OT: A call to arms (sort of)

From: William Donzelli <aw288_at_osfn.org>
Date: Wed Jul 7 14:47:11 1999

> Of course they do. But I won't accept that the service manual gives them
> that much help. And I can't think of a case where a service manual has
> been the cause of a clone.

Not clones - that is lawsuit country - but certainly improvements. Just
about every other week *EE Times* has an article on company X dragging
company Y to court for stealing technology. Documentation is often the
main bit of evidence, including, yes, service manuals for the techs.
 
> Given a working example of a product you can 'reverse engineer' it to
> better than the level of information in the service manual in a couple of
> weeks. Using only a DMM and a simple logic analyser. The former to get a
> rat's nest of the connections, the latter to assign useful
> names/functions to pins on gate arrays. Believe me, it's not hard.

A "couple of weeks" can make or break a product, back then and today.
Time to market is often the overriding factor in any engineering project.
Shaving off some time by snatching a manual and getting clues, sometimes
just subtle hints, can make a lot of difference. Also, engineers are not
cheap, and having a small team buzzing out circuit boards is not exactly
cost effective.

Also, there are lots of things that show up in service manuals that are
completely non-obvious to someone examining a board. Diagnostic ports,
used during the manufacturing process, are sometimes are very well hidden
in the normal vias of a circuit board. One thing a manufacturer does not
want a competitor to get at is the JTAG scan ports, as they
really let someone look deep into a machine.

Another thing found in service manuals are the tricks that allow
technicians and engineers to crack passwords and use backdoors. Imagine
the fun a company would have trying to patch a PR disaster because some
high school kid keeps fooling around with spanning trees and access lists
on switches and routers, with techniques available in the back of a manual.

There is a lot of sensitive information in these service manuals that can
really hurt a company if it falls into the wrong hands. The easiest
solution, and probably the best as well, is to restrict the manuals.

> Therefore, the service manual isn't _that_ valuable. Not having it
> doesn't stop the above. And from what I've heard (and observed on classic
> computers), you would be very unwise to rely on a manual. Errors creep
> in. Things are missed out. Important details of things like signal timing
> can only be deduced from the product itself. In other words, even with
> the manual you are going to want to dismantle and analyse a real machine.

Oh yes, but lots of clues and shortcuts can be found in the pages.

> I will assure you that restricting the service manuals like this doesn't
> hinder 'the competition' that much at all. What it does hinder are
> non-official service agents, though. That, IMHO is the main reason for
> restricting them. And I am not happy with that.

No. Companies simply do not care about "non-official" service agents
anymore because there are not that many of them to worry about, and the
companies can always clobber these "non-officials" on replacement parts
costs, technical support, and the continuation of warranties.

There is a very good reason why one of the first pages in any military
electronics technical manual is devoted to destruction of the equipment
and the manual in an emergency. War and the market aren't much different.

William Donzelli
aw288_at_osfn.org
Received on Wed Jul 07 1999 - 14:47:11 BST