25 switching systems used to distribute news wire services
became overloaded and this triggered a deeper panic. The
sudden drop in price triggered programmed trading, with dropped
prices even further. And the inadaquate clearing systems made
the market even more jittery.

Ironically, the one source of information that DID work was the
network of UNIX systems connected via TCP/IP links. Very
quickly,
the NASD decided to standardize on UNIX as their system for
trading
stocks. Before long, more and more companies were adding UNIX
to their list of "must have" systems. In many cases, UNIX was
used
to integrate the other systems, to provide a "front end" that
could
make Mainframe data available to the minicomputers, and
minicomputer
information available to the mainframe.

In 1991, IBM had literally "Bet the Farm" on MVS 4.0 and OS/2
2.0.
They assumed "If you build it, they will buy". Unfortunately,
the
cost of the software and hardware upgrades exceeded the cost of
the original hardware. In additon, OS/2 2.0 was having
reliability
problems. Many IT managers were being pressed for $5 million
per
mainframe, and had to upgrade 5-10 mainframes. Meanwhile,
Oracle
and Sybase were touting "mainframe class" databases available
on
UNIX systems. The market was very competitive. UNIX
administrators
sheepishly suggested that a Pyramid with 16 processors might be
able
to take some of the load off the mainframe, but it would cost
nearly
$250,000 (a very high price for UNIX systems in those days).
Even
if it would only save them from having to upgrade ONE
mainframe, or
at least delay the upgrade, the machine would pay for itself
nearly
20 times compared to the MVS upgrade. (If this is sounding a
bit
like Microsoft's XP upgrades, keep reading).

IBM's stock fell, and UNIX stocks started moving up.
Administration
costs were surprisingly low, the UNIX teams were remarkably
responsive
(largely due to their large cache of Open Source tools -
courtesy of
BSD and GNU). Very quickly, UNIX got very popular.

Microsoft and IBM went their separate ways, OS/2 going to IBM,
and Windows 3.1 going to Microsoft. Microsoft decided to
strike
early and hard, before IBM could establish it's OS/2 market.
And
IBM's competitors were more worried about IBM controlling the
market
than Microsoft. The fact that Windows 3.1 was bundle-priced
with
Word, Excel, and Powerpoint (not quite the whole office), was
inconvenient,
but necessary since Lotus, WordPerfect, and Borland were intent
on
supporting BOTH Windows and OS/2.

By 1993, Open Source was already having an impact on the
economy.
UNIX systems were being used to provide e-mail and newsgroup
services
to millions of PC users. Many of these users actually accessed
the
"Internet" via BBS systems such as FIDO and Wildcat. They
would
dial-in, pull down their mail, and disconnect. Often, users
would
pay $20 to get set-up, and then after a few months, would stop
paying, but would keep downloading.

By early late 1993, the BBS operators had begun to offer
"real-time"
services, such as access to web browsers. But since the users
would
be tying up phone lines for longer times, they were more strict
about
collecting their access fees. Most of these early POPS charged
a flat
rate of $20-$30/month. Every user after the first 20 was pure
profit,
they used SLIP or POP to their Linux box, and connected to the
net
via X.25 TCP/IP connections.

Open Source was crucial to the success of the internet, and the
economy that began booming shortly after. Most of the eariest
sites were old Windows 3.1 PC's, 80386 machines mostly, with
Linux, and maybe an extra hard drive (made swap faster).
They could download the system over a week or two, and could
get the whole thing installed in a few days. Eventually,
they grew enough, and generated enough income to fund
Sun SparcServers, then E-450s, and many eventually grew
into E-10,000 servers. Ironically, in an interesting twist,
many of them are switching to Linux on Z-900 mainframes.

> It seems to me that when recession hits, corporations
> are going to try to cut costs. One of the ways to do this
> is obvious to you and me but is it to them?

4 years ago, most CIOs didn't even know they had Linux systems
in house. The Linux systems were so cheap, and most of the
machines were running reliably, that they never had to sign
any purchase orders. In many cases, the Linux license was
charged as "Office supplies", or "petty cash software".
The server was usually a machine that couldn't run Windows NT.
In some cases, it was a minitower that wouldn't run Windows 95.
About the only time the were even noticed, was when the CIO
would
come down to tour the computer room. He'd go to the very back
of the room, notice a few very old minitowers sitting under a
table, and ask "What are THOSE" (astonished that anything that
old was still running). When he was told that these were
Linux machines, he said "get rid of them". When he was told
that these little Linux machines had been providing e-mail,
firewall,
and intranet web serivices for the last 2 years, he said,
"put them on better hardware". After all, if you had a winning
solution and it failed, it was going to be a bit hard to
explain
that they needed to find an 80486 chip at a hobby shop.

One year later, 17% of the CIOs surveyed, said they were using
Linux,
but not for "mission critical systems" (Oracle databases and
such),
the fact that Linux was providing firewalls, DNS, POP mail, and
the local mirror to the company web site (via rsync) wasn't
considered
"Mission Critical" while Oracle databases running on Sun
E-450's were
considered Mission Critical.

The following year 24% of the CIOs surveyed said they were
using
Linux for numerous utility tasks. In fact, many times these
Linux servers were even providing fault tolerance componants
for other systems (Dynamic DNS, Web Browser load balancing,
SOCKS...).

> Being that Linux is a child of the internet

If you are referring to the combinition of TCP/IP Arpanet and
Usenet
that became known as "the internet" back in 1984, you're
correct,
Linux is a child of the internet.

If you are referring to the World Wide Web, E-commerce,
business-to-business communications, real-time streaming, and
the $1 Trillion industry, then Linux is the MOTHER of the
Internet (And UNIX is the father).

Linux created internet access for more corporations, created
more first websites, and created more first Intranet servers
for more corporations, than ANY other operating system. Time
after time, corporate leaders who couldn't even see the
possibilities of millions of customers ordering products via
the internet, would deny a request for a $25,000 SparcServer as
"rediculous nonsense". But any good manager knew a good
"scrounger" who could find out where the PC that had been
replaced by the VP's newest PC was buried. Usually, the
machines were
so old that the biggest challenge was making sure it didn't get
tossed out
as garbage.

In many cases, the guy who put together that first Linux server
eventually
ran the company's web site. Today, many of these guys are
CTOs, or will be soon. This year, nearly all companies are
using Linux in some capacity.
Nearly 60% plan on using even more Linux next year. And this
survey was
conducted before Microsoft's "upgrade to XP or Die" ultimatum.
Many are
even exploring the possibility of Linux on the Desktop. Some
managers
are even adjusting corporate policy to make it more "Linux
Friendly".


> there is'nt afaik a precedent for this kind of situation.

As you can see above, there are precedents. And yes, when
money gets
tight, and the next round of cuts is going to be in management,
the
managers get very interested in switching to Open Source
solutions
where they can see a fit. More importantly, the seeds for
recovery
are often found in the unconventional solution (such as Open
Source).

Other precedents of this broader "breakthrough thinking"
concept include:

In 1974-5, the country was threatened with a terrible gas
shortage (it hit
Colorado first, other states much later), the guys driving
volkswagons,
civics, and corollas were laughing at the guys driving the huge
GM Muscle
cars. A guy came up with a way to get "200 miles per gallon"
from a
car. It was essentially a fuel injector. There are portions
of his patent
which could be implemented later, if the country's fuel needs
required it.

In 1980, the economy was a mess, and Personal Computers were
just "toys".
But companies who couldn't afford to keep upgrading their "big
iron"
began to experiment with CP/M and MP/M systems. Very quickly
the Z-80
based CP/M system was threatening to wipe out the PDP-11 and
Series 1 markets.
One of IBM's key concerns in 1982 was that they didn't want
MS-DOS to be
a multi-user operating system. They really wanted it to be
just a bit
smarter than a dumb terminal (the original PCs didn't even have
hard drives,
they were loaded from audio-cassettes).

Those early CP/M programmers got the preview of MS-DOS. By the
time MS-DOS
came out, they had the tools, and the skills to write very good
MS-DOS
software. They created a market that threatened IBM itself,
from "toy
computers".

> Your thoughts please.

Perhaps times have changed. Those of us who grew up with Howdy
Doody
(born 1950-1955) were active participants in some of the most
radical
changes in history. In 1959, we did duck-and-cover drills, so
that
we knew what to do when the A-bombs hit. In 1963, we watched
the first
news coverage of JFK's murder (where you could easily see that
multiple
shots were fired from different directions), we watched
brothers and
older friends go to Vietnam, we watched day after day of REAL
people
suffering from REAL wounds, and REAL prisoners being killed by
REAL
soldiers with REAL pistols. And then we'd watch Elmer Fud blow
Daffy
Ducks beak off, 20-30 times an hour. And many of our
schoolmates came
in the next day with bruises all over their bodies, where their
drunken
fathers or step-fathers had beaten them, with belts, boards, or
electrical
cords. And the teacher would say nothing.

We watched sisters get sexually molested (or hear about it),
the school would warn us that there were "bad people", and not
to get into their cars. In many cases, these "bad people were
stepfathers, "uncles" and other friends of mom. The boy scout
leaders were gay, but since they were married, and good friends
with the preacher, no one would believe you. Girls wore
dresses and skirts, boys wore long baggy wool pants.

We protested the war, we disobeyed laws, we challenged
authority. Some of us rioted in the streets, we had been
"waiting for the bomb" for so long that death was no longer
something we feared. We knew we were going to die, the
question was "When". Some of our friends used drugs, we drank
even though we were too young, and we eventually learned that
even the other protestors couldn't be trusted.

We lived through the "Me Decade", with everyone so busy trying
to find the meaning of life, trying to get what they wanted,
trying to get paid, get laid, and not get caught. We listened
to Rock Music when it wasn't cool, and we had 4 channels on TV,
and one of them was "educational programming". The Television
told you what to think, when to think, and how to think it.

We watched as we learned that the President of the United
States had tried to cover up the fact that his aides had hired
thieves to alter an election. He ended the war, opened the
door to china, made friends with the Russians, but the Congress
made it look like he was trying to overthrow the government.
We watched a remarkably successful President, resign in
disgrace.

Then we watched the next President, a really honest fellow,
speak of the "Malaise", he talked us into a recession, we
watched interest rates soar, housing costs soar, and saleries
and wages stayed flat as a pancake. Much of this was because
we were shifting from a War-based economy to a peace-time
economy, and we watched as the economy threatened to collapse.
We even watched Iran take our embassy hostage.

We watched as Ronald Reagan spoke, we didn't know if he was
senile or crazy, but everyone was afraid that he would "push
the button". The economy improved as we went from being
"Hippies", to being "Yuppies". We started having children and
became parents to the second largest "baby boom" in U.S.
history. We earned, we worked, we created the computer
industry, the PC industry, the Cable TV industry, our rebel
nature enabled us to challenge the norms of corporate
executives who assumed there was only one way to do things. We
made millions, even billions by outmaneuvering the "command and
control" organizations where initiative was punished, and
conformity (Blue Suit, White shirt, Red Tie, black leather
shoes, the "corporate uniform") was rewarded. We bought our
own PCs and brought them to work with us. We created
spreadsheets, charts, diagrams, and prose in various fonts, and
each report and presentation looked like it had been prepared
by a madison avenue marketing firm. You didn't even have to
know what you were talking about, if you could "make it look
good", there was always some executive who would tell you what
to say.

We took the internet, which we used mainly to discuss sex and
politics while we were in college, and turned it into a way of
doing business. We redefined business completely.
Those of us with families spent more time working from home.
Those of us without families,
(divorced) spent more time travelling and visiting customers.
We even watched as world political and economic policy was
formed in internet discussions. We saw the emergence of a
whole new form of democracy, in which those who really cared
about an issue, could get directly and actively involved is
solving the problems and getting the solution implemented.

And today, we have Bill Gates, one of our own generation,
attempting to try and get us back to the "good old days" of "Ma
Bell", "3 networks, all alike", and total conformity, and
thinking that he should BE the monopoly power that Ma Bell, Big
Blue, and NBC/ABC/CBS used to be. He wants to remove the
choices. And he want to do it when budgets are tight, and he
offers nothing in return.

Do you really think that our generation would let him do that?


--
Rex Ballard
IT Architect
Leader in commercial use of the Internet and Open Source.
http://www.open4success.com