[Discuss] Self-introduction and more on software patent

[bogstad at pobox.com (Bill Bogstad)]

On Fri, Sep 30, 2011 at 11:46 PM, Hsuan-Yeh Chang <hsuanyeh at gmail.com> wrote:
> On Fri, Sep 30, 2011 at 8:23 PM, Bill Bogstad <bogstad at pobox.com> wrote:
>>
>
> Software developers are smart people.? It would not take too much of your
> time to learn the basics of patent application drafting.? The USPTO's
> website also has many introductory tutorials which may be downloaded for
> free.? I would suggest that anyone interested in this topic should check it
> out.? Or, spend an afternoon in Barnes Noble, and you should have some
> better ideas on patents.

I believe that software developers for the most part have zero
interest in patents except to the extent that they are being told that
if they don't they may suddenly discover that it is illegal for them
to use a program that they wrote entirely themselves without recourse
to other's ideas.  Based on your resume you are also a smart person.
How long did it take you to become familiar with the basics of patent
language, application process, etc.?   How long would it take you to
draft a patent application covering every idea in the largest program
for which you have seen the source code?  (How large was that
program?)

> As I said previously, the person knowing the rules plays the game better.
> Filing a patent application can serve different functions, which include: 1)
> to have the idea published as a preventive measure, and 2) to ultimately
> seek patent protection.? The presumption here is that you are filing a
> patent application for protection purposes.? Examiner's would not even
> bother looking at your applications, if you expressly abandon them AFTER
> publication at the 18th month.

I'm confused.  Are you saying that just filing an application won't be
sufficient to make it prior art as far as the patent office is
concerned?  If so, how much more time/money would typically be
required to push an application to the point where it would be
consulted by patent examiners for prior art.  Or are you just saying
that they would just reject them for bad form, but would still use
them for prior art searches?  (Which would be fine from my
perspective.)


[HYC, depending on how busy you are, you might want to skip the rest
of my response and jump to the very end.]

> To change law is one thing, to abolish an entire body of law is another.
> The patent laws and rules has been changed so many times over the past
> hundred years, but none of the changes has delineated too much away from the
> core value of the patent law.

Sure.  In fact, I don't think that patent and copyright law should be
abolished.   Just tweaked.   But, you still seem to be saying that the
Constitution REQUIRES such laws to exist as opposed to ALLOWS them.
Do you see the difference between those two possible interpretations?
I'm not asking about how politically difficult it would be to pass
such legislation through Congress and get a President to sign it.  I'm
asking under what basis the US Supreme Court would decide that
Congress wasn't allowed to do that.

>?Yes, you are right in that the Constitution
> does not recite "patent" nor "copyright."? However, how would you propose to
> the Congress what to do to promote the progress of science and arts by
> granting "exclusive rights" for "limited time" for people's writings and
> discoveries?

Much like the Commerce Clause it can probably mean as much or as
little as Congress wants it to mean.  In the case of software, I've
already suggested that copyright is sufficient and software was
clearly written before patent protection became available so I don't
see how it is unreasonable to suggest returning to that model.

> We should thank Disney for the recent change of copyright law for the
> extended protection.? Mickey mouse is about to enter into public domain
> under the old law.? Can you image what would happen when everyone in the
> world can copy mickey mouse without paying Disney a nickel?

I don't seem to have the same horror of this that you do.   If
somebody wants to make copies of 80 year old black and white films
that nobody but film buffs care about anymore, why should I care.
Frankly, I'm not even sure why Disney cares all that much.  I'm
assuming that Disney was careful to file for copyright for every
subsequent work so even if the original fell out of copyright the more
recent stuff would still be protected.  (If I'm wrong about the way
that copyright and derived works please let me know.)  I guess they
might be concerned about works solely derived from the original
damaging their brand (i.e. XXX rated mouse films), but I really can't
get that worked up about that either.   It's not like Disney didn't
use other peoples ideas as the basis for some of their most most
famous works.  Why should they be perpetually immune to the same
reuse?  In any case, how is this relevant to software patents?

> I know "intellectual property" is somewhat a misnomer, but I don't see any
> problem using the term "intellectual property" to refer to a set of
> intangible assets.

But you used it in a context where we were talking about changes to US
patent law and how other countries would react to changes in those
laws.  Why the change of nomenclature?  By doing so you implied that
all US "intellectual property" complaints would suddenly be at risk
because we decided to tweak one area of US patent law.   I suppose it
is possible that some countries might take that tact, but I provided
what I consider to be a plausible response to such an attempt.  What's
wrong with it?

> Patent and copyright are two different bodies of law
> which are designed to protect different things.? It is often said that
> there's already copyright protection for software, and people don't need
> double protection.? This is a somewhat loose statement as many things are
> protectable by more than one set of intellectual property law.? For example,
> closed-source software codes can be protected by both copyright and trade
> secret.? Can you argue that we already have copyright protection, why trade
> secret?

I don't know.   My knowledge of trade secret is fairly minimal, but as
I see it the fundamental difference is that copyright/patent
protection apply to everyone while trade secret protection only exists
when the parties have explicitly entered into a legal agreement.  If I
find your trade secret protected widget on the road where you dropped
it, I believe that I'm perfectly free to do anything I want with it.
This isn't true when something is protected by patent/copyright.   So
everyone involved with a trade secret needs to be careful.   On the
other hand, trade secrets can be permanent whereas patent (and
theoretically copyrights) have time limits.  In any case, I never said
anything about trade secrets and don't see why I should have to say
anything about how they relate to copyrights when we are talking about
software patents (and to some extent copyright).  What I and RMS are
saying is that copyright appears to be sufficient to encourage
innovation in software and that patents are an unnecessary additional
expense which may in fact retard innovation.

> Also, copyright protection for software codes can be problematic by itself.
> The copyright law is designed to protect writings and artistic works that
> have "originality" -- meaning the works must have original creative value to
> be protectable, such as poem, fiction, paintings, etc.? Certain writings,
> such as legal documents or collection of data (like white page), lack
> originality and thus cannot enjoy copyright protection.? Many people have
> challenged the originality of computer software codes.? Some people have
> argued that computer codes are merely a set of instructions that ask a
> general-purpose machine to perform certain acts.? For example, can anyone
> tell me where's the "originality" of my following code:
>
> ?? main()
> ?? {
> ?? ? ? int i;
> ?????? printf("Hello BLU!\n");
> ?????? for (i = 0; i < 10;i++) {
> ? ?? ???? printf("i = ", i , "\n"); }
> ?? }
>
> I would say the "originality" of my above silly code is ZERO.? Yet, current
> copyright law still protects it and prevents anyone from copying it without
> my permission.? How would you argue for the propriety of this part of the
> copyright law?

Why should I be required to do so?  I never suggested changing what
should be subject to copyright (although I do think the current time
limits are too much).
Are you suggesting that something which you imply doesn't meet your
personal test for what you would want subject to copyright should be
eligible for patent protection?   Again, you seem be bringing things
up that are not in debate.  Personally, I don't care if that is
copyrightable or not.   It's trivial to generate a program that would
generate equivalent results.  I would suggest allowing it to be
copyrighted neither hinders nor encourages innnovation.   OTOH,
software APIs which specify how you call other programs/functions are
apparently not subject to copyright.  I believe the courts decided
that APIs were primarily about methods and not originality and
therefore not eligible for copyright protection.   (Although an actual
implementation of an API could still be copyrighted.)

> Back to patents, I don't want to argue if the patent system would promote or
> demote the innovation of software industry, because I can't.? I agree that
> many existing software patents shouldn't have been allowed in the first
> place.? But in the past decades, almost no one in the open source community
> has reacted to it in a proper and feasible way (namely, react by taking the
> lowest energy path).? Also, almost no one in the open source community seems
> to have really understood how to ride on the current patent system so as to
> protect the interests of open source community, and help the patent office
> to eliminate bad software patents.

Fair enough.   I can't argue that the approach that open source has
used have been that successful.

> My two cents are to file more patent applications by the open source
> community.? If you don't have money, try to get your ideas (not just codes)
> published (preferably by the patent office) to bar others from getting a
> patent against you.? If you have money, try to get one or two patents for
> yourselves and license them back to the open source community.? You may have
> thousands of reasons why you don't want to pursue a patent.? But to the eyes
> of proprietary software vendors, there's only one reason...
>
> Patents are really like weapon competition.? You don't get to discharge your
> enemy's nuclear weapon by asking the united nation to change international
> law.? You get your enemy to discharge their nuclear weapon by owning more
> nuclear weapon.? Should you have none, you are a loser forever.? I know, you
> are likely to challenge the propriety of this analogy, but you know my
> point.

No argument here on the problem.  Still trying to understand the
time/money cost to try it your way.  At least with patents, it seems
likely that we aren't going to get "unlimited" time limits via
perpetual extensions.  BTW, if this is taking up too much of your
time; I would be happy to restrict further discussions to the details
of your suggestion on pushing prior art into the patent office.

Thanks,
Bill Bogstad