Free?

[Steven Schveighoffer]

On Thu, 27 Oct 2011 09:52:18 -0400, Kagamin <spam at here.lot> wrote:

> Steven Schveighoffer Wrote:
>
>> You can profit from the fruits of your invention *without* patents.
>
> If a bigger corporation doesn't steal your invention.

Truth be told, patents don't even protect you against this.

What happens when you sue the larger corporation:

1. They have better lawyers
2. They probably have a patent library they can throw at your other code.

Software patents favor the big corporation, not the lowly developer.   
Consider that it costs about $10,000 just to *obtain* a patent these days,  
much less defend it.

>> >> Add that to the fact that software
>> >> patents are *rarely* beneficial to the community.
>> >
>> > Does the community want benefits at the expense of the inventor?
>>
>> The *point* of patents is to benefit the community.  The price society
>> pays to the inventor is granting a monopoly.  I'd argue that a 17-year
>> monopoly on software technology and algorithms is too high a price to  
>> pay
>> for knowing a "secret" you can't use until it's very obsolete.
>
> Patented technology can be used under terms of GPL right now as example  
> of x264 shows. With GPL patent holder can be sure he still can make  
> money on commercial patent users.

Again, this is a choice of the patent holders of x264, this does not  
universally apply to all software patents.

The holders could have said "free to use in any open source software that  
is available at no charge".  This would cover all open source licenses,  
even ones that can be shipped as binary-only.

There is no inherent clause for GPL in the patent system.  Stop using this  
argument, it proves nothing.

>> >> They are mostly used as
>> >> weapons to stifle innovation from others.  In essence, software  
>> patents
>> >> have had an *opposite* effect on the industry compared to something  
>> like
>> >> building cars.
>> >
>> > Let's look at the H264 technology. Would it exist in the first place  
>> if
>> > its creators had no chance to patent it?
>>
>> What if is not a fair game.  It's impossible for me to say because I did
>> not invent it.  But I believe most people who come up with ideas for
>> software are not in it for the patents.  Even in the company I worked  
>> for
>> which got several software patents, they were an afterthought --  
>> Software
>> is invented to *solve a problem* which needs to be solved whether it can
>> be patented or not.  Did the inventors of H.264 do it for the patents?
>> Maybe.
>
> If it's impossible to say, then your opinion has the risk to be  
> unfounded. If we eliminate patents, it will be impossible to say,  
> whether things became better or not - who knows what inventions were not  
> invented because their authors had no resources for it.

It's impossible for you to say also.  So is your opinion unfounded?

In all likelyhood, a patentless, but trade-secret-enforced video standard  
would have emerged, done well, then an open source equivalent would have  
emerged.  Both the H.264 owners and the community would be in competition  
to see who can make the better video stream, everybody wins.

> I'm not sure h264 solved a problem. Video encoding worked just fine  
> before it. It's just a better algorithm. The experts may be not for  
> patents, but they are paid by big companies which are for profit.

Better algorithms solve the problem of efficiency, implementability, and  
features.  It's like saying DVDs didn't "solve a problem", because there  
was already Laserdisc.

Big companies still have incentive to create a better codec for video,  
because it helps them sell videos.

>> But I firmly believe if software wasn't patentable, we would have
>> equivalent video streams today (maybe even better than what we have),
>> because it *solves a problem*.
>
> Equivalent - yes, but not today. The story of h264 became at 1998, it  
> took years to complete it. It also took quite a while to get Theora  
> right. Innovation in XviD were incremental and backward compatible with  
> stock MPEG4 ASP decoder.

Are you sure the open source community would not have done a better job at  
fixing the codec?  Or any other myriad of companies who currently do not  
hold that patent?  As you loosen the grip on your IP, the benefit you see  
is that more people can help you improve your software -- open source  
development has shown that it works.

>> This is a strawman -- GPL is not required by patent law to be licensed  
>> at
>> no cost for software patents.  The inventors of H.264 have chosen this
>> route, so good for them.  But it is not a benefit of GPL or a strike
>> against boost, it's just what they chose.
>
> Can you make money with boost license?

Yes.  You can make money with any license.

>> You cannot copyright a design.  You can copyright implementation.  And  
>> if
>> you don't make the design public, people have to spend vast amounts of
>> time and effort to just *figure out* your design, then they have to  
>> write
>> their own implementation (which is not cheap).  Meanwhile, you have
>> improved your design to something better and already sold thousands or
>> millions of copies, sucking up all the market share.
>
> So open source is out of game?

huh?

>
>> > The same is for software world. A program may require quite a large
>> > investment before it could be made usable. Let's consider D: who would
>> > get quality implementation first - Digital Mars or Microsoft? If DM
>> > doesn't patent D, it will sell *nothing*.
>>
>> I think if Microsoft decided to implement D, Walter would be the first  
>> one
>> jumping for joy :)
>
> That's only because he doesn't sell D.

You sure about that?  I think Digital Mars would sell support contracts  
for D, and businesses would pay for those if they were to use the language  
in any real capacity.

>> > Even if DM manages to get some market share, it won't survive
>> > competition and eventually lose. IE lost its market share because  
>> there
>> > was more effort put into Firefox than IE.
>>
>> DMC is still being sold AFAIK.  There is always a market for cheaper
>> software, or a more agile software company.
>>
>> One might pay for DMD if one gets specific support.  For example, if I
>> wanted to buy a D compiler for ARM, would Microsoft be willing to
>> implement it for a fee?  Would they even respond to my request?
>
> Windows 8 supports ARM for some reason.

Then some other arch.

>
>> > I suppose trivial patents are also a problem for physical industry as
>> > the wheel patent shows.
>>
>> The wheel patent is a test of a poorly designed patent system (as the
>> article indicates).  It is not representative of most patent systems.
>>
>> See this quote from your linked article:
>>
>> ===========
>> Keogh, who is a freelance patent lawyer himself, says that he applied  
>> for
>> the patent in order to test this new class of new patents. He says that
>> innovation patents are not examined in detail by the Australian patent
>> office.
>>
>> "The patent office would be required to issue a patent for everything,"  
>> he
>> told The Age newspaper. "All they're doing is putting a rubber stamp on
>> it."
>> ===========
>>
>> Note that this is not a trivial granted patent because of a flawed  
>> review
>> process -- THERE IS NO REVIEW PROCESS, ALL PATENTS ARE GRANTED!  This is
>> not a fair comparison of well-established patent systems.
>
> Do you call patent systems granting trivial (software) patents  
> well-established?

Regardless of the quality of the review, at least there *is* a review.   
What I meant was, a review-less patent "rubber-stamping" system is not  
comparable to one which goes through a review process to weed out trivial  
patents.

>> >> When was the last time you did anything with a patented software
>> >> technology except *avoid it like the plague*?
>> >
>> > I would like to avoid H264 but unfortunately I can't.
>>
>> Right, and if software patents did not exist, the web would have
>> standardized on some other video codec, which would be freely available  
>> by
>> now.
>
> I actually avoid h264 in the web :)
> Well, in fact I use firefox and avoid flash, which results in avoiding  
> h264.
> Webm is enough for me in the web.
> I can't avoid H264 for "real" video.

I meant as a developer, not as a user.

-Steve