🚀 go-pugleaf

>>>> Though if two (or more) people come up with the same idea
>>>> independently of one another (and none knowing of the
>>>> work/existence of the other(s)), does this not hint that the patent
>>>> might be invalid because of the requirement that it be innovative
>>>> and non-obvious (to a skilled practitioner in the field)?

>> No. there are loads of examples of simultaneous invention-- eg the
>> telphone is one, and in another area, Evolution is another.  That two
>> people independently come up with an idea is not necessarily an
>> indication of obviousness.

mark> Maybe what happens is that something becomes obvious (to a
mark> suitably skilled person) once some different invention becomes
mark> known or some data is published.

No, what happens is that there is more than one bright person in the
world working in the same field and all are in a position to get to the
same conclusions. Conclusions don't happen; they are often the result of 
the state of the art. As somebody once said, when it's steam engine
time, it steam engines.

The patent systems, and this is something lots of people forget, does not 
reward invention -- for invention is at best hard to define, and its
value is even harder to define.

What the patent systems reward is the publication of trade secrets; for
publication is easy to define, and if something is a trade secret
probably it has some value.

This also acknowledges that what really benefits society is not
invention in itself, but the publication and circulation of the
information embodies in inventions; the reason is that usually
inventions eventually happen (when it's steam engine time, someone
invents steam engines), one way or another, but if everybody keeps
inventions as trade secrets then further progres is much harder.

In old English patent law for example invention patents were awarded to
the first person to publish a valuable trade secret, *even if s/he was
not the inventor*, for it is publication that is the overriding public
interest.

Thtat patents are granted usually to the first (inventor) to publish
nowadays in most countries (in the USA it's different) serves this
interest: if two companies invent the same thing one after another, both 
will feel a need to patent, and thus publish.

If the first inventor does not publish, and keeps the invention secret,
anybody subsequent reinventor that publishes and patents not only makes
the trade secret moot, but can prevent the original inventor from
continuing to use the trade secret.

The system is based on the idea that publication of valuable trade
secrets is in the public interest, and that it must not be compulsory
but rewarded.

The flaws of any patent system are that inventor lobbies try to raise
the reward as high as possible, and single inventors try to pass
everything they can as valuable trade secrets.

The public interest would of course be the opposite: to offer the
minimum reward necessary to induce inventors to publish their trade
secrets, and to grant the reward only to those publish actually valuable
trade secrets.

The first issue is one of public policy, and here the public interest is
usually screwed pretty hard: inventors are a big, rich lobby and patents
are a reward that does not cost anything to the state budget, only to
society as a whole. The result is that the scope for patentability and
the duration of patents have steadily increased over time (often
retroactively!), as political factions find it a relatively obscure way
of granting large amounts of money to their lobbies.

The second issure is supposedly left in the care of the patent offices
and of the courts: patent offices are supposed to refuse granting a
patent to obviously unvaluable trade secrets or things that are not
trade secrets at all, and the courts are supposed to cancel patents for
those that not so obvious.

However this does not work either that well:

* Currently most patent offices finance themselves with patent
  application fees, and every time they reject a patent application they
  discourage further application and thus further sales (each patent
  application is a sale), and thus their own survival; therefore many
  patent offices tend to grant patents whenever possible, to encourage
  further applications.

* The courts then are the last call, where invalid patents should be
  canceled even if the patent office had granted them to make more
  sales. Unfortunately patent holders have large advantages when an
  attempt is made to cancel a patent; the first one is that obtaining a
  patent is much much much cheaper than having one canceled by a court,
  and the second is that in order to have a patent canceled the burden
  of proof is on those trying to have it canceled.

  Eventually this burden of proof amounts to having the money to pay
  expensive, famous experts to testify that the trade secret it
  publishes is rubbish; while the patent holder usually has more money
  or a greater incentive to spend lots of it in even more expensive and
  famous experts that testify that the invention described in the patent
  was actually very valuable.

The end result is that the best way to act is to patent as much as
possible: first of all it's easy to patent almost anything, even the
obvious, and if someone then sues you for infringement of one of your
dubious patent you can try to find out they are infringing one of your
own, and thus retaliate.

The end result of all this is that two types of entities profit from the 
system:

* if a company has the money to invest in a scattershot of patents one
  can then usually enter in crosslicensing agreements with competitors
  and avoid trouble, and keep out lesser competitors that cannot afford
  that;

* if someone has money to invest in legal fees, one can *buy* a
  scattershot of patents and then sue them for infringement; there are
  law practices that do nothing but buy and litigate patents, and they
  usually become very very rich.

In other words the patent system ultimate effect is not exactly in the
public interest, as it results in restraint of trade and the reward of
rentiers.