www-legal-discuss mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Robert Burrell Donkin <rdon...@apache.org>
Subject Re: LGPL software behind an isolation layer
Date Thu, 17 Sep 2009 19:36:41 GMT
Hash: SHA512

fpslpb@member.fsf.org wrote:
> Hi Ceki,
> Please see my responses inline below.
> On Wednesday, 16 September 2009, Ceki Gulcu wrote:
>> Given:
>> 1) A java library called logback licensed under LGPL 2.1.
>> 2) Logback implementing an interface called SLF4J, licensed under the
>> MIT license. Logback being implementation among many others of the
>> 3) Let FooCorp be a company and Frobber be some software developed by
>> FooCorp. Frobber codes against the SLF4J API without ever directly
>> referencing logback.
>> My Question: Should Frobber be considered as derivate work (of
>> logback)?
>> Your Answer: If (logback) is not required for Frobber to operate as
>> expected (i.e., Frobber will work without (logback), or with an
>> equivalent different implementation of the MIT SLF4J), then it seems
>> reasonable that this is sufficiently arms-length, and a case could
>> probably be made that providing the source code for the SLF4J
>> implementation would satisfy the requirement.
>> --
>> I hope my editing was faithful to the original and corresponds to your
>> intent.
>> Your answer is interesting because although not from an official
>> representative of the FSF, it originates from a person close to that
>> organization, and also because it answers my original question about
>> whether Frobber is derivative work or not.
>> If Frobber is not derivative work, then LGPL requirements do not apply
>> to Frobber (as long as it remains non-derivative).
> The first analysis of Frobber is to determine if it is a "derivative work", 
> which has a particular set of requirements; Frobber might or might not be a 
> derivative work of (logback) depending on how Frobber uses (logback). In your 
> example above, one reasonable interpretation is that Frobber is not a 
> derivative work. However I provide a different example below in which Frobber 
> might be considered a derivative work.
> If it is reasonable that Frobber is not a "derivative work", then further 
> analysis invokes Section 5 of the LGPL, which defines Frobber as a "work that 
> uses the library". This requires that Frobber + SLF4J + (logback) allow 
> sufficient reverse engineering to debug its function with a modified version 
> of (logback). A reasonable interpretation of that provision would require 
> FooCorp to ship Frobber with the source code to the implementation of SLF4J 
> and a way to link Frobber with a modified/debuggable version of SLF4J as well 
> as the modified (logback) library. This for example may imply separate .jar 
> files for SLF4J and (logback), and a way to invoke Frobber so that it can 
> include modified versions of those .jar files.
> Please note that in any analysis Frobber must either be a "derivative work" or 
> a "work that uses the Library", there is no third choice. 

this sounds incorrect to me: aggregation is also possible

- - robert
Version: GnuPG v2.0.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org


To unsubscribe, e-mail: legal-discuss-unsubscribe@apache.org
For additional commands, e-mail: legal-discuss-help@apache.org

View raw message