MySQL author gives view on Patents

Michael Widenius, the main code-contributor to MySQL says that that software patents are bad for the community, should not be awarded for software ideas, user interfaces, standards or interoperability. Moreover if a patent were to be awarded, it should only be for major work (something that can't be done by a single individual within 50 years) and last 2-3 years at most.

Who the fsck are you, troll?

[pingouin]

Learn how to spell "fascist" - better yet, learn what it means - before using the word. Then maybe I (and I'm not a Free absolutist ...yet) would pay attention to your adolescent nonsense. If becoming a "billionaire" is that important to you, maybe you should stop trolling and begin schmoozing a few drug cartels.

--

mysql non-free -- so?

[Matthew Weigel]

Yeah, it would be nice if mysql were really free. But it's not, and that's the author's choice. I would think that that would not invalidate his opinions on software patents, however. In fact, I think it's telling if even some proprietary developers think such patents suck ass.

Who determines how "hard" the achievement is?

[sql*kitten]

that's very naive. communism and fascism are simply variations of collectivism: the idea that the wishes of the individual are worthless and exist only to serve the state/the church/whatever.

Who determines how "hard" the achievement is?

[cthonious]

no one is going to become a billionaire providing service and support to RedHat users).

Actually, I think you could, just need more users.

... and for the record, nobody needs to be a billionaire, especially you. Sorry.

Patent duration should vary by category

[A nonymous Coward]

200 years ago, even 100 years ago, 17 year patents made sense. Everything was a mechanical invention, society was slower moving.

Some thing still apply. Pharmaceuticals, for instance, take so much money and time to develop and get regulatory approval.

But business practices? XOR cursors? auctions? These are obvious to anyone even not so skilled in the art. Other patents, like the suid/guid bit, strike me as a true clever invention (altho I don't know the state of the art at that moment) but surely a couple of years would be sufficient.

--

Who determines how "hard" the achievement is?

[RedOctober]

"All communists... are facsists [sic] deep down".

This A/C went on to prove that true is false, night is day, white is black, before being run over on a zebra crossing. [Apologies to Douglas Adams]

Who determines how "hard" the achievement is?

[jinx]

I don't like what I'm seeing here.

First of all I'm all for OSS and its proven its case, time and time again in regards to functionality and stability. Its certainly managed to keep large corporation in check with the new pressure of producing good code for a better price.

My problem is this "We don't need a new billionaire" or "you don't ~need~ to be a billionaire approach". This in its self ~is~ communism. First of all, we better thank those non government oriented corporations and individuals for
~being~ billionaires, which just so happens to layout the fundamental basis of economy which allows smaller companies to grow, for people to have decent jobs that they can feed their family with conformably.

Not everyone want's to be a government employee making the same salary or a salary geared towards popular acceptance.

You want to be a billionaire, then be one. I know its will circulate into my pocket and everyone's pocket... its the law
of economy presuming we don't chase their investment into other countries due to tax punishment for having that money.

The patent laws have gone off the wall, this is true. But to say one cannot be awarded a patent based on the fact that they are an individual or haven't spent 50 years in development is complete non-sense and "is" communism. At least in America, this is a land of the "individual", not organizations, corporations and government. Let it be up to the individual what to do with his/her invention or patent. But I would agree that no one needs a patent on an ashtray umbrella.

In my opinion, Linus should have full rights to patent linux. Whether he wants to charge for it or give it away under a gnu license, that "his" business, not some community or a group of gimme free mongers.

You demand free stuff, then sit "your" ass down, code it "yourself" for free, and give it away. Don't tell "me" I can't patent/sell and commercialize my software.

Note:
My comments are to only those who apply.

I wonder...

[Athos]

what would happen if this continues. Eventually, every possible algorithm and informational concept will be patented.

When governments and corporations cannot possibly know even a fraction of the patents they are violating simply by doing business and without intention, I wonder how long it will take to get this mess cleaned up.

You know, wait for a critical mass of this stupidity, then declare all software patents void or good only for 6 months or somesuch.

Or am I dreaming again?

I wonder... (governments are special)

[Athos]

Unfortunate.

I guess the lawyers will have work for life. Several lives, actually.

"Yes, I'm a startup, and I need $1 billion in venture capital." "Why so much?" "Before I can hire desingers I have to hire patent lawyers."

Is 'Intellectual Capital' beneficial for society??

[wakebrdr]

Is it beneficial for society when we encode intelligence and then regulate the use of that intelligence?

If I can patent an algorithm, why can't I patent one such as, 'Remove keys from pocket, use key to open car door, get in car, fasten seatbelt, etc...' ??? Is the validity of a patent only governed by how clueless the patent office is about the industry the patent applies to?

If you consider that computer code is simply encoded intelligence, at what point are we limiting the technological innovation of the human race in the name of Intellectual Capital?

More bangs for the buck!

[takshaka]

This is so exciting! I love to punctuate! Even when I write a long sentence that couldn't possibly be spoken with the emphasis that I think it should have, I still end it with an exclaimation point! Bang! Bang! Shebang#!

Er...

But, really, patents should only be awarded for things that are specific and truly innovative, not for "the transfer of music from one computer directly to another" or some other such nonsense. The idea is to protect the inventor, not screw the world.

-tak

silliness

[geophile]

In software, a patent is one of two things: 1) A marketing gimmick, which doesn't fool anyone but investors and the lower IQ part of the customer base. 2) A bargaining chip. If your company threatens my company by waving a patent at us, then my company can respond by waving a patent back at your company.

Your argument makes some sense in the physical world. But in software, there are lots of ways to solve a given problem, and it's just too easy to code around a patent. Especially the notoriously lightweight software patents that are routinely awarded. Also, by the time a patent is issued (probably 2-3 years after the application date), the patented technology is likely to be obsolete.

Do you really think Microsoft or IBM got rich from their patents? Or that Microsoft would not have written Word or Windows but for the protection of its patents? Or that patents are really preventing anyone from competing with those companies? I suggest you tune in to the Microsoft antitrust trial.

No, the only guys making money from patents are patent attorneys. Three years ago I left my job and wrote some software to start a new business. My major expense was paying a patent attorney several thousand dollars to get an "opinion letter" which does nothing but buy some insurance in the extremely unlikely case that a patent infringement case does get to the point of a court verdict. That was quite enough for me; I couldn't even begin to contemplate obtaining a patent on my own work.

Here's a suggestion that should appeal to the open source crowd: If you have a patentable idea, PUBLISH IT. Then no one gets to patent it.

WHO CARES? PostgreSQL that's who!

[memra]

There is an OSS alternative to MySQL and it is a darn good product too.

Try out PostgreSQL [postgresql.org]. It is an object-relational database and I have personally used it with PHP, Python, Java and through the ODBC driver on Windows machines. I have set it up on Linux machines, FreeBSD 2.1 and FreeBSD 3.0 machines. And I have used it in commercial projects and will do so again.
--
Michael Dillon - E-mail: [e-mail]