Re: Patent on aligned memcpy

[N8TM at aol dot com]

In a message dated 8/20/98 12:19:43 PM Pacific Daylight Time,
geoffk@ozemail.com.au writes:

> finding
>  prior art for software is not something the patent office is good at.
A few years ago, a co-worker of mine who was a patent attorney informed me
that patent attorneys are obligated (at least in USA) to assure themselves
that a suitable effort was made to determine the validity of a patent
application, including determining whether it violates the "prior art"
provisions of patent law.  The filing attorney is liable for negligence if a
suitable effort was not made; worse, in a case like this, where it could
probably be proven in court that he ought to have known he was filing an
invalid patent.

One of the basic principles of turbomachinery aerodynamics was published in a
French paper in 1934, invented independently by Frank Whittle in England
shortly after and patented by him, although it was already in use in the USA.
Supposedly, Sir Frank performed his own research by conversing over the back
fence with his neighbor, the chief steam turbine designer at English Electric.
Assuring himself that his neighbor, who presumably knew all there was to know
about his subject, knew nothing of this, Sir Frank concluded that he had hit
upon something new.  When Frank came to the USA to consult, he found out
quickly that his invention but not the existence of his patent was well known.
He threatened legal action but had the grace to admit his mistake when shown
the French publication.  But, 10 years later, the US patent office issued the
same patent to an obscure US consultant who threatened to sue anyone who
mentioned this principle in a publication, no matter that it was already at
least 20 years old.  His patent attorney, however, was never available to
discuss the merits.