Forgot your password?
typodupeerror
Patents Businesses Google The Internet

Northeastern University Sues Google Over Patent 159

Posted by ScuttleMonkey
from the who-isn't-suing-google-these-days dept.
kihbord writes to mention that Boston's Northeastern University and Waltham, Mass. based company Jarg have brought suit against Google for apparently infringing on a distributed database system developed by Kenneth Baclawski. "The patent describes a distributed database system that breaks search queries into fragments and distributes them to multiple computers in a network to get faster results. The patent was assigned to Northeastern University, which licensed it exclusively to Jarg, according to the lawsuit, filed last Tuesday with the U.S. District Court for the Eastern District of Texas."
This discussion has been archived. No new comments can be posted.

Northeastern University Sues Google Over Patent

Comments Filter:
  • by MontyApollo (849862) on Monday November 12, 2007 @05:06PM (#21328841)
    It looks like Jarg is a company with actual products, so I don't know if you can really call them a patent troll (unless you are Google fanboy).
  • by addikt10 (461932) on Monday November 12, 2007 @05:18PM (#21328995)
    If the courts of Marshall Texas are so popular for patent trolls, and the way you justify jurisdiction is to claim that "services are available there", blacklist every Marshall Texas ISP's IP range in that area.

    If one of those IP addresses tries to access your service, put up a nice, static HTML page declaring that it due to the local court's ignorance of patent issues and the resulting popularity of those courts for patent litigation, it isn't a good business decision to provide services to that area.

    Have a nice day.

  • by MozeeToby (1163751) on Monday November 12, 2007 @05:31PM (#21329139)
    Yeah, but on the other hand, sitting on a patent for years and years while you know its being violated by numerous other companies IS patent trolling.
  • Are you kidding? (Score:1, Interesting)

    by Anonymous Coward on Monday November 12, 2007 @05:33PM (#21329175)
    That they got a patent for something so obvious is absurd. One could call upon the millions of beowulf clusters out there that split up a computing job among (tens/hundreds/thousands) of computers to speed the work, or alternatively, any load balancing application (be it hardware or software) for quicker response to web queries. Even Slashdot could be prior art! The US patent office needs to be pulled over and given a sobriety test!
  • by Nom du Keyboard (633989) on Monday November 12, 2007 @05:33PM (#21329185)
    I thought Google was highly secret in what they did and how they did it. The way I've read this, it seems like a pishing expedition of Google needing to prove that they don't violate the patent, and not that there is yet any proof that they do violate it.

    And then there's always the specter of Prior Art raising its unwanted head.

    Have these guys ever built such a database system themselves for sale?

  • by rhombic (140326) on Monday November 12, 2007 @05:33PM (#21329189)

    Universities are corporations, nothing more or less.

    Except Uni's are completely free to ignore everyone else's patents in the course of their research, have access to all scientific software at much much cheaper "academic" rates, and can pay grad students slave wages ($15,000 per year for a 3000 hour work week is well below the Federal minimum wage). So while they behave in many ways like corporations, they have a number of government-issued advantages in the competition. Who'd have thunk it, the government giving itself an advantage.

  • Prior Art? (Score:5, Interesting)

    by CustomDesigned (250089) on Monday November 12, 2007 @05:36PM (#21329237) Homepage Journal
    We had a system in our office in 1985 that distributed records for each table to N processors via a hash function, where N could be a large as you liked. Queries were sent to all nodes and run in parallel, and the results combined (since SQL is set based, this works perfectly). Queries on any size database could be made arbitrarily fast by adding more nodes. The only bottle neck was the band width to the control processor and any order by clauses, which was proportional to the result set size, not the database size.
  • by MontyApollo (849862) on Monday November 12, 2007 @05:40PM (#21329285)
    That's a silly argument. If you sue anybody, you want to do it where its most favorable for you.

    A patent troll is someone who never intends to develop a patent but just sits on patents with the only purpose to sue those who infringe. In this case, the professor did not sit on the patent; he licensed it to Jarg, which is a company with real products.
  • by EEPROMS (889169) on Monday November 12, 2007 @05:43PM (#21329337)
    In my postgrad years in the early 80's I worked for a now defunct computer company that often supplied minicomputers for Government departments. The patent is an exact copy of how we did information searches across a network.Honestly there is no other way of doing this type of search efficiently back then and I dare say the same method pre dates my early years within th computer industry. Think about it, user enters search criteria into stand alone PC/Terminal then it is passed onto a primary node minicomputer, if the information is not available on hand (cached common searches) the search engine software then quires all the other minicomputers computers attached, it's a no brainier. What Google has done is refine the search algorithms and used more generic hardware but the concept hasnt changed.
  • by The Empiricist (854346) on Monday November 12, 2007 @05:46PM (#21329371)

    The /. posts labeling Northeastern University as a patent troll or claiming that the patent should not have issued have been posted too quickly to be credible assessments of the morality of this suit and the worthiness of the patent. The current Wikipedia definition [wikipedia.org] is that a patent troll is "a person or company that enforces its patents against one or more alleged infringers in a manner considered unduly aggressive or opportunistic." If the patent was obtained through lawful and ethical means, is valid, and is infringed on by Google, then how is it "unduly aggressive or opportunistic" for Northeastern University to enforce the patent?

    Some argue that a patent troll is merely a person or company that seeks to enforce a patent but does not practice the patent. Maybe Northeastern University is not practicing the patent. Then again, the mission of most universities seems to be conducting research, not applying and commercializing research. Licensing research to companies that can and will apply and commercialize that research is one way that universities fund additional research. Maybe universities should have to give all of their research away for free. But currently they do not. And it seems unfair to fault Northeastern for exercising its rights while not pushing the scope of its mission.

    Given how quickly Northeastern was accused of being a patent troll, and given that there was no discussion about the proper role of universities or any real analysis of the worthiness of the patent (which was filed in 1994...almost 4 years before Google was founded), it seems likely that some people consider a patent troll to be any person who tries to enforce any patent rights.

    Maybe Northeastern is acting like a patent troll. And maybe their patent is worthless. But it takes more than a quick glance at the 20-page issued patent or the 6-page complaint against Google to come up with a reasonable assessment of these issues.

    Some analysis of the complaint would at least show that it doesn't look like Northeastern really knows the details of Google's search infrastructure:

    13. A part of its search engine services, Google uses one or more hashing algorithms.
    14. As part of its search engine services, Google returns search results responsive to user queries.
    ...
    16. Google maintains and operators clusters of networked computers to provide search engine services to users.
    ...
    26. Google has directly and/or indirectly infringed on one or more claims of the '593 patent, and Google is continuing such infringement by practicing or causing others to practice one or more of the inventions claimed in the '593 patent. For example, Google makes, uses, imports, sells and/or offers for sale search engine services and systems that infringe or that are used in ways that infringe one or more claims of the '593 patent in this district and elsewhere in the United States.

    Then again, Google's code is not open to the world. If it was, more detailed analysis would be possible. How can Northeastern try to get access to the code? By suing and demanding it as part of discovery. Does this make Northeastern a patent troll? Maybe. But the alternative (aside from discarding the patent system altogether, at least for software innovations) is a system that rewards patent infringers who keep their source code inaccessible to patent holders.

  • Re:Interesting Dates (Score:5, Interesting)

    by bstone (145356) on Monday November 12, 2007 @06:01PM (#21329565)
    Aside from obviousness, the idea has been around forever. nCube built a business around it in the early 80's.
  • by raddan (519638) on Monday November 12, 2007 @06:04PM (#21329587)
    The fact that a university is behind this patent lawsuit is the most disturbing part to me. I probably have no real basis for this opinion, since I only spent three semesters at the school (I transferred to another school), but I had the overwhelming feeling that Northeastern U's primary concern was: bring in the money. That's why I transferred to UMass Amherst. In my experience, Northeastern was overpriced, and filled with students and teachers who just couldn't give a damn about academics. UM Amherst's admission standards were certainly lower than NEU's, but I found that the students who wanted to be there were really motivated, and that the teaching, in general, was outstanding.

    That's not to say that NEU didn't/doesn't have some strong departments, nor do I mean to disparage anyone who is presently working their ass off there. I just didn't see it. This article strengthens my opinion of NEU as essentially "for profit" and not "for education".
  • Re:Former student (Score:3, Interesting)

    by Curze (1166985) on Monday November 12, 2007 @06:07PM (#21329637)
    I am a recent former student, and my experience is quite opposite of yours. Prof Baclawski, besides having the added benefit of speaking English natively, was an excellent teacher both for myself and the handful of others in my class that I spoke with regularly. He usually referenced "back in the day" stories, but in a humorous way that lead into the subject at hand, and I ended up learning more than most in that class. Also rediculously apparent was his apathy for monetary compensation, he just wanted to teach and research new things, hell the guy wore the same damn dirty coat for the 5 years I was there! The move on google is most likely a move by the NU corp, as they are bloodthirsty when it comes to money, just look at tution costs of NU.http://www.neu.edu/admissions/costs/tuition.html [neu.edu]
  • Re:Prior Art? (Score:4, Interesting)

    by yorugua (697900) on Monday November 12, 2007 @08:07PM (#21330633)
    Well, TFA mentions Dec 7 1997 as "the day". In 1996, I worked in a project which involves Oracle Parallel Server on a RS/6000 SP2 parallel system. OPS used "I/O shipping" to send to several nodes the I/O request for a given query, and used the VSD subsystem for doing that (VSD stands for Virtual Shared Disk. The disk were attached to several SP2 nodes, which did the actual I/O while the many OPS nodes all saw the complete database as directly connected to it). Back then, 1996, the other "standard" of doing the parallel-database-thing was called "function shipping", in which not the IO, but the query was split among the nodes that had the tables and/or data involved. Of course, in both cases, all the nodes were networked using the SP2 switch in order to have huge bandwidht and low latency. Sounds familiar...
  • by caffeinemessiah (918089) on Monday November 12, 2007 @08:23PM (#21330781) Journal

    The patent was for taking a single request, breaking it up into subrequests, then distributing the subrequests amongst multiple servers and then gluing the results back together.

    How is this different from any parallel divide-and-conquer algorithm?

Little known fact about Middle Earth: The Hobbits had a very sophisticated computer network! It was a Tolkien Ring...

Working...