Tgppl, a new type of open-source license • Hacker News Board

Re: Tgppl, a new type of open-source license

#31

Post by tptacek » Sun, Sep 06, 2020, 11:59 PM UTC

iameli wrote: post #23

Haven't gone line-by-line on this one, but I like the new trend of "time limit" licenses. If I'm trying to start a business, something AGPL-ish might make sense in the short term to avoid bigger competitors copying everything we're doing. But that doesn't mean I want my great-grandchildren to be able to sue for violations 90 years from now. "AGPL for now, MIT later" seems like a good compromise.

You don't really need a new kind of license for that; you can just AGPL now, and release MIT later (assuming you're assigned the copyright for all the code).

Re: Tgppl, a new type of open-source license

#32

Post by rpdillon » Mon, Sep 07, 2020, 12:04 AM UTC

GranPC wrote: post #25

nordsieck wrote: post #24

This seems like a fundamentally bad idea for software. Simply put, living software contains a constant stream of patches, which "reset" the clock continuously. It would be one thing if this license were used on a book or some other work which is substantially finished. Once the book is published, the clock starts ticking and users eventually get to the MIT pot of gold at the end of the timer. With software, the only…

I haven't read the license fully, but wouldn't that just mean you would be able to use releases that are older than the time period specified?

That's my reading as well. It's actually kind of funny to me: the term of copyright has grown so long that folks are essentially looking for ways to reimplement reasonable copyright term length inside of the construct of copyright protection itself.

Re: Tgppl, a new type of open-source license

#33

Post by pas » Mon, Sep 07, 2020, 12:17 AM UTC

tptacek wrote: post #31

iameli wrote: post #23

Haven't gone line-by-line on this one, but I like the new trend of "time limit" licenses. If I'm trying to start a business, something AGPL-ish might make sense in the short term to avoid bigger competitors copying everything we're doing. But that doesn't mean I want my great-grandchildren to be able to sue for violations 90 years from now. "AGPL for now, MIT later" seems like a good compromise.

You don't really need a new kind of license for that; you can just AGPL now, and release MIT later (assuming you're assigned the copyright for all the code).

That doesn't guarantee downstream users that later they'll in fact have the code as MIT licensed. The author can change their minds, or just thinks it isn't worth the hassle. (Or some big corp buys them and they won't care.)

Re: Tgppl, a new type of open-source license

#34

Post by ocdtrekkie » Mon, Sep 07, 2020, 12:22 AM UTC

nsajko wrote: post #9

This (under the same name and version 1.0) has already been presented to FSF and OSI in the 2000s and 2010 on their mailing lists and other places, so I guess it didn't get approval. Since it thus isn't an open-source license, why present it as such in article title - starting out dishonest is probably not a good idea. If I'm wrong and it is approved by OSI, that info should probably be more prominent in the blog pos…

I don't feel like we're going to make any progress in this space while a single entity is considered the sole authorizer of something being open source or not.

Re: Tgppl, a new type of open-source license

#35

Post by giancarlostoro » Mon, Sep 07, 2020, 12:44 AM UTC

tptacek wrote: post #31

iameli wrote: post #23

Haven't gone line-by-line on this one, but I like the new trend of "time limit" licenses. If I'm trying to start a business, something AGPL-ish might make sense in the short term to avoid bigger competitors copying everything we're doing. But that doesn't mean I want my great-grandchildren to be able to sue for violations 90 years from now. "AGPL for now, MIT later" seems like a good compromise.

You don't really need a new kind of license for that; you can just AGPL now, and release MIT later (assuming you're assigned the copyright for all the code).

At that rate I'd prefer close sourced now, MIT later.

Re: Tgppl, a new type of open-source license

#36

Post by mncharity » Mon, Sep 07, 2020, 12:50 AM UTC

Animats wrote: post #29

nordsieck wrote: post #24

This seems like a fundamentally bad idea for software. Simply put, living software contains a constant stream of patches, which "reset" the clock continuously. It would be one thing if this license were used on a book or some other work which is substantially finished. Once the book is published, the clock starts ticking and users eventually get to the MIT pot of gold at the end of the timer. With software, the only…

Yes, that looks like a problem: ...to distribute or communicate copies of the Original Work and Derivative Works to the public, with the proviso that copies of Original Work or Derivative Works that You distribute or communicate shall be licensed under this Transitive Grace Period Public Licence no later than 12 months after You distributed or communicated said copies It looks like the effect is that the public versi…

> public versions [...] lag

That was the ghostscript licensing for many years. Recent versions available under a commercial license, and older versions as open source.

Nowadays would have to worry more about security and backporting fixes.

One challenge is it discourages contribution.

Re: Tgppl, a new type of open-source license

#37

Post by t0astbread » Mon, Sep 07, 2020, 1:31 AM UTC

pas wrote: post #33

tptacek wrote: post #31

Earlier quoted context omitted.

You don't really need a new kind of license for that; you can just AGPL now, and release MIT later (assuming you're assigned the copyright for all the code).

That doesn't guarantee downstream users that later they'll in fact have the code as MIT licensed. The author can change their minds, or just thinks it isn't worth the hassle. (Or some big corp buys them and they won't care.)

Isn't that always the case if the author is the sole copyright holder (or all copyright holders agree)? (Meaning that regardless of the license the copyright holder can always choose to not distribute new versions under that license anymore.)

Re: Tgppl, a new type of open-source license

#38

Post by gregwebs » Mon, Sep 07, 2020, 2:03 AM UTC

linkdd wrote: post #11

Reminds me of Sentry's model: https://blog.sentry.io/2019/11/06/relicensing-sentry

Their BSL has a 3 year period (whereas this license has a 1 year period). Sentry also admit it is by definition not an open source license since it has restrictions on particular uses.

TGGPL claims to be open source and probably is because everyone gets the same rights. But you are always potentially looking at code that is a year behind.

Re: Tgppl, a new type of open-source license

#39

Post by pas » Mon, Sep 07, 2020, 2:42 AM UTC

t0astbread wrote: post #37

pas wrote: post #33

Earlier quoted context omitted.

That doesn't guarantee downstream users that later they'll in fact have the code as MIT licensed. The author can change their minds, or just thinks it isn't worth the hassle. (Or some big corp buys them and they won't care.)

Isn't that always the case if the author is the sole copyright holder (or all copyright holders agree)? (Meaning that regardless of the license the copyright holder can always choose to not distribute new versions under that license anymore.)

But that's completely different than having a released version that we know will be available as MIT in X months.

It might mean you can start developing something based on the code and when it becomes MIT you can launch it and keep your additions to yourself. (Let's say in case of a server side thing.)

Re: Tgppl, a new type of open-source license

#40

Post by geofft » Mon, Sep 07, 2020, 2:59 AM UTC

nsajko wrote: post #15

geofft wrote: post #13

Earlier quoted context omitted.

There's a number of other reasons the OSI doesn't approve licenses, such as avoiding license proliferation https://opensource.org/proliferation . As a simple example, there are a huge number of variants of the 4-clause BSD licenses where the advertising clause lists more and more people besides the University of California; they're all open-source licenses under the Open Source Definition, but the OSI isn't intereste…

Still, the claim of "open source" or "libre" or "free software" doesn't hold water effectively if there is nobody to verify it (lawyers are probably needed, too). License proliferation is a separate issue. BTW, the OSI people on their mailing list weren't convinced that the license is conformant (during the grace period) in 2008, 2009, or 2013. Hypothetically there could have been a change of heart since, but I can't…

> BTW, the OSI people on their mailing list weren't convinced that the license is conformant (during the grace period) in 2008, 2009, or 2013.

Thanks, that's interesting. If you happen to have the links to the discussion it'd be worth figuring out what they thought was nonconformant (and whether it can be fixed).