www-jcp-open mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Geir Magnusson Jr." <g...@pobox.com>
Subject Re: Contributing to RI (SUN) and Apache...
Date Thu, 27 Mar 2008 15:18:14 GMT

On Mar 27, 2008, at 11:09 AM, Wade Chandler wrote:

> --- "Geir Magnusson Jr." <geir@pobox.com> wrote:
>> On Mar 27, 2008, at 10:35 AM, Wade Chandler wrote:
>>> --- "Geir Magnusson Jr." <geir@pobox.com> wrote:
>>>> I don't believe so.
>>>> As much as I hate/admire (Good Geir  / Evil Geir) what Sun is doing
>>>> here ("everyone works for us for free!"), it should have no  
>>>> effect on
>>>> anything you do at the ASF.
>>>> You might want to run this past legal-discuss to be doubly sure,
>>>> but I
>>>> don't expect any problems.
>>> I don't see the "everyone works for us for free" part. All the SCA
>>> does is assign joint copyright
>>> through agreement and signature versus relying only on a license for
>>> contribution and covers
>>> their butt legally to use the sources someone else volunteers to add
>>> to one of their projects
>>> without being bogged down by future claims.
>> They also take that source and re-license under commercial terms to
>> others, because they can (and no one else can).  It's a brilliant
>> strategy :)  Got the MySQL crew a cool $1B.
> Yes, but MySQL had commercial licensing the whole time ;-). It was  
> very expensive to include a
> copy of MySQL with a commercial product. I wrote them a long time  
> ago and asked for a project we
> were working on. The reply was they wanted us to pay roughly $400.00- 
> U.S. over a few years ago
> for each copy we distributed in a piece of commercial software which  
> didn't cost 50 itself, so we
> went with Sybase SQL Anywhere.

Sun has commercial licensing for the Java codebase.  Have had it the  
whole time.

> Anyways, others have actually taken Sun projects and sold them  
> commercially. There are groups
> selling OpenOffice as a differently named packaged software product  
> along with other softwares
> and their value adds. Others do this with NetBeans as well. With  
> NetBeans one can take the full
> IDE if they want to, call it something else, add a couple other  
> modules to it, and sell their own
> branded IDE for money with added terms and an overall different  
> license with the core obviously
> being covered by the GPL w/ClassPath Extensions and CDDL.

Really?  try shipping a derivative work under a closed source license.

> You may be referring to something else though, but the Sun projects  
> code released under one
> license can't be changed on the given release, so I don't really see  
> a real difference in others
> selling and licensing a product commercially or not. It seems much  
> like what Linux distros do.
> Someone has to pay the devs and they all have to eat :-D.
>>> The sad reality is this should be the case when one commits
>>> something, but with the way legal
>>> systems work sometimes not based on laws but opinions or feelings it
>>> can certainly feel necessary
>>> to corporations to go the extra mile to make sure everyone is at an
>>> understanding. Anyways, the
>>> SCA is simply a signed statement of what it really means to put your
>>> code in someone else's
>>> project.
>> No, that's not true.  At the ASF, we don't get "joint copyright"
>> assignment when code is placed in our projects.  The copyright  
>> remains
>> distributed among those that contributed it, and the project  
>> asserts a
>> collective copyright on the entire work.
> This is what the joint copyright is.

No it isn't.

> Those who contribute their own code have a copyright on
> their contributions, and the project has their copyright on the  
> collective work.  Seems the SCA
> asserts the same thing...just someone has to sign it.

Nope.  They are not the same thing.  The collective copyright is the  
copyright on the *set* of elements, but the authors still hold the  
copyright on the elements themselves.  Think about it as an anthology  
of poetry or something.  The editor can assert copyright on the book,  
but each poet still maintains individual copyrights on the poems  

Sun can take the OpenJDK codebase and remove any obligations imposed  
by the GPL and give you a license to use that code in any way you  
chose (or in reality, as they allow you...)

No one can take an Apache project and do that.  Granted, the  
obligations placed upon the licensee by the Apache License are far  
lighter than that of the GPL so that it really doesn't matter, but  
still - there is a clear difference, and that is why I contend that if  
you want a level playing field, some sort of agreement between the  
entity that is being granted joint copyright and the community is  
necessary to keep all actors equal.

>> I agree that having a single copyright holder is probably useful for
>> dealing w/ license transgressions (to give a single entity standing  
>> to
>> do something about it) - I think that some kind of "no special
>> benefit" covenant between the copyright holder and the community  
>> would
>> be better to have a level playing field.
> Sure. Hard to say though. At that level someone could contribute a  
> couple lines of code then be
> able to do what they want with the rest. However, with open-source  
> projects which are more
> liberal than the GPL one can do many things with the project and  
> commercially turn a profit. IBM,
> Oracle, Sun, etc all do this with the Apache projects, and, at least  
> for IBM and Sun...I don't
> know about Oracle, they contribute to different ones.

They all use our stuff.  And that's ok.

> For instance, we use Oracle AS (Web Server), and it is Oracle OC4J  
> and Apache HTTPD with some add
> ins. They then have a bunch of other open-source libraries in use.  
> Reading the licensing for the
> product one can't tell anything about the underlying contributions  
> unless they dig down and see
> the licenses in the individual pieces directories etc, and you  
> certainly pay for the
> product...$$$, but others can do the same thing with the base open- 
> source.

Yep.  :)


> Wade

View raw message