Subject: Re: Journal / Trade Mag?
From: "Karsten M. Self" <>
Date: Wed, 9 Mar 2005 21:14:52 -0800
Wed, 9 Mar 2005 21:14:52 -0800
on Wed, Mar 02, 2005 at 08:06:52AM -0500, Robin 'Roblimo' Miller (

> Another thing to note is that most well-promoted open source projects
> are either headed by non-coders (Mitchell Baker, Mozilla) or have a
> non-coder onboard to handle "community relations" a la Louis
> Suarez-Potts at OOo. These projects have funding, of course, which
> helps, but there are people out there, including PR pros, who will do
> pro bono work for volunteer projects if they are asked.


Interesting examples, 'coz OOo and Mozilla have somewhat checkered

OOo, which *really* ought to be a no-brainer (MS Office compatible!
Cross platform!  Free!) remains something of a hard sell outside
technical circles.  I really don't think OOo or StarOffice have the
visibility they deserve.  And they  are  backed by what's still a pretty
considerable firm, shrinkage and losses notwithstanding.  Of course,
Sun's laser-like focus on objectives is a topic for another

Mozilla's greatest success is a rebel project that broke away from the
mainstream browser.  And yes, Firefox *has* taken the world by storm,
though I'm not sure how much of the publicity it managed to garner was
self-directed or came out of the Mozilla project proper.  Mind:
capitalizing on successful forks is  strongly  recommended, and there
don't appear to be  too  many overt hard feelings between the Mozilla
and FFX communities, though some frictions have surfaced.

If we look back on notable free software successes, we have:

  - Apache, webservers.
  - Firefox, browsers.
  - GNU/Linux, servers, clusters, embedded systems.

Firefox offers the strong advantage that browsers are not generally
tightly coupled (despite Microsoft's best efforts), and drop-in
replacement is possible.  This coupled with an ongoing rash of
exceptionally bad security issues with MSIE, as well as Microsoft's
total abandonment of browser development for the past four years.  The
opportunity was open, Firefox took full advantage.

Embedded systems (at the unit level) are similarly modular, though
development efforts require some supporting groundwork.  Webservers are
increasingly an interdependent component, but Apache offered strong wins
early, and is holding and extending dominance.  Interdependence in
serverspace varies, but can be low in many areas.

Desktop space remains difficult, but we're chipping away at it.  And
there are those of us who've run nothing but for approaching a decade
(eight years, for myself).

The point is that marketing's part of the picture, but viability is very
much a part as well.

> But the biggest barrier to effective marketing for open source
> projects is the, "Our stuff is great, and great stuff doesn't need
> marketing. And marketing is nothing but lies anyway," attitude so many
> FOSS people seem to have.

That and the fact that real ads cost real money.  But yeah, there's
"project" and "product" as Brian mentioned elsewhere.

Still, tools like Freshmeat and Sourceforge, highlighting popular
projects, etc., are at least promoting what's currently popular to the
already-clued technical market.

> Umm.... whatever.  Go on believing that in a world where inferior
> products commonly rise to the top of the marketplace while superior
> ones tank, and you may be happy but you will be WRONG.

For a really interesting case study of this, dig up ESR's "PC Unix
Buyers' Guide" FAQ from the early 1990s.  Lots of people were producing
x86 Unix at the time, little of it Free Software.  The market leader
grossly lacked features of many lower-priced products.  Oddly enough,
their name was SCO, which I may have heard elsewhere....

> The belief that lying is an effective marketing tactic is sooooo
> untrue that I'm suprised when I run into people who still believe this
> canard.  The reality, especially for a small/newcomer company or
> product, is that underpromising and overdelivering is more likely to
> generate long-term success than doing things the other way around.

I was just commenting today to someone who's been coming on-board in a
big way, that this is one of GNU/Linux's (and Linus's) strengths.  Both
consistently under-promised and over-delivered.  And while "this doesn't
suck as much as I thought it would" doesn't sound like a ringing
endorsement, when it starts hitting with monotonous consistency, and the
alternatives  don't , it starts to be persuasive.  Unfortunately for the
impatient, it's something of a long-acting approach.

> Indeed, I believe the main reason Linspire (formerly Lindows) isn't
> the most popular Linux distro in the world  is that Michael Robertson
> made so many untrue claims early on that the company lost all
> credibility with journalists and trendsetters -- and still hasn't
> overcome its credibility gap.

That's another discussion I've had this week.  Subscriber to the
debian-user list mentioned that he had significant problems selling
Lindows/Linspire, even at a steep discount, compared with Debian.  The
fact that a steep discount (loss) was $50, vs. $9.50, may have had
something to do with it.  But the name also mattered, as well as the
recommendations of the seller -- above and beyond Lindows/Linspire's
slick marketing.


Karsten M. Self <>
 What Part of "Gestalt" don't you understand?
    > That's nice, but totally beside the point.
    If you put it on top, it will just fall off.
    - Nick Moffitt strikes again.

["application/pgp-signature" not shown]