(I use the car analogy almost every day in my small computer and network
maintenance company, because most of my customers drive, and they
recognize that as drivers they are responsible for wisely using a
technology that they don't understand.)
Hallam-Baker, Phillip wrote:
The voices that need to be heard most in the IETF are the 95% or more
who have difficulty with the technology foisted upon them.
A century ago car breakdowns were invariably considered the fault of
the driver for failling to maintain their automobile sufficiently.
Yes, but ONLY because, a century ago, cars were so unreliable that
only a mechanic would buy one, except for those who were so wealthy that
they could afford to hire a mechanic to maintain it. Of course you then
blame the mechanic.
Today we blame the manufacturer if the car breaks down within the
first four years for any reason.
Yes, but the manufacturers did not solve this problem by themselves.
They were trying to sell an immature device to an unprepared public, to
run on an unsafe platform.
First, the platform -the roads- was made safe for cars by getting
everyone to accept responsibility for improving them. City gov'ts paved
city streets, regional or national gov'ts built highways, and
individuals and companies paid to have their driveways paved. Don't
forget that this "improvement" came at the expense of making these roads
LESS useable by the previous standard, the horse (If you don't believe
that, try running barefoot on grass or dirt for a minute, then shift to
concrete or asphalt. Your joints won't like it, and neither do horses).
The vast majority of internet users run our products on an unstable
platform. I believe that, until the world's most common OS is made
stable (or is replaced), the IETF's efforts will be ineffective, just
like better cars kept failing, on rutted dirt roads. The IETF has no
handle on the creators of that platform to make them fix it. That's
reality, and many people are trying other platforms, while the
manufacturer of that platform concentrates on winning this fight by
advertising, instead of fixing the problem.
Next, "cars were made more reliable" by governments and the insurance
industry working together to get untrained/unsafe/argumentive drivers
and cars off the road, by licensing both cars and drivers, and
periodically inspecting both. In our case, we have a lot of drivers on
our road that try to cause wrecks, and you cannot blame the factory for
the damage that results.
Last, the builders started making special-purpose vehicles. "Our
passenger cars are only to be used on paved roads. If you want to go
off-road, you should buy this line of vehicles over here. It's not as
nice, and it's a lot more expensive, but it won't break down as fast."
Manufacturers design cars that are expressly designed to deal with
users who are by any standards unreasonable.
Yes, but our insurance and court system accepts that the warranty is
void if the user does something that he shouldn't (tow a 20-meter boat
with his Ferrari, put leaded gasoline in an unleaded-only engine, take
his Mustang dune-hopping, or get drunk and go driving). A large part of
our solution will have to include education for prospective drivers, and
practical punishment for wrong-doers who cause wrecks and vandalize
other cars.
Any machine built by humans can be broken by humans, whether they are
sufficiently evil, or simply sufficiently stupid. You are overly
imaginative if you think that it is possible to build something that bad
players cannot tear down.
To be blunt, conspiracy theorists can make a decent case that the
Internet is already being used for information and economic warfare
between nations and cultural groups, and even economic groups inside
nations. Do we blame the protocol's manufacturer, when anyone, anywhere
in the world, is allowed to do anything he can think of, to your
protocol stack? And, would you trust someone to test and license
prospective Internet users?
Sooner or later those standards are going to apply to the Internet.
Yes, but we're going to need some outside help making it happen.
I want that to happen sooner and for the IETF to be an enabler of
that transformation rather than an obstacle.
I'm with you there, but simply crying "It's the designer's fault!"
will not solve this problem.
--
Unable to locate coffee.
Operator halted.
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf