21 Jun
2004
21 Jun
'04
3:49 a.m.
Product documentation and schematics
First hint "never divulge all of the details in a patent".
Actually I know when we applied for patents and provided details of a
product we deliberately left out some key nuances and facts so that
somebody else could not totally reengineer the product from the details.
Much of the process of the internal workings and timings was not
detailed.
Our product was originally created and manufactured by some astronomical
engineers; they/we went out in the field to customer sites and made them
work. The system was based on a PDP-11/04 with all of the programming
stored in a PROM board. The entire software was one large program, no
OS. Original test units had been PDP-11/05's with core memory. Output
was on a VT52 with a built in hardcopy printer.
There were very accurate drawings and documentation however when it was
turned over to the medical manufacturing team the biggest problem was
the difference in the experience levels of the personnel and the
tweaking required to make each unit "work".
There is a big difference between 10-15 prototypes and 100 regular
customer units. It's also easy to retrofit, and keep 10-15 units
synched up, almost impossible with 100 units.
Support is also a problem, initially we had the designers and
engineering staff answer any support calls and troubleshoot problems, we
later converted to customer support to screen the calls. With 10-15
customers the initial adopters are motivated, usually analytical and
willing to work on problems to solve them. Regular customers just want
it to work.
"Institutional memory gets lost"
Mike
21 Jun
21 Jun
5:10 a.m.
>>>> "Mike" == Mike McFadden
<McFadden> writes:
Mike> Product documentation and schematics First hint "never divulge
Mike> all of the details in a patent".
Mike> Actually I know when we applied for patents and provided
Mike> details of a product we deliberately left out some key nuances
Mike> and facts so that somebody else could not totally reengineer
Mike> the product from the details. Much of the process of the
Mike> internal workings and timings was not detailed.
If a person "skilled in the art" cannot replicate your invention from
the patent description, then that patent is invalid.
I suspect this is hard to enforce, but (IANAL) that's what I'm told
the law says.
paul
10:38 a.m.
Product documentation and schematics
First hint "never divulge all of the details in a patent".
Thankfully nobody told HP that in the early 70's. The patents for the
98x0 series read like service manuals with schematics and ROM source
listings (the one of the 9830 is something like 900 pages long). The
info doesn't quite apply to production machines, but it's a heck of a lot
better than nothing!
Actually I know when we applied for patents and provided details of a
product we deliberately left out some key nuances and facts so that
somebody else could not totally reengineer the product from the details.
As I understnad patent law (and IANAL), if somebody 'competent in the
field' can't recreate the invention from the patent, then the patent is
invalid. Of course this would probably involve a long and expensive court
case, but anyway.
-tony
22 Jun
22 Jun
10:10 a.m.
As I understnad patent law (and IANAL), if somebody
'competent in the
field' can't recreate the invention from the patent, then the
patent is invalid.
I'm not sure that's quite it. I thought that the patent
would be invalid if somebody "competent in the art"
could be reasonably expected to have thought of it
(but not necessarily done it) so you'll obviously
never see a patent for moving a cursor by using
an xor :-)
The few patents I've seen try to be as broad as
possible (without going quite far enough to be
thrown out as too broad) and give a few specific
examples.
So if you find that you can make your calculator
last for years on a single sunbeam if you use
Peruvian Bat Guano as the keyboard membrane,
what you do is patent the use of Peruvian
Bat Guano as a keyboard membrane in any way
shape or form. You give an example which
involves a 0.1mm layer of said Guano and
claim (correctly) that it improves battery
life. Since your patent covers all use of
such guano, you don't bother to mention
that the best guano is that collected
during the summer nights, that a 0.15mm
layer produces much more beneficial
effects and that only the guano
collected by a specific tribe doesn't
smell.
Of course, by the time the patent agent
has finished with the application you
won't find simple words like "guano"
or "calculator" or "keyboard" in it
anywhere at all. I always assume that
if I can understand anything at all
about a patent on first reading,
then someone made a big mistake
somewhere :-)
BTW: In case bat guano (peruvian or
otherwise or indeed any biological
waste product [babel fish anyone?])
does prove to be useful
in any form of computing device,
then please remember that this
message counts as publication,
so you are too late to make a
fortune from my ideas!!!
Antonio
--
---------------
Antonio Carlini arcarlini(a)iee.org
10:18 a.m.
>>>> "Antonio" == Antonio Carlini
<arcarlini(a)iee.org> writes:
> As I understnad patent law (and IANAL), if
somebody 'competent in
> the field' can't recreate the invention from the patent, then the
> patent is invalid.
Antonio> I'm not sure that's quite it. I thought that the patent
Antonio> would be invalid if somebody "competent in the art" could be
Antonio> reasonably expected to have thought of it (but not
Antonio> necessarily done it) so you'll obviously never see a patent
Antonio> for moving a cursor by using an xor :-)
Antonio> The few patents I've seen try to be as broad as possible
Antonio> (without going quite far enough to be thrown out as too
Antonio> broad) and give a few specific examples.
There are a bunch of requirements -- as I understand it (IANAL), they
are roughly like this:
1. It must be useful.
2. It must be "not obvious to one skilled in the art" (that's the
one you mentioned).
3. It must be novel.
4. The description has to disclose the invention sufficiently for
one "skilled in the art" to reproduce it.
5. The description has to be of a "preferred embodiment", i.e.,
one you'd pick, not a variant you'd reject in favor of something
better you didn't describe.
6. You can't claim what you didn't describe.
That's not to say that reality is all that close to this, but this is
what I've been told the rules are.
paul
7489
days inactive
7490
days old
4 comments
4 participants
participants (4)
-
arcarlini＠iee.org -
ard＠p850ug1.demon.co.uk -
mmcfadden＠cmh.edu -
pkoning＠equallogic.com