Weekly Edition
Return to the Press pageSponsored link
Vyatta –
Linux & Open Source
Alternative to Cisco –
Advanced Routing,
Firewall, VPN, QoS..
Free Download ->
| |||||||||||
Two patent decisions from the U.S. Supreme Court
The U.S. Supreme Court has issued two decisions, both of which weaken the current patent regime somewhat. The San Jose Mercury News covers the ruling in ATT v. Microsoft, which decided that Microsoft is not responsible for violations of U.S. patents which happen elsewhere in the world. "'The presumption that United States law governs domestically but does not rule the world applies with particular force in patent law,' Justice Ruth Bader Ginsburg wrote in the majority opinion."
This Bloomberg article covers the second ruling, which states that simply combining two inventions in a trivial way does not create a new, patentable invention. "'Granting patent protection to advances that would occur in the ordinary course without real innovation retards progress,' Justice Anthony Kennedy wrote for the court." (Log in to post comments)
Two patent decisions from the U.S. Supreme Court Posted Apr 30, 2007 16:57 UTC (Mon) by tetromino (subscriber, #33846) [Link] Groklaw has links to the pdfs of the actual rulings: http://www.groklaw.net/article.php?story=20070430121005424
The second decision is extremely important. Before this ruling, in the US, you could combine two existing inventions into a patent as long as there was no written "teaching, suggestion or motivation" to suggest the two inventions could be combined. This obviously gave a lot of freedom to patent trolls. But now thanks to the new Supreme Court, combining two inventions into a patent is permitted only when such a combination would not occur "in the ordinary course". So (at least some) obvious things are no longer patentable!
Two patent decisions from the U.S. Supreme Court Posted Apr 30, 2007 17:02 UTC (Mon) by JoeBuck (subscriber, #2330) [Link] A very large number of software patents would seem to fall into this category; they are variants of the form "do X over a computer network", where the patent claims not only the particular way that the submitter did X over a computer network, but the very concept of doing X over a computer network.
Two patent decisions from the U.S. Supreme Court Posted Apr 30, 2007 19:04 UTC (Mon) by ms (subscriber, #41272) [Link] Here are some bits. Here are some more bits. If I combine them, they do both the things that they individually did before. You mean, I *can't* patent this any more? The mind boggles!
Two patent decisions from the U.S. Supreme Court Posted Apr 30, 2007 20:57 UTC (Mon) by drag (subscriber, #31333) [Link] There are two things to keep in mind here... (also keeping in mind IANAL)
Basing a new invention on a old one is still perfectly valid under the law. It's fundamental to how the patent system operates. Take for example, the Microsoft FAT patent. When it expires Microsoft can release a new FAT patent by adding a new idea to make the old invention a new one. Get a fat++ patent or whatever.
The second thing to keep in mind is that what a patents _says_ it covers and what it actually _does_ cover can be two entirely different things. Most people read the abstract and assume that is what the patent covers, which is not realy true. The abstract is just a overview.. it gives you a idea about what the patent is about.
To find out what the patent actually covers you have to examine the 'claims' section.
Now it looks like this:
2. ----
3. ----
etc etc.
So each claim is what the patent actually covers. The a/b/c/etc part is stipulations (I am pretty sure that isn't the correct word for it) on that claim.
So if you have a product which is not covered by claim 1, but is covered by claim 2.. then you violate the patent. So each claim is independant and you can fall underneath any one of them.
But in order to violate the claim you have to still fufill all the stipulations.
So if I have a product which looks like it violates only claim '1', but only matches what is found in 1.a and 1.c, but not 1.b.. then I do not violate the patent. You have to match all the bullet points in a paticular claim in order to violate the patent. If any one of them does not apply to you then your not covered.
So in summary you have to violate at least _one_ of the claims AND your product has to fuffill all the bullet points listed under that claim in order to violate the patent.
The abstract of the patent is just worthless for understanding. It's just a 'execuative summary' type thing. So patents are usually quite a bit more focused then they seem at first blush.
Of course software patents are still a travesty.
Two patent decisions from the U.S. Supreme Court Posted Apr 30, 2007 21:32 UTC (Mon) by Los__D (subscriber, #15263) [Link] Basing a new invention on a old one is still perfectly valid under the law. It's fundamental to how the patent system operates. Take for example, the Microsoft FAT patent. When it expires Microsoft can release a new FAT patent by adding a new idea to make the old invention a new one. Get a fat++ patent or whatever.But then what they patent would only be the new parts, right? Mentioning that it's based on the old one would then just be a part of the explanation of how the new patent works, and legally tell about the earlier patent as prior art.(AFAICS, IANAL, etc.,etc.)... - But then again, I have no idea of what I'm talking about when it comes to law... :) Dennis
Two patent decisions from the U.S. Supreme Court Posted May 1, 2007 0:43 UTC (Tue) by drag (subscriber, #31333) [Link] yep. The new patent would only cover the new stuff, the old invention that it's based off of would be, when it expires, public domain.
Microsoft/AT&T patent decision from the U.S. Supreme Court Posted May 3, 2007 21:41 UTC (Thu) by giraffedata (subscriber, #1954) [Link] decided that Microsoft is not responsible for violations of U.S. patents which happen elsewhere in the world. That was not the holding at all. It is already well established that violations of US patent law can't happen outside the US. This is a case where the alleged violation took place in the US -- Microsoft prepared and shipped, in the US, a version of Windows that uses an AT&T speech compression invention. This isn't a violation of ordinary patent law, because the law views software as just information, not a device, so preparing bits doesn't violate a patent. To infringe the patent in this ordinary way, Microsoft would have to build a computer that runs that Windows. But there's a special provision in US patent law adding to the protection of a patent the shipping from the US of a major component of an infringing device. So the question is whether what Microsoft shipped was a component of the foreign computers that ultimately got built with that version of Windows. The court said information itself can't be a component. The computers' disk drives and CDROMs containing Windows are a component, but Microsoft didn't ship those; Microsoft shipped a master and the foreign manufacturers copied the information onto CDs and disk drives. The court likened shipping of a Windows master to shipping the blueprints for a machine. Blueprints are understood not to be a component of the machines you build with them. The view that software cannot be a device seems primitive to me, but until the law gets with the times, this holding should be good for software developers, regardless of the international aspect. It says if you keep your work sufficiently soft (never send hardcopy; maintain distance between you and its ultimate installation) you can avoid patent liability.
Microsoft/AT&T patent decision from the U.S. Supreme Court Posted May 5, 2007 12:26 UTC (Sat) by ddaa (subscriber, #5338) [Link] So, according to you, the ruling establishes that:
That seems strangely contrived. It appears to me there is some subtext here:
That, or the court is truly being a vehicle of the of the law, and I am completely imagining this subtext. DISCLAIMER: Being a continental European, I am probably unable of understanding the anglo-american law system.
Microsoft/AT&T patent decision from the U.S. Supreme Court Posted May 5, 2007 21:23 UTC (Sat) by giraffedata (subscriber, #1954) [Link]
Well, the case doesn't establish anything like that, but maybe you meant my interpretation of the law surrounding the case. In that case, this is basically right. But:
As I said, none of the above is new in this case; it's all from previous cases. What's new in this case is that shipping of a master copy of software, copies of which are to be built into computers, is not the same thing as shipping a component of a computer.
That's no loophole, it's a complete contradiction of the assertion that you can patent software. First of all, there's the terminology issue from above -- you can't patent software (or hardware). You can patent only an invention. Second, while you can with a patent stop someone from building an apparatus, you cannot stop someone from writing a program. No twisting, no loophole. It's always been that way. What is relatively new in the US and thought to be screwy by many people, is that you can with a patent stop someone from building an apparatus that uses software to implement an invention. What LWN calls a "software patent" is a patent on an invention that is realized by loading certain software into a computer. As opposed to, say, soldering certain wires into it. There used to be a loophole (not related to software -- it was before software was a big deal) in patent law in that the same computer that would be illegal to build in the US and ship to Mexico, you could build in two individually useless parts in the US, ship those parts separately to Mexico, and assemble them there. Congress attempted to close that loophole, but left open to interpretation whether shipping a master CD to Mexico, to be copied into a computer there, is covered.
|
|||||||||||