Re: What's wrong with the AGPL?

On 14.06.24 01:48, Matt Wilson wrote:
> First, I think it's important to recognize that when I post things
> on Twitter from my personal account, I'm doing so in an individual
> capacity.

It wasn't just you. In the end I mostly got "everyone believes it is not 
working so it is not working." Social proof does not cut it for me here.

> Indeed, the lack of a "viral" effect for the copyleft obligations of
> AGPLv3 licensed programs to "infect" obviously independent creative works
> of software is not only a common position of many who offer software

To an engineer writing client code to a copyleft-licensed library, their 
client code is probably "obviously independent". But not to a strong 
copyleft license of that library. The Linux kernel has long upheld this. 
Client code becomes covered work.

> The thing that
> is different about AGPLv3 is an additional obligation to provide the
> complete corresponding source code to modified versions of the program
> that are made available over a network to the user(s) of the software.

If by "made available" you mean "conveyed" (as used in the license 
definition) and all of this corresponds to using the functions of 
running object code on someone else's servers, this would be cloud 
copyleft (in my definition).

> Note however that it is NOT required that applications using mongo
> be published. The copyleft applies only to the mongod and mongos
> database programs. This is why Mongo DB drivers are all licensed
> under an Apache license. You application, even though it talks to
> the database, is a separate program and “work”.
> (the typos in the original have been preserved)
>
> My interpretation of their historical position is this: the APLv3 applies
> to the "core database source code", and not to anything that is a separate
> "work". To make it abundantly clear that AGPLv3 obligations do not 
> propagate
> to other works, like applications that connect to the database, they
> published Apache v2 licensed drivers. My personal position is that this
> is not strictly necessary, but it's nice to have that clarity from the
> producer of a AGPLv3 licensed program.

Yes, different problem though. Just because the AGPL doesn't cut it for 
MongoDB's business model (it can't distinguish between 
users-who-might-become-customers and the hyperscalers) doesn't mean the 
license is broken. All single-vendor open source cloud infrastructure 
COMPONENT providers have this problem and need permissively license 
shims to not shy users-to-be-customers away. Very different from the 
original single-vendor open source APPLICATION providers who originally 
created this IP strategy and who didn't need permissively licensed 
decoupling layers.

> In my personal view, the narrative that AGPLv3 is defective is a 
> flawed one.
> AGPLv3 is apparantly functional for its purpose: to give clear 
> permission to
> offer FOSS as a service (one among many useful purposes), as long as you
> fulfill the obligations (importantly, to provide the complete 
> corresponding
> source for a modified version to those who can access the program over
> the network).

Interesting. My point resp. my original assumption: The AGPL is just fine.

We only seem to differ in what is covered code. Which means I'm more 
confused now ;-)

My current state of thinking. The AGPL works as intended, however, 
because it does not allow the vendors of the currently dominant 
commercial open source paradigm to achieve their business goal, it has 
been dismissed off-hand. The IP strategy of the vendors should have been 
dismissed, not the license.

There are some negative consequences of this confusion: Application 
vendors who could open source choose to go directly to source-available. 
Example on my mind: GitButler; could have been plain AGPL, is SSPL.

Enjoyable as always. Hope there will be a resolution though.

Dirk

-- 
Confused about open source?
Get clarity through https://bayave.com/training
--
Website: https://dirkriehle.com - Twitter: @dirkriehle
Ph (DE): +49-157-8153-4150 - Ph (US): +1-650-450-8550


_______________________________________________
The opinions expressed in this email are those of the sender and not necessarily those of the Open Source Initiative. Official statements by the Open Source Initiative will be sent from an opensource.org email address.

License-discuss mailing list
[email protected]
http://lists.opensource.org/mailman/listinfo/license-discuss_lists.opensource.org