Re: License for Google's patent

http://mail-archives.apache.org/mod_mbox/hadoop-general/201004.mbox/%3C121803A3-CFB9-489B-96EF-027234E55D25@apache.org%3E

by helwr • 16 years ago

30 41 16 years ago

mail-archives.apache.org

jacquesm12• 16 years ago

For some reason this strikes me as really weird.

We should be happy that google grants a license for something they shouldn't have had a patent on in the first place?

Continued generosity indeed.

fierarul2• 16 years ago

Didn't Google actually introduce mapreduce into the mainstream and thus Hadoop is just an open-source clone of Google's system ?

Given that patents are a way of live in the US, why wouldn't Google patent this ? Would you prefer someone else instead ?

On a related note -- how many patents does the Apache Foundation own ?

jacquesm12• 16 years ago

Mapping and reducing have been part and parcel of programming for the longest time. Packaging the concept in a framework and patenting it should not have been possible.

jfoutz• 16 years ago

C* and zeta lisp both supported parallel map reduce at the language level.

brainlounge3• 16 years ago

Apache owns zero patents.

emarcotte2• 16 years ago

Indeed, however, I suppose another company could have applied for and gotten the same patent and started trolling with it.

jacquesm12• 16 years ago

But that's not really an excuse. Bogus patents should not be applied for and when they are applied for they should not be awarded. Applying for them on the grounds that someone else will do it is not a valid reason, and google is not exactly the most transparent company in the world. If they're so generous they should donate the patent to a neutral party.

Most - if not all - software patents are bogus.

gfodor2• 16 years ago

You can't be seriously suggesting that Google, a multinational corporation, cease to apply for software patents? The problem isn't that people are applying for them, it's that they're being granted.

Or as they say, don't hate the player, hate the game.

jacquesm12• 16 years ago

About 15 years ago I was in a position to apply for a patent on the 'streaming webcam', live video over the internet. I found a way to stream video to netscapes new browser ('Navigator'), without using plug-ins.

I expressly did not do that because to me it was an obvious extension of being able to do 'stills'. After all, simply making it move does not require much ingenuity, merely a couple of nights of hard work.

Looking back I'm happy I didn't do that. I refused to 'play the game', even if it could have made me a boatload of money I still stand by that decision.

Whether google is a multinational or a one man shop, to apply for a patent simply because of scaling an existing algorithm up to 'cluster size' in my opinion is not the nicest thing. For them to 'grant' licenses like these and then to receive the kind of groveling gratitude present in that email in return is really disgusting.

madair2• 16 years ago

At a certain point one needs to refuse to play an unethical game. Until then, everyone blames everyone else for their own unethical behavior.

instcode• 16 years ago

Not really. Pinch & Zoom is a very basic gesture that any multitouch screen devices need. Then Apple came & just destroyed the game...

richardw3• 16 years ago

No large company can afford to ignore IP these days, partly because they might need a defensive portfolio to negotiate with, when another large company sues them. Look at Apple & Nokia.

wingo• 16 years ago

What does that mail really mean? Does the patent grant extend to end users of Hadoop, or just the Apache foundation? What about users who patch their Hadoop? Is it revocable?

I'm sure there is good will here, but the system is disgusting.

brainlounge3• 16 years ago

Google granted the license under the Apache License terms, and through distribution of Hadoop these extend to the end users of Hadoop under the very same Apache license. (AFAIUI, IANAL)

Re: License for Google's patent

Related Stories