Hi,
I'm packaging golang-github-hashicorp-lru-arc-2 [0] as a dependency of symfony-cli [1]. The code is licensed under the MPL-2.0, but the documentation of the library mentions the IBM patent of ARC [2].
I just noticed it and suspect it's not allowed in Fedora, but I'd like confirmation.
Thanks in advance.
Kind regards, Arthur Bols fas/irc: principis
[0]: https://github.com/hashicorp/golang-lru/tree/v2.0.7/arc [1]: https://github.com/symfony-cli/symfony-cli [2]: https://en.wikipedia.org/wiki/Adaptive_replacement_cache
Hi Arthur:
On Fri, Jan 19, 2024 at 10:41 PM Arthur Bols arthur@bols.dev wrote:
I'm packaging golang-github-hashicorp-lru-arc-2 [0] as a dependency of symfony-cli [1]. The code is licensed under the MPL-2.0, but the documentation of the library mentions the IBM patent of ARC [2]. [...] [0]: https://github.com/hashicorp/golang-lru/tree/v2.0.7/arc [1]: https://github.com/symfony-cli/symfony-cli [2]: https://en.wikipedia.org/wiki/Adaptive_replacement_cache
FWIW, per the Google page [3] for the patent [4] this IBM patent is set to expire on 2024-02-02 ... so in just a few days! ;)
With this said, I wonder what is the interplay between the patent-related terms of the MPL in Hashicorp's code and the patent from IBM who is not Hashicorp. And what is the impact on the MPL licensing? A well known example is FFmpeg with L/GPL-licensed but patented media codecs. [5]
[3] https://image-ppubs.uspto.gov/dirsearch-public/print/downloadPdf/6996676 [4] https://patents.google.com/patent/US6996676B2/en [5] https://ffmpeg.org/legal.html -- Cordially Philippe Ombredanne
+1 650 799 0949 | pombredanne@nexB.com AboutCode - Open source for open source - https://www.aboutcode.org ScanCode - scan your code https://github.com/nexB/scancode-toolkit https://github.com/nexB/scancode.io DejaCode - What's in your code?! - http://www.dejacode.com
Thank you Philippe.
On 20/01/2024 10:51, Philippe Ombredanne wrote:
Hi Arthur:
On Fri, Jan 19, 2024 at 10:41 PM Arthur Bolsarthur@bols.dev wrote:
I'm packaging golang-github-hashicorp-lru-arc-2 [0] as a dependency of symfony-cli [1]. The code is licensed under the MPL-2.0, but the documentation of the library mentions the IBM patent of ARC [2]. [...] [0]:https://github.com/hashicorp/golang-lru/tree/v2.0.7/arc [1]:https://github.com/symfony-cli/symfony-cli [2]:https://en.wikipedia.org/wiki/Adaptive_replacement_cache
FWIW, per the Google page [3] for the patent [4] this IBM patent is set to expire on 2024-02-02 ... so in just a few days! ;) ... [3]https://image-ppubs.uspto.gov/dirsearch-public/print/downloadPdf/6996676 [4]https://patents.google.com/patent/US6996676B2/en
What a coincidence, that's great to hear! I assume that from February 3 the package can be included in Fedora?
With this said, I wonder what is the interplay between the patent-related terms of the MPL in Hashicorp's code and the patent from IBM who is not Hashicorp. And what is the impact on the MPL licensing? A well known example is FFmpeg with L/GPL-licensed but patented media codecs. [5] ... [5]https://ffmpeg.org/legal.html -- Cordially Philippe Ombredanne
Could indeed be interesting, but I'd rather be safe than sorry! ;)
Arthur
Arthur:
On Sat, Jan 20, 2024 at 12:32 PM Arthur Bols arthur@bols.dev wrote:
On 20/01/2024 10:51, Philippe Ombredanne wrote:
On Fri, Jan 19, 2024 at 10:41 PM Arthur Bols arthur@bols.dev wrote:
I'm packaging golang-github-hashicorp-lru-arc-2 [0] as a dependency of symfony-cli [1]. The code is licensed under the MPL-2.0, but the documentation of the library mentions the IBM patent of ARC [2]. [...] [0]: https://github.com/hashicorp/golang-lru/tree/v2.0.7/arc
FWIW, per the Google page [3] for the patent [4] this IBM patent is set to expire on 2024-02-02 ... so in just a few days! ;) ... [3] https://image-ppubs.uspto.gov/dirsearch-public/print/downloadPdf/6996676 [4] https://patents.google.com/patent/US6996676B2/en
What a coincidence, that's great to hear! I assume that from February 3 the package can be included in Fedora?
That's my understanding, but these patent documents are all going through a round of obfuscation, two rounds of encryption, and three rounds of compression to make them hard to decipher for the law school dropouts without a decoder ring! ;)
On another note, it could be that this patent is part of some patent non-assert promise of IBM or some patent pool, possibly making the patent issue moot. IBM or RedHat lawyers would know best.
-
Arthur Bols -
Philippe Ombredanne