|
Authored by: Marc Mengel on Monday, April 08 2013 @ 08:08 AM EDT |
Well, There was a lot of rambling in there about hardware shift registers (which
could be one of these or one of those types, or any other type of hardware)
being attached to some central timing clock for the whole system. I'm pretty
sure a couple of PC's with Hadoop installed on them don't suddenly develop such
hardware.
Also, so far as I know, there is no error reporting for the memory on the disk
drives other than that listed in the SCSI specifications (which I'm guessing
predate this patent).
Basically, this reads as a hardware patent, and Hadoop is a software-only setup.
So there are hardware details in each of these patents, where particular pieces
of hardware are connected to particular other ones, and you ought to be able to
go non-infringement on any of them.
But we shouldn't let this distract from the prior art search. If patents
asserted against open-source projects are routinely destroyed via prior art,
perhaps it will help keep the trolls at bay. And it is in this party's interest
to file the prior art for invalidation at least as a plan B.
[ Reply to This | Parent | # ]
|
|
Authored by: Anonymous on Monday, April 08 2013 @ 03:11 PM EDT |
IANAL, but unless HDFS is prior art to the issuance of the
patent, I believe arguing that the patent describes it
doesn't help any. At best, you're claiming someone else
violates the patent as well.
Arguing that others had similar conduct that (under the
infringement theory presented) infringed the same patent in
a similar way, and the patentee knew about it, and
deliberately decided not to pursue those claims, you might
have an argument for either estoppel or waiver (the
patentee's conduct lead you to believe they wouldn't enforce
the patent in this way, you relied on that to your
detriment), but AFAIK it doesn't help for invalidity. [ Reply to This | Parent | # ]
|
|
|
|
|