Re: [RFC 0/1] Contributing a compound object to the libpthread

[Patrick McGehearty <patrick.mcgehearty@oracle.com>]

On 6/19/2018 8:09 AM, Adhemerval Zanella wrote:
>
> On 19/06/2018 09:42, Oleh Derevenko wrote:
>> Adhemerval,
>>
>>
>> On Tue, Jun 19, 2018 at 3:07 PM, Adhemerval Zanella
>> <adhemerval.zanella@linaro.org> wrote:
>>
>>>> Regarding the note of DJ Delorie -- the patent covers the
>>>> synchronization method (the approach), not a particular
>>>> implementation. You will not be able to create an implementation of
>>>> yours without violating the patent anyway. So, don't worry about the
>>>> effects of looking in the code.
>>>>
>>>> That said, I understand that my idea is not going to be accepted at
>>>> this time. I would like to apoligize for you time I have taken. In
>>>> case anyone is going to change their opinions on the matter feel free
>>>> to contact me.
>>> To be clear: it is not your idea that is being rejected, but rather
>>> your imposing rules we need to accept so you can disclosure your idea.
>> All the patent texts (except for those that contain sectecy matter)
>> are publicly available from the USPTO or similar corresponding
>> national/international patent management offices. In particular, there
>> is a search form at the USPTO website you can use to view any present
>> or past patents and published applications. The sole idea of patents
>> is to create and save descriptions any other skilled individuals can
>> later use to re-create the inventions. Just, an exclusive use period
>> is granted to authors as a reward for their binging the inventions to
>> public. Later, after patents expire, they can (and should) be used to
>> advance everyday life of all the humanity (or other more humble
>> purposes -- whatever they are good for). In my opinion, most of the
>> patents in the database are good for nothing and are just wasted
>> money. Or they can be treated as enforced losses businesses have to
>> take due to the competitive environment they exist in. But sometimes
>> there are things that are worth the application/maintenance fees paid.
>>
>> So to resume, there is no problem in disclosing the idea. And it
>> already was disclosed here and is permanently being disclosed at the
>> USPTO website. I posted a link to it.
>>
> I am far from Copyright and patent expert and correct if I am wrong,
> but my understanding is for GNU projects all your description does not
> mean anything if you, the one holding the patent, do not sign a
> Copyright assignment for FSF so any projects which the idea or code
> derived from it are protected for Copyright enforcements by third
> parties (you inclusive).

I also am not a Copyright or Patent expert, but I can share
what corporate lawyers for multiple large companies tell engineers:

"Don't look at or search patent applications without direct
instructions from higher management."

It's due to the concept of "willful infringement". If it can
be shown that your company has "willfully infringed", any court
finding of patent infringement will result in "enhanced damages".
If I do a patent search that happens to include a patent which
some other engineer at my company coincidentally infringes
while producing a high value product, the patent holder might
claim my company willfully infringed, even if I never spoke
to the other engineer. It is difficult to prove the negative.
Given that some high profile patent cases have resulted in billion
dollar judgements, you can see the potential seriousness.
Willful violation of company policy that has such high
consequence is usually grounds for dismissal.

For me and most responsible engineers working large companies,
it means:  "I don't look at patent applications for fear of the
possible effect on future lawsuits."  Sad, but true.

- Patrick McGehearty