I'm not sure it is, and I'd argue the situation is unclear. I'd be keen to see any examples though – it would be awesome if there was something to fall back on.
All I can find is evidence that it would be at best very risky to rely on an implied patent grant.
For example, it's considered unclear enough that the ClearBSD license was explicitly created to clarify that it doesn't offer patent grants - http://directory.fsf.org/wiki/License:ClearBSD
Even more damning:
In the absence of an explicit patent grant, but considering the word use in the license, can we assume that the BSD license impliedly grants enough of whatever patent rights the Univer- sity of California then owned that a licensee may use the soft- ware as it was originally distributed by the University? Most licensees under the BSD assume it does on the theory that oth- erwise the copyright license would be of no value. What good, they say, is software that can be copied but not used?
Such a conclusion is not based on the law of licenses. Indeed, a bare license of copyright need not include a bare license of patent at all. It is only if the BSD is viewed as a contract that we can introduce contract law principles such as reliance or reasonable expectations of the parties. If software is licensed under the BSD without forming a contract between licensor and licensee, the extent of any patent grant is at best ambiguous.
As to whether an implied grant of patent rights extends to versions of the software with modifications, that’s an even more complicated question. The BSD license is silent about a patent license for derivative works. So if a licensee improves the origi- nal Berkeley Software Distribution in a way that infringes a patent owned by the University of California, there is no easy way of knowing whether an implied BSD patent license includes a patent license for that improvement.
Since courts are likely to construe implied grants of license narrowly, a licensee should consider obtaining separately from the licensor an explicit grant of patent rights that might be needed for modified versions of BSD-licensed software.
""I'm not sure it is, and I'd argue the situation is unclear."
I wouldn't.
What you've quoted is Larry Rosen's view. Larry is a wonderful guy, but his views are pretty far outside the norm for open source lawyers.
To start "If software is licensed under the BSD without forming a contract between licensor and licensee, the extent of any patent grant is at best ambiguous."
This is now settled since he wrote this.
It is in fact a contract, that, if breached, leaves the licensor without a copyright license (causing both infringement of copyright and breach of contract).
So you don't have to worry about this.
The latter is a real issue, but one that most explicit grants don't solve either.
In particular, apache/et al have explicit grants do not cover modifications by others that suddenly encompass patents.
I'm not sure it is, and I'd argue the situation is unclear. I'd be keen to see any examples though – it would be awesome if there was something to fall back on.
All I can find is evidence that it would be at best very risky to rely on an implied patent grant.
For example, it's considered unclear enough that the ClearBSD license was explicitly created to clarify that it doesn't offer patent grants - http://directory.fsf.org/wiki/License:ClearBSD
Even more damning:
In the absence of an explicit patent grant, but considering the word use in the license, can we assume that the BSD license impliedly grants enough of whatever patent rights the Univer- sity of California then owned that a licensee may use the soft- ware as it was originally distributed by the University? Most licensees under the BSD assume it does on the theory that oth- erwise the copyright license would be of no value. What good, they say, is software that can be copied but not used?
Such a conclusion is not based on the law of licenses. Indeed, a bare license of copyright need not include a bare license of patent at all. It is only if the BSD is viewed as a contract that we can introduce contract law principles such as reliance or reasonable expectations of the parties. If software is licensed under the BSD without forming a contract between licensor and licensee, the extent of any patent grant is at best ambiguous.
As to whether an implied grant of patent rights extends to versions of the software with modifications, that’s an even more complicated question. The BSD license is silent about a patent license for derivative works. So if a licensee improves the origi- nal Berkeley Software Distribution in a way that infringes a patent owned by the University of California, there is no easy way of knowing whether an implied BSD patent license includes a patent license for that improvement.
Since courts are likely to construe implied grants of license narrowly, a licensee should consider obtaining separately from the licensor an explicit grant of patent rights that might be needed for modified versions of BSD-licensed software.