Patent nonsense in the data warehouse DBMS market
There are two recent patent lawsuits in the data warehouse DBMS market. In one, Sybase is suing Vertica. In another, an individual named Cary Jardin (techie founder of XPrime, a sort of predecessor company to ParAccel) is suing DATAllegro. Naturally, there’s press coverage of the DATAllegro case, due in part to its surely non-coincidental timing right after the Microsoft acquisition was announced and in part to a vigorous PR campaign around it. And the Sybase case so excited a troll who calls himself Bill Walters that he posted identical references to it on about 12 different threads in this blog, as well as to a variety of Vertica-related articles in the online trade press. But I think it’s very unlikely that any of these cases turn out to much matter.
I’ve been a software industry analyst since 1981, covering DBMS throughout that period. Perhaps I’m getting senile, but I can recall only 1 software patent case in those 27 years that wound up having any material effect on enterprise software users — Marcam/Ross Systems, about process manufacturing ERP software. Business Objects/Cognos doesn’t count; Business Objects enjoyed the cash, but Cognos easily afforded the payments. Lotus/Borland about Quattro Pro’s compatibility with 1-2-3 doesn’t count either; the issue there wasn’t patents.
So I haven’t studied the patents and cases in question seriously myself, since I think it’s very unlikely the trouble to do so will prove worthwhile. That said, DATAllegro CEO Stuart Frost told me today that Cory Jardin’s patent was something he thought was invalid before it ever was submitted, because of prior art, and that the application was formally submitted without reference to prior art. (Bad idea if you want the patent to hold up to reexamination when you try to enforce it later.) He further claims that DATAllegro doesn’t come close to infringing anyway. Needless to say, he also rejects the allegation that his own patents or applications in the area are for similar claims to those Jardin asserts.
As for the Sybase case — I don’t know which specific patents are alleged to be infringed. That said, US Patents 5794229 and 5918225 (issued in 1998 and 1999 respectively), both read in their first claim as if they pretty much patent the idea of columnar SQL DBMS with compressed data, especially dictionary/token compression. The “SQL” part is important, because otherwise they’d be blown out immediately by all sorts of prior art, e.g. from Essbase and Verity, or even Excel. Still, given the prevalence of ODBC access even in the 1990s, I’m hard-pressed to imagine how those patents could have any significant validity.
Also, the Sybase case was filed in Texas, which seems to be an oddly irrelevant venue. I wonder what their motive, or even excuse, for that choice was.
Related links
- Eric Lai’s article on the DATAllegro suit, with further basic links
- Stuart Frost’s short blog post responding to the suit
- What appears to be docket information for the Sybase/Vertica suit
Comments
7 Responses to “Patent nonsense in the data warehouse DBMS market”
Leave a Reply
Patent cases are often filed in (East) Texas, because its judges are known to rule in favour of patent owners.
Thanks. I was guessing it was something like that. 🙂
I think Stuart is barking up the wrong tree with prior art. If he knew prior art was missing from the submission “before it was submitted” as he claims then he was free to submit his own prior art. Anyone can submit prior art against a patent claim. He could even do so anonymously.
Furthermore, under laws of disclosure, if the patent attorney who submitted the claim knew of any prior art they were legally obliged to disclose it. If there was any blantant prior art out there at the time they would have been foolish to ignore or hide it.
If Stuart thought the patent submission was invalid you’ve got to wonder why two years later he submitted his own patent with much the same technology – “Ultra-shared-nothing parallel database”.
This case has a smoking gun, if Microsoft cannot invalidate the patent they are gone. Injunction. However it’s going to take several years to get to that point and they may settle before then.
Vincent,
You are, of course, correct in stating that inventors have an obligation to disclose prior art to the patent office. However, it’s very difficult to prove that someone was aware of a particular set of documents (unless you can prove a clear link via, say, an advisory relationship between the inventor and one of the gurus in the field).
In the patent in question, all of the prior art listed was raised by the patent examiner (shown by having * next to the reference).
As for whether I should have submitted prior art to the patent office, that would be a very unusual step. However, I’d agree that hindsight is a wonderful thing. Also, there’s no restriction on anyone putting forward prior art even after a patent is granted and asking for a re-examination.
Just because I didn’t do it earlier doesn’t make it wrong now. Nor does it say anything about the validity or strength of my position.
Stuart Frost
CEO, DATAllegro
Thanks for responding Stuart. I think we would be better off with less software patents and I hope this doesn’t get in the way of Microsoft rolling out some useful database acceleration.
[…] in August, 2008, I pooh-poohed Sybase’s patent lawsuit against Vertica. Filed in the notoriously patent-holder-friendly East Texas courts, the suit basically claimed […]
[…] manufacturing, and the co-products/byproducts it entailed. This led to the one significant patent suit outcome in enterprise software history, in which Marcam really did chase Ross Systems’ product off the […]