Date: Mon, 6 Apr 1998 16:11:48 -0700 (PDT) From: Tom Lord <lord@emf.net> To: freebsd-questions@FreeBSD.ORG Cc: lord@emf.emf.net Subject: not exactly a question Message-ID: <199804062311.QAA03700@emf.emf.net>
next in thread | raw e-mail | index | archive | help
A Strategy For Winning Big With Freeware
by: Tom Lord <lord@emf.net>, <lord@lana.lanminds.com>
1810 Francisco St. #2 / Berkeley, CA / 94703-1313
date: 6 Apr 1998
----------------------------------------------------------------
"The primary goal of the summit is a high-level exchange of ideas
and strategies for expanding the acceptance of popular freeware"
Tim O'Reilly (http://www.oreilly.com)
I think it is too hard to build a small network using publicly
licensed software. To wire up, for example, a small office, there
are many programs to install and configure, and if you don't get
them just right, the whole thing fails to work. Each application
has its own installation procedures and requirements. Documentation
is generally poor and scattered.
Here's a scenario: suppose you want to wire up a new CD store so
that the cash registers all send transaction reports to a single
server that generates a journal tape that the (off network) inventory
database analyzes at the end of the day. To me, that sounds like
a job that should take perhaps 2-4 hours, if the cable has already
been installed -- loading standard software onto the machines,
perhaps setting a few parameters, and double checking to make sure
everything works. But with freeware: ok, I'll look into Emacs modes
and Tcl/Tk scripts for cash registers while you install Linux
everywhere and configure the net; we'll meet back here in a couple
of days.
I think the best "strategy for expanding the acceptance of popular
freeware" is simply to identify a whole bunch of similar scenarios
and create "pre-fab" solutions for them. For example, it might be a
good idea to publish (compatible) CD-ROM and documentation sets for
a specialized unix, pre-configured to boot running exactly one
application: a cash register interface or a database server.
From a technical point of view, producing such a series seems almost
goofily easy, so in my opinion it would be a lot of fun to do
_extremely_ well.
I propose that we form a general agreement to cooperate in building
a series of software and documentation publications. The series
to be constructed in steps:
1. Identify a small number of network-building scenarios, like
wiring up a retail store, that we want to optimize.
2. Design simple component specifications for components which
solve the scenarios and that are easy to build.
3. Independently build components to fit the specifications.
Build redundant components. Build components that interact in
nifty ways using protocols expressed as standards.
Simultaneously, integrate and test the independently
developed components; prepare them for publication.
4. Publish the software, documentation, and "research and
development reports" about the internal standards developed,
etc.
In addition to being generally beneficial and possibly effective in
"expanding acceptance of freeware", such an exercise would be
profitably repeatable an essentially indefinite number of times.
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199804062311.QAA03700>