At 09:00 01/12/99 -0500, Andy Kunz wrote:
>That's where "trade secret" comes into play.  If you don't disclose
>something, it can be considered "trade secret."  People are free to reverse
>engineer that, even if it _could_ be patented but wasn't.

as far as i understand it, copyright is automatic if it applies, but patent
not. so the fact that something could be patented has no legal implications
(other than that it could be patented). or am i wrong here?


>>the picstart+. in neither case there seems to be anything published about
>>the respective interfaces, so i don't understand how the interface could be
>>copyrighted.
>
>Not published doesn't mean something isn't copyrighted.  A language spoken
>between two devices is considered patentable, but usually not copyrighted.
>
>The author of a work is automatically granted a copyright for his work (for
>the source code, or which the object is a "derivative work," not for how it
>speaks to a device).  So the interface, or language, is not copyrighted but
>_could_ be patented.  If not patented, the company can consider it trade
>secret.  Then so long as you reverse engineer it (ie, not steal the
>company's internal documentation) you are free to sell your product without
>recourse from the originator.

so back to the original question about problems with using the (obviously
neither patented nor otherwise published) protocol between the picstart+
and the driver software: when i apply your reasoning, it seems that there
is no reason to expect legal troubles writing a driver software which uses
the protocol to speak to the programmer, and neither a problem with writing
a driver for using the picstart+ to program whatever other chip it was not
originally designed for: because the protocol may or not be patentable, but
in any case isn't patented, and copyright would be on the original code
(both driver software and firmware), but not on the protocol itself.

ge