Keys to Effective Linux Advocacy Within Your Organization
by Russell C. Pavlicek of Compaq Computer Corporation



INTRODUCTION:

The opinion of many market analysts:

  "No technology can invade the desktop unless they have a huge marketing 
  organization."

The opinion of an advocate:

  "No marketing organization can withstand the effects of a community 
  which generates an ever increasing number of effective advocates."
  
We have come to a time in the Linux movement when we see major corporations
jumping on the Linux bandwagon.  We see the advent of press coverage and
advertising campaigns by major players in the computer industry.  We see 
organizations with big budgets telling the world they do Linux.

The tendency for Linux Advocates is to sit back and say, "Look at all the
marketing and advertising!  We never had the benefit of those things before.
We can sit back and take it easy; our job is done."

Nothing could be farther from the truth.

Does the Linux community need marketing and advertising?  Of course.  But it
needs Advocates more than ever.  In this paper, we will discuss what
Advocacy is, why it is important, how it is done, and who should do it.


BACKGROUND:

In 1995, I was an OpenVMS technical consultant with a good background in
system internals.  My manager approached me and asked me to go to a training
session in the Sector 7 toolkit, which migrates OpenVMS code to Unix platforms.
I didn't know too much about Unix at the time, but I went to the training
anyway.

It happened that a PC mailorder catalog that I brought with me on the 
airplane had an ad for this PC-based Unix called "Linux".  It cost about
$25.  I figured it would be a good investment so I could develop some
Unix skills.  While I was at the class, I showed the catalog to my classmates,
many of whom ordered it as well.  Without realizing it, I had begun a pattern
of Linux Advocacy that would characterize the next few years of my 
professional life.

After I received the CD (Yggdrasil Plug-and-Play Linux, Fall 1994, kernel 1.0),
I began to use it on my lowly 386SX/40 -- a boat anchor even then.  Later that
year, I needed to go on a two week training course.  I set up a system by
which I could dial into my Linux box and exchange email with my wife during
that period (remember that the Internet was not a household item in 1994).
One night I dialed in and discovered that my wife was the system at the same
time I was.  That's when it finally hit me: this PC was acting like a REAL
computer!  It was doing many things at the same time with no problem.  All
it needed was a good operating system: Linux.

In 1997, I attended the first Atlanta Linux Showcase.  It featured only 
about 20 vendors and about 500 attendees.  Even so, I was electrified!  I 
heard stories of people doing great things with this operating system and 
I saw how far some of the newest software had come.  This movement was
going somewhere and I wanted to be a part of it.

When I returned to work, I assembled a 486 out of discarded junk I found
in our lab.  I installed Red Hat 4.2 with Apache and created a box that
became my workstation and intranet web server.  I put up a web page with
as many Linux links as I could find, focusing on articles in the trade
press.  I made this piece of junk into a quiet witness to the abilities
of Linux.  Some of my coworkers laughed a bit about my goofy little machine
with its strange little operating system.

I wanted to see how stable the operating system was, so I purposely did not
shut down the box for any reason.  Finally, after 267 days, it went down -- 
when a severe thunderstorm took out all the power in the building!  People
weren't laughing so hard anymore.

Around this time, I began to write notes in one of our internal online
conferences about building a business out of Linux.  People laughed and
scoffed.  "That hacker's toy?  Don't bother us with that!  That's for
geeks to play with!  You can't do business with that!"  About a year later,
the tone had changed to, "Well, what would a business plan look like?
How could you make money with free software?  That makes no sense."
Finally, in the fall of 1998, high level corporate managers began to
accept the concept that maybe there was a business in this Linux thing,
after all.  And when the opportunity opened up, a number of active Linux
Advocates run into the gap to begin pushing the new Linux strategy, led
by Jon maddog Hall.

Oh, remember that 486 junk system I built?  It's been upgraded to a Pentium
90.  It serves as my workstation, my intranet web server, a Majordomo
server, and a host to a web-based discussion forum.  It's main web page
registered over 9200 hits in its first two years in operation -- not
bad for an unsponsored intranet page.  Yes, I could now get approval to
upgrade it to faster hardware, but why bother?  A Pentium 90 providing
all those functions can be quite impressive.

I am writing this paper because I know what it is to be a Linux Advocate.
I have seen what works and what does not work.  And I need to share that
with the community so that more effective Advocates can rise up in their
organizations and introduce Linux to their world.


WHY IS ADVOCACY IMPORTANT?

Advocacy is powerful -- extremely powerful.  And now is the time to increase,
rather than decrease our efforts.  Regardless of marketing and advertising,
we need effective Advocates.

Advocacy becomes highly effective when it comes as the personal 
recommendation of a trusted friend to try a product or service.  Which of
us has not stayed away from a service provider who gave a friend a hard
time, or used a product which was recommended highly by someone we know 
well?

Years ago, while I was working at a software house, a sales rep who was 
approaching retirement said something to me that I will never forget.  He said,
"Russ, when you show me a new product, I want you to do everything in your
power to convince me that this is the best product of its kind on the 
market.  When I am in a selling situation and I know that I am doing a
disservice to that customer if I let him buy our competitor's inferior
product, there's a look in my eye and a tone in my voice that customers
can detect.  They hear me and understand, 'This man isn't just selling me
something, he is convinced that his product is what we really need.'  And
that makes all the difference in the world!"

That is when Advocacy is at its peak: when the person your are talking to
can detect that you are absolutely convinced that you are speaking the
truth, and you have a passion that ranges far beyond the desire to "feed"
off of the person's decision (as you find in many sales situations).

Advocacy is also important because it is exponential in its growth.  This
is an important point: do not confuse the method with the result.  Just
because Advocacy is often done one-on-one, that does not mean that the 
results are linear.  When you practice effective Advocacy, you grow more
effective Advocates.  And each of them begin replicating themselves in a 
similar fashion.  The result is an exponential growth pattern which can
bring amazing results.


WHAT IS ADVOCACY?

Advocacy is very different than marketing.  It generally has a much smaller 
focus of one-to-one or one-to-a-few.  As such, some people say Advocacy is
not as "efficient" as marketing -- you are not reaching vast audiences in
a short period of time.  But, as I mentioned earlier, the end result is
actually much more impressive than what appears to be the case at first
glance.

Because Advocacy works person-to-person, it is generally not as costly as 
marketing.  But Advocacy rides high on something that most marketing can
barely produce: TRUST.  Trust is the lifeblood of Advocacy.  Without a
sense of trust, advocacy is almost useless.  But when someone trusts you,
you have the ability to sway them with your conviction about what can and
will help them.

Because Advocacy rests on trust, it is imperative that you get your facts
straight.  You cannot trust someone's message when they have the facts
wrong.  So get your facts straight and do not overwhelm your listener with
them.  If you spout too many facts, the listener will become saturated and
will not be able to absorb what you are saying.  Make sure your facts
are in order and then dole them out in digestible portions.

Stories are also an incredibly strong part of Advocacy.  Marketing people
use stories much of the time to make their point because they are so
powerful.  Why do you suppose they fight hard to find "reference accounts"
-- people who can tell how the vendor's product has made a difference in
their operations?

There is incredible power in first-hand stories, especially if the person
telling the story is someone you trust.  Herein lies one of the most
powerful techniques in Advocacy: establishing yourself as a trustworthy
person and then telling persuasive first-person stories.  When someone you
trust tells you something that they did, what choices do you have?  You
might think that they are lying, but you are unlikely to think so because
you trust them.  You might think they are crazy, but again you are unlikely 
to do so.  So, when someone you trust tells you about something they did, 
the natural response is to believe them.

It is therefore imperative that we establish ourselves as trustworthy people
and that we learn our first-person stories.  We need to be able to recite
things that we ourselves have done, so that we can persuasively attest to
the potency of Linux.  That's also why I began this paper with a number of
first-person stories: it helps to establish context and trust, as well as
delivering information that cannot be easily dismissed.

Advocacy also thrives when you spend time to build relationships with the
people you are trying to persuade.  Good marketing cannot overcome a good 
relationship.  Think of how many corporate decisions are made every year
on the golf course.  So, in Advocacy, the rule of thumb is this: Win friends, 
then reap converts.


WHAT NOT TO DO:

Don't spread FUD (Fear, Uncertainty, and Doubt).  If you can't deal from 
the deck of truth, don't play the game.  Let the FUD be characteristic of 
your opponent.  Remember: you need to build trust.  If the listener knows
that you won't resort to cheap tactics, his ears will be wide open.  If he
realizes that your opponent will do anything in his power to win the point,
even resorting to unsavory tactics, the listener will be more cautious
about accepting what he has to say at face value.

Don't waste time knocking Microsoft.  Anyone can shoot fish in a barrel -- 
at close range -- with a bazooka.  Microsoft has plenty of flaws that you
could try to exploit.  But you are likely to build enemies rather than real 
victories.  If I say to a Windows engineer, "Windows sucks", what have I
done?  I have told him that his education, his certification, even his job
are all worthless!  What I have done is to build an enemy who will stand
against me until his last breath.  He will never yield to me, even if I
prove my point beyond a reasonable doubt.

Also, knocking Microsoft makes Linux look like just another ABM (Anything 
But Microsoft) proposal.  It sounds like you are saying that you hate
Microsoft so much that you are willing to settle for an inferior substitute.
Focus on the positives instead.  We have so many good things to talk about
that we do not need to bother ragging on Microsoft.


HOW DO YOU DO ADVOCACY?

Know your goals, both long and short term.  They may be quite different
depending on how much time you have.  And there is no need to advertise 
your goals either.  Know what you are trying to accomplish so you have a
chance of actually succeeding.  Just like the old adage says, "If you
aim at nothing, you will hit it every time."  Know what needs to be done
and when.

Anyone who has heard maddog Hall talk about his Linux Advocacy at Digital and
Compaq and heard me talk about the same could assume that we were talking
about entirely different companies.  In fact, we had very different goals
in the short term, although our long term goals were quite compatible.
Maddog would act as an Advocate to his management to protect the Linux/Alpha
team that was in place.  I, on the other hand, worked almost strictly on
a wide group of peers within the corporation.  I had no product team to
protect and sought to grow more Advocates from the lower ranks of the 
corporation.

Knowing your goals leads into the subject of knowing your timeline.  Use 
the time you have available.  Don't waste time; invest it.  I could afford to
spend 2 years discussing the concept of a Linux business with members of that
electronic forum I mentioned earlier.  Good Advocacy should steep like tea 
when possible.  A quick, "fast food" approach will inevitably yield "fat" 
results in improperly set expectations or poor grasp of important concepts.

Know when it is best to "fail".  You can sometimes lose in the long run 
when you win in the short term.  If you seek to win the immediate point at
all costs, you can make enemies.  And an enemy in the way can increase the
likelihood that you will fail in your long term goal.  It's often better 
to walk away from an argument than make an enemy over it.

Talk it up.  Remember that Advocacy is about relationships.  You need to
talk about Linux if you expect to make an impact on the subject.

Help out.  You need to realize that you are the face of Linux to those 
who are listening to you.  Don't let your own limits scare you.  Learn
to find answers for people by asking for help in newsgroups or doing
"egrep -i keyword /usr/doc/HOWTO/*" (in Red Hat) to find HOWTOs that
discuss the issue at hand.  I used that method when I began to function
as an Advocate, not hiding the fact that I limited experience but I was
willing to look for answers to questions posed to me.  People realized
that I was working hard to find them answers -- and they appreciated that.
That opened their ears to me even more.

Consider distributing free or "at cost" CDs.  You might think that this is
worthless, since you can purchase GPL CDs cheaply from several sources on
the web.  But your friends will trust you; they don't know Linux Mall or
Linux Systems Labs or Cheapbytes from a hole in the ground.  Order a bunch
of CDs yourself and either give them away, or sell them at cost so that
you recoup your investment, but don't make a profit.  Remember that you
are not here to "feed" off people.  Don't even attempt to make dollar on
the deal.

Look for the "small win".  There are two forms of this: fixing the known
problem or fixing the unknown problem.  For example, I once fixed a known
problem having to do with Windows NT servers from two different Windows
domains that would not pass print jobs to each other.  I ended up taking
an old 486, setting up some print queues pointing to both servers, tearing 
the monitor and keyboard off of it and stuffing it under a table.  The end
result was a Linux printserver that would happily farm jobs to either NT 
server -- and a lab full of people who saw how Linux could handle their
problem.

An example of fixing an unknown problem was running my 486 for 267 days
continuously.  Until I had done that, people assumed that PCs HAD to
crash; PCs were not seen as stable enough to run for months on end.  By 
providing proof that PCs can run without crashing, it opened people's
eyes to the fact that PCs crash frequently when they do not need to.

Understand your audience.  They may not understand the value of stability 
and flexibility.  Swiss cheese may look badly flawed to someone who has
only seen cheddar.

Understand what is meant when some managers balk at "unsupported" software.
You may have spent 15 minutes going over support options with them, and yet
they make a statement about not wanting to use "unsupported" software.
Often times, the support that is lacking is the support of the MANAGER, not 
the PROJECT.  If something goes wrong, the last thing that many managers
want is to stand alone before high-level management and give an accounting
of the failure.

It used to be that a manager could say, "I bought IBM and they are working
the problem.  What more could I do?"  In more recent years, the common line
is, "We purchased that industry-standard Microsoft software and we are working
the issue with them."  But Linux has no Fortune 100 vendor who is responsible
for fixing things when they break.

The solution in this case is to offer to engage top-tier vendors such as 
IBM, Compaq, or HP to assure the manager that the project will be completed 
properly and that there is someone to sue if something goes wrong.  Top-tier
firms are now offering Linux consulting and support.  Propose to management
that they use these vendors if they are concerned about implementing the
system without support from a large corporate partner that the top 
executives know and respect.


WHO SHOULD DO ADVOCACY?

You!  You can make a difference.  Who else can talk to people with your
own style?  You are capable of carrying the message to those folks you
know.