[Approval Request] Kallisys Reflexive License (Draft 2)

[Paul Guyot]

Hi all,

I would like to get the Kallisys Reflexive License (KRL) approved by 
the Open Source Initiative.
This submission (for the Draft 2) replaces the previous submission 
for the Draft 1.

Changes for Draft 2:
* Abe Kornelis cleverly remarked that the idea that one could modify 
the covered work wasn't explicit. This has been fixed by defining use 
(which includes making modifications) and rephrasing section 4 
accordingly.
* Some more or less obscure passages (including those remarked by 
David Turner) were rephrased.
* Victor Rehorst submitted several corrections for the American 
English version.

>1.	Put the license on a web page in HTML form. We will convert 
>it into the same style as the existing approved licenses. You can 
>help us by publishing it in that style yourself to save us the 
>conversion step.

The KRL can be found here:
http://www.kallisys.org/reflexive/reflexive.html (2 columns)
http://www.kallisys.org/reflexive/reflexive-1.html (1 column)

The 1 column presentation is close to the approved licenses format.

>2.	Tell us which existing OSI-approved license is most similar 
>to your license. Explain why that license will not suffice for your 
>needs. If your proposed license is derived from a license we have 
>already approved, describe exactly what you have changed. This 
>document is not part of the license; it is solely to help the board 
>understand and review your license.

The KRL isn't derived from any license. However, it has a 
contaminating clause, like the GPL.
The GPL creates some problems on some OSes such as NewtonOS (e.g. 
tools to create software cannot be included in the software source). 
The KRL fixes this problem with a strict definition of notions such 
as Development System or Host System. Also, it seems that the GPL 
isn't compatible with a lot of licenses such as the Apache license. 
The KRL, with its Annex A, solves the problem.

Additionally, the KRL uses the contaminating feature to add a 
positive side-effect. Code released under the KRL must complain a 
certain number of objective criterias which I believe will guarantee 
a certain quality of software (of source code first and, as a 
consequence, of the software as a whole).
Finally, the KRL is based on a conception of open source software 
(which is a superset of the OSI definition, hence my approval 
request) which is different from any other approved license. For 
example, the development software should either be free/included with 
the system or there should be at least two concurrent commercial 
development systems.

>3.	Explain how software distributed under your license can be 
>used in conjunction with software distributed under other open 
>source licenses. Which license do you think will take precedence for 
>derivative or combined works? Is there any software license that is 
>entirely incompatible with your proposed license?.

The KRL, like any contaminating license, will take precedence over 
any license it is compatible with. To improve compatibility, the KRL 
includes an Annex A to be completed with compatible clauses required 
from other licenses (e.g. Apache's license conditions). Cf 4.3 & 
Annex A.
Please note that compatibility with a license doesn't mean 
compatibility with software released under this license, i.e. you 
cannot re-license to KRL without ensuring that the criterias are 
respected.
I think that the KRL is compatible and incompatible with the approved 
licenses mentioned there:
http://www.kallisys.org/reflexive/compatibilite/

>4.	Send your proposed license by email to 
>license-approval at opensource.org. Indicate in the email whether you 
>want the license posted to the license-discuss list with your 
>identification or anonymously. (We are willing to consider licenses 
>that the author doesn't want posted at all, but since community 
>review is an important part of the approval process, we will have to 
>circulate such licenses privately to individual reviewers: because 
>of this, licenses not posted to license-discuss at all may take 
>longer to approve, and are likely to require more interaction with 
>you.) You are invited to follow discussion of the licenses by 
>subscribing to license-discuss-subscribe at opensource.org. This 
>mailing-list is archived here.

I'm posting myself to license-discuss (note to license-discuss 
members: no need to CC me, I subscribed to the list).

>8.	Once we are assured that the license conforms to the Open 
>Source Definition and has received thorough discussion on 
>license-discuss or by other reviewers, and there are no remaining 
>issues that we judge significant, we will notify you that the 
>license has been approved, copy it to our website, and add it to the 
>list below.

I'll then change "Projet 2" and "Draft 2" to "Version 1.0".

Paul
-- 
Home page: http://www.kallisys.com/
Newton-powered WebServer: http://newt.dyndns.org:8080/
--
license-discuss archive is at http://crynwr.com/cgi-bin/ezmlm-cgi?3