[ On Monday, September 1, 2003 at 18:53:54 (+0100), John (TJ) Penton wrote: ]
> Subject: Re: [OT] Re: [Exim] Is exim in trouble?
>
> And therein lies the problem.
> I don't think there is anything fundamentally wrong with patents. They
> allow inventors to see return on the investment they put in to solving
> problems. In this way they stimulate innovation.
Except this not only doesn't work when it comes to software, but in fact
it stifles innovation (unless you count all the creative effort put into
a few select technologies to create totally free and unencumbered
equivalent technology to supplant that which has unfortunately already
been patented to death out of pure greed, e.g. OGG Vorbis vs. MP3).
Often, especially in the digital world, there's no point to trying to
create a competing "standard" as ultimately only one will win out in the
marketplace. Worse it's often the least innovative technology that's
pushed out to the marketplace (perhaps only because it ends up being
less expensive to license).
(and you'll never convince me or most anyone else that the true
innovation in a technology such as MP3 encoders is worth anywhere near
the price its supposed inventors are trying to extort from the whole
world)
Software innovation is stifled by software patents because many
innovative people who are living in jurisdictions (dare I say "under
regimes") where software patents are already prevalent have clearly
stated that they are afraid to work on new technology which may turn out
to have already been "protected" by someone with far deeper pockets than
they have. I know I'm more than a little leary of the threat to my own
free software development efforts posed by the software patent system
used by our neighbour to the south (i.e. the USofA).
> Aside: the alternative to patents is to keep your ideas/inventions
> secret.
Sure, no problem. That is the best solution in the digital realm. If
you're not developing free software then you're not going to let just
anyone read the source code in the first place anyway.
However if someone reverse engineers your algorithms and re-implements
them in either free or for-profit software, perhaps specifically to
compete against you, then the best you _should_ be able to do is to try
to compete fairly with th superior service, quality, documentation,
support, integration, etc., that you should have been able to attain by
having been first in the marketplace with your previously unique
solution.
> In this way patents actually promote the sharing of ideas - the
> inventor discloses the detail of their invention in return for some legal
> protection.
The only thing that needs disclosing in software in order to share ideas
are the pure algorithms, and if I understand correctly there have
already been cases in the USA of greedy non-innovative corporations
patenting algorithms that were long ago published and freely available.
If you don't want to publish new and innovative algorithms then your
lack of willingness won't necessary stifle innovation -- indeed it may
actually do quite the opposite by encouraging others to create competing
software. If someone does a clean-room re-implementation of some
software system then why should they pay a royalty to someone else if
they didn't derive any knowledge from that other party?
> The holder of a patent is required (I believe) to sell
> licences to use that patent at a reasonable rate.]
While that may be true the practices of many software patent holders has
stretched it well beyond all reasonable limits. _Well_ beyond!
> The problem is that (in recent years, and particularly in America I
> believe) patents have been granted for obvious ideas, ideas which are
> already widely assumed, or ideas which have prior art. In this way,
> companies which gain such patents attempt to destroy other (smaller?)
> companies which did not steal the idea and have a business based on it.
That part of the problem is only the tip of the iceberg when it comes to
software (and business) patents.
The biggest problem I see (and I see it directly) is the fear they
instill in developers who may be able to re-invent and re-implement
techniques without ever being aware they were already patented.
> The whole process seems to just feed the lawyers, and stiffle small
> innocent business. What is needed (I think) is just a review of how
> patents are researched and granted, and not the throwing out of patents
> altogether (don't throw the baby out with the bathwater).
I'm reasonably certain that even in other fields than software the fear
of being second to the table with a new idea, even though one may have
done all the work independently, prevents as much innovation as the
thrill of the race encourages. The follies of the race to patent a new
idea is well documented over and over in all fields where patents are or
have been used to "promote" innovation.
Also the immense expense of researching patents over and above the other
R&D costs to develop an idea also seems to be an increasing deterrent to
innovation. It's one thing to go through the academic literature
describing new ideas in ones field of endeavour, but quite another to
try and sift the real meat out of patents which have been purposefully
obfuscated by their authors. If I had to research patents the world
over to see if my independently created free software was truly
innovative or not then I may as well just move back to the farm and hoe
potatoes for a living as I'll be a lot better off all around.
Worst of all, especially to a global endeavour such as software
development, is the fact that there's not just one patent system to deal
with, but one in every potential jurisdiction where one might wish to
vend one's wares (which scares the bloody pants off anyone trying to
develop and maintain free software!!!!).
I.e. I do think the very idea of patents has long been over-extended
past its useful lifetime. It is long past time to throw it out, baby,
bathwater, and all. The baby was stillborn anyway. If companies such
as your own cannot continue to thrive on their own merits then perhaps
they weren't actually making as fair a use of the system as you make
them out to have been doing.
--
Greg A. Woods
+1 416 218-0098 VE3TCP RoboHack <woods@???>
Planix, Inc. <woods@???> Secrets of the Weird <woods@???>
