What lessons can we learn about open source
software from the development of the Internet?
"The value of the Internet as a case study is that it shows us the value of peer
review. This is a lesson we should actually know from other branches of engineering and
from experimental science. Which is that the way you get reliability, even when you are
dealing with complex problems that are not well characterized or understood is by doing
experimentation and then subjecting the results to independent peer review...You don't get
to pour concrete for a hydroelectric dam or build a suspension bridge until the blueprints
have been reviewed by engineers who were independent of the original design group. If we
look at software engineering on the other hand, throughout most of software engineering we
do not institutionalize peer review. And in one corner of software engineering we do
institutitionalize peer review and that is Internet software. And we look at the rest of
software engineering and we see reliablity that is extremely poor. And we look at Internet
software and we see reliability that is extremely good. Coincidence? I don't think so.
What are some examples of open source
software that people use on the Internet everyday?
"We generally take for granted being able to get e-mail anywhere on the Internet
at any time with very high levels of reliability. That level of 999's reliability is there
because the e-mail transports and the DNS software and the other components are all open
source. Another good example is Web service. Apache, the open source Web server, depending
on how you count has between 54 and 60 percent market share...We generally count on being
able to get to Web servers almost all the time, unless the network itself is overloaded.
It's not the servers that go down. That's why Linux is worth looking at even for
consumers. Because consumers may not care about the techie aspects of Linux that make it
interesting to people like me. But I think consumers have a right to an operating system
that doesn't crash or blue screen of death three times a day."
Do you think intellectual property rights
should be changed for software?
What are the limitations of open source
software?
"I used to have a really elaborate set of beliefs about that. But everytime I made
a prediction that some kind of software would not be developable within the open source
mode I turned out to be wrong. So I've stopped maintaining those types of beliefs."
Were you suprised by the Halloween
documents?
"I won't say the surprised me exactly it's more like they confirmed the myth that
we'd sort of believed all along without having a whole lot of really hard evidence. And
then I read the memoranda and it was like "Wow it really is as bad as we thought and
in fact it's worse." Worse in that Microsoft is every bit as malevolent as we thought
and the dirty tricks they are willing to stoop to are every bit bit as low. And in fact
they are more malicious and willing to resort to even cheaper, dirtier tactics than we
thought."
What are Microsofts tactics?
How do you counter Microsoft?
"Sunshine. By sheding light on what Microsoft is trying to do. And by educating
people to demand products that are in conformance with open standards...What we need to do
is get the word out people that if you buy into these proprietary Microsoft protocols as
replacements for the open Internet ones what you are doing is mortgaging your soul to
Redmond for the rest of time."
What advantages does open source
software have for average users?
"If people go with the open protocols, better software, lower prices, higher
reliability. And most importantly because you have the source, you're not in a position of
being completely beholden to a monopoly supplier anymore...So the thing I always ask
business people is 'in what other area of your business would you tolerate being beholden
to a single supplier?'"
How far do you see open source software
going?
"...I think we're looking at 'open source world domination in servers' - that's a
phrase that Linus coined a while back. He gives a speech called 'World Domination 101' It
used to be a joke. Now it looks like it's not a joke anymore. I think we're looking at
world domination in servers, in the most demanding end of the business market, in maybe 18
months or so. The desktop is maybe 3 years out."
Could you describe some of the
business models for open source software?
Why did you personally decide to pick up
the banner of 'open source software'?
What are the biggest
misconceptions about hackers?
"In the context of what we've been talking about probably the biggest
misconception is that all hackers are hostile to intellectual property, capitalism and the
whole commercial thing. And that's not really true, most of us really aren't. That's a
misconception that's come up because historically there has been a very vocal minority
within the culture that was hostile to intellectual property and hostile to commerical
software and the rest of us just uneasily kept quiet... Most of us are pretty comfortable
with commercial, what we're not happy with is proprietary. There's a big difference there.
Most of us are happy to see people make money from software anyway they can, we just want
to see the good engineering outcomes that come from publishing source...
Other misconceptions? The big one that tends to exercise me is the way a lot of
journalists and ignorant people abuse the term hacker. A lot of people have the notion
that a hacker is a person who breaks into computer systems and does mischief. This is not
true. This is an incorrect use of the word that was promulgated by lazy journalists in the
late 1980s. A hacker is properly a programming enthusiast. A person who loves the art and
craft of programming not just for the results, but for the process and who is fascinated
by puzzle solving. The way I like to summarize things is that hackers build things,
crackers break them. When you talk about hackers you talk about people who built the
Internet, and who built unix, and who built the World Wide Web, and who are building Linux
today. And please don't confuse us with those lousy crackers because that annoys us a
lot."
Top of Page
Please direct questions or comments to iota.webmaster@umich.edu.
Last Updated January 29, 1999
