the SaaS-wraps-FLOSS loophole

[JanZerebecki]

From my reading

parity-public-license/LICENSE.md

Line 21 in 3353644

[Contribute](#contribute) software you develop, operate, or analyze with this software, including changes or additions to this software. When in doubt, [contribute](#contribute).

does not ask you to contribute software that you use to operate this software. Only software that is operated by this software.
I think a more unambiguous wording of equivalent meaning is:

Contribute additions to this software. When you use this software to develop, operate, or analyze other software contribute that software. When in doubt, contribute.

Thus it seems this license does not prevent the SaaS-wraps-FOSS loophole. I.e. a cloud service making a script to operate running this software more automatically as a service would not have to contribute the script to do this. As it is operating on not being operated on by Parity licensed software.

A similar example, but where to my understanding the intent is to not cover it: Using a keyboard with firmware to operate a Parity licensed software.

My understanding from our prior discussion in #60 is that the intention is to have the loophole fixed. Some wordings that fix this loophole risk over reaching so far that it becomes impractical to find a real world system where all software (firmware in disks, USB-C cables, network cards, switches, routers, etc.) is available under compatible licenses.

Should I have interpreted the wording differently, how?
Is there an easy way to fix this without resorting to line drawing like a system library exception?

[orlandohill]

I also find this ambiguous. My interpretation was that the 'with' in "develop, operate, or analyze with" was bidirectional, otherwise you could link with a Parity licensed library without the need to contribute. My assumption was that the requirement to contribute extended to wrapper scripts and software running over RPC protocols (avoiding the shim trick used to circumvent the GPL).

[nyabinary]

Is Parity even alive anymore in general?

[kemitchell]

@nyabinary very much so. I continue to maintain the website. These repos aren't going anywhere.

The language has felt more and more stable. No license is perfect, and a copyleft license like this has to drop some lines that pure-permissive forms don't. But I'm getting more comfortable with the level of generality it's at now.

I had taken a bit of as break earlier this year, when my father died and I had to settle all his affairs. I am still behind on some e-mail.

[nyabinary]

@nyabinary very much so. I continue to maintain the website. These repos aren't going anywhere.

The language has felt more and more stable. No license is perfect, and a copyleft license like this has to drop some lines that pure-permissive forms don't. But I'm getting more comfortable with the level of generality it's at now.

I had taken a bit of as break earlier this year, when my father died and I had to settle all his affairs. I am still behind on some e-mail.

Will we ever see another version of Parity that addresses these issues, btw?

[kemitchell]

It's been a while since #60 closed, and since this #75 opened.

The intent behind Parity was clearly not to insist that Parity-licensed software only be run on fully "open" systems or platforms. For reasons @JanZerebecki pointed out in #60, that could lead to absurdities, like arguably prohibiting use on computers using USB cables with closed firmware.

At the same time, I don't think I agree that this license wouldn't reach additional software written to "service-ize" a Parity-licensed application. Here again is the the Copyleft rule of version 7.0.0:

Contribute software you develop, operate, or analyze with this software, including changes or additions to this software. When in doubt, contribute.

If I'm building a web service version of an existing, Parity-licensed application, I would read that as software developed with the Parity-licensed software. I could also read it as an "addition" to the Parity-licensed software. And the final "if in doubt" interpretation rule would shore up those readings.

[nyabinary]

Any updates :P, also
https://www.theregister.com/2023/12/27/bruce_perens_post_open/
https://www.theregister.com/2024/04/30/bruce_perens_post_open_license/

I wonder if we can integrate some of these ideas into a new Parity ver? Bit off topic but yeah.