[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [dvd-discuss] Court Sides With Geac in MainframeSoftwareCase



Roy Murphy wrote:
> I think it's quite reasonable for them to
> require non-disclosure agreements from third-party consultants.


I'm sorry but that just is not sensible.  The software is "published" as
in made visible to persons buying the software.  If by inspecting the
published software one can maintain the software, then their should be
non requirement of an NDA.

NDA's are for exchange additional, non-public information.  For example
if ".002 Consulting" wants to maintain code by poking around at the
symbol tables and decompiling -- there is no way I need an NDA for
reverse engineering.  If however, I want (in order to do my job better,
and be paid better) access to unpublished information (such as a header
file or internal documentation) -- then an NDA is appropriate and
mutually beneficial.  If we allow click-wrap or other license agreements
to interfere with the ability to reverse engineer, then we impede
progress, and ignore the objective reality.  Publish software is visible
to whoever has access.  Just because GEAC management doesn't think
object code is a readable, transparent representation of their supposed
"trade secret" doesn't make it so.  The concept of a "trade secret"
existing within the object code of published software is oxymoronic. 
Certain (half-mad, IMHO) software engineers like reading the object, and
their first response to a problem or challenge is "let me run that
through the decompiler." To assert that object code keeps a secret
preserved in light of these professionals, ignores objective reality.

Now, having said all that, the truly interesting part of this case is
the use of "residual knowledge" by former employees to profit by taking
business away from the former employer.  For that it depends on the
contractual obligations between these employees and GEAC -- and the
enforceability of those obligations. (In certain states, "non-competes"
are not enforceable for example).

.002