So it looks like open source is turning a new corner and headed further into the giant cow pasture in the sky. The newest twist?? Steal someone's product, put your name on it, and open source it before anyone notices. While this used to be called theft open source makes it OK.

Open Source. Slave to the economy.

Oh please, don't be rediculous.

Odds are that ZFS either doesn't violate these patents, or if they do, the engineers (and thus Sun) who developed it were unaware of the patents at the time. No, NetApp is simply setting up a very classic response to Sun's lingering threat. Look for a nice out of court settlement that sees both parties agreeing not to sue each other or their customers.

Though, truth be told, I wouldn''t much care if there are some patent violations here - software patents are bunk, and should be eliminated. The only thing here that really disturbs me is Sun's behaviour - both of their legal team over the last 18 months (assuming NetApp is telling the truth), and of their PR department. I mean, their response had so much spin on it I'm surprised it didn't impact the earth's rotational speed.

-daniel

I'm puzzled as to why anyone would be so willing to join the likes of other anti-open-source idiots such as Enderle, DiDio and Lyons. Oh well, to each their own.

people are so conservative about file systems at least the Linux community tends to wait and natter about it for years other things go out very quickly but file systems if someone starts a patent lawsuit about that the work you did can be completely wasted Solaris can be the guinea pig for this thing I will be watching with a tiny bit of interest. As far as the whole pay up or else thing SUN pulled I think we have a legal department that needs some castor oil applied and NetApp seem to be coming foreward with a spoon.

As ZFS source is open, NetApp will have no problems at all pointing to the parts of the code that violate their patents. NetApp will also be able to indicate the parts of the independent BSD and Linux implementations that are covered.

As with SCO, put up or shut up.

Are you sure you aren't Bill Gates or Steve Ballmer? You certainly seem to be toeing the company line.

I haven't tried to look at the patents, but we all know that a large portion of software patents are worthless. In the web postings, it seems that NetApp is making a big deal about copy on write. While not applied to filesystems, copy on write has been a software technique since the Multics days, at least. That's the 1960's, kids. Techniques for maintaining consistency of trees are old, too. Patents are not supposed to be granted on obvious combinations of well-known techniques. That's what engineers do all day.

I think that part of the patent examination process should be to give the problem the patent claims to solve to a bunch of engineers, and if they come up with a solution that is anything close to what the patent wants to claim, that should be enough to reject the patent as being obvious.

Not only were the Sun engineers working on ZFS fully aware of NetApp's WAFL filesystem, there's reason to believe they took the fundamental ideas from it and stuck them in ZFS. I suggest you read Hitz's blog entry - especially you, g lane - and the ZFS white paper he links to.

It looks like NetApp are more than willing to point out specific aspects of ZFS that they think infringe on their patents, and they're not just making a big deal about copy-on-write - it's about the specific way ZFS implements copy-on-write and filesystem consistency. To quote the Sun whitepaper:

"The file system that has come closest to our design

principles, other than ZFS itself, is WAFL[8], the file

system used internally by Network Appliance’s NFS server appliances. WAFL, which stands for Write

Anywhere File Layout, was the first commercial

file system to use the copy-on-write tree of blocks

approach to file system consistency. Both WAFL

and Episode[4] store metadata in files. WAFL also

logs operations at the file system level rather than

the block level."

GRUB is in there but I cannot see any source code provision.... or is this version not under the GPL?

I got this from their latest OnTAP release tree:

/etc/boot/NPM_FCSUM-mips.sha1.asc:d7b1a4e3dd5f20b2821e942df7e506fbe8a2108f /etc/boot/grub/stage1

/etc/boot/NPM_FCSUM-mips.sha1.asc:3695fcf4bb62fedf62025ea33453d8ccd29ba87f /etc/boot/grub/stage2

I cannot find any mention of source code on netapp.com even if I login to download the code...

The U.S. Patent system and the patent systems of those who followed the American method of handing out patents on all and everything is a big steaming pile of shit. It stinks and it's starting to hit the fan. There are a multitude of well publicized patents that have earned infamy but I doubt that those numbers represent more than the very tip of the iceberg. The problem is that the iceberg is in the shipping lane and when it begins to break up it's going to wreak havoc.

Recently there have been signs that the American government and judiciary are aware of the problem but the hands off, lassie fair approach underlying American beliefs seems to think a monetized, adversarial system can straighten it all out. OTOH when the law suits start flying hot and heavy and are interdependent then the genius of it all will be apparent in the ability of the American government to launch a new industry comparable to the military industrial complex. Maybe it all leads back to political contributions. It's a great circus act to follow.