Software grants Was: a few steps before approving a project

[Henri Yandell <fl...@gmail.com>]

On 9/1/05, Justin Erenkrantz <ju...@erenkrantz.com> wrote:
> --On September 1, 2005 12:28:13 PM -0400 James Carman
> <ja...@carmanconsulting.com> wrote:
> 
> > So, does that mean that Jakarta Commons Proxy will have to go through the
> > Incubator?  Right now, it's a commons sandbox project, so it's not
> > officially supported.  The code first lived in my "syringe" project I
> > created over at java.net.  It was all developed by me under the Apache
> > License 2.0.  Will Commons Proxy have to go through the Incubator to get
> > into commons proper?  Or, should it be in the incubator now?

Figuring out how the sandbox(es) and the incubator work together is
increasingly flagging on my attention list. Sam's list of rules is the
most concrete thing I've seen so far:

* If the SVN tree was always on ASF infrastructure, 
* the code was always under the Apache License, 
* and the committers were all ASF committers,
=> then no trip through the incubator is necessary.

<= If any ONE of these things are not true (example: code on CodeHaus
 created by ASF committers with Apache license), then incubator needs to
 be involved to ensure that there is a proper audit trail.

This is pretty painful, write a class outside the ASF that you own the
copyright to, then go through pain to get it in. Still, it's the
system and we need to try the system before we try to break it :)

> The Incubator (with its delegated responsibilities from the board)
> currently specifies that all external code imports that is going to be
> maintained at the ASF must have a 'software grant' on file:
> 
> <http://www.apache.org/licenses/software-grant.txt>

How is this submitted? Does it have to be fax/postal (ie very slow
feedback loop)? Do we send to Jim or the Incubator?

> Whether anything else is required is a decision for the Incubator PMC after
> knowing more about the situation.  In a lot of cases, that's all that is
> needed.

Didn't get anywhere with the Commons CSV stuff earlier, but I probably
didn't understand the right questions to ask. Currently I'm preparing
an Incubator proposal and hoping it'll just amount to a software grant
and then we can be out of the Incubator.

So another question here, is the de-facto new project template the
correct way to bring component/code into the sandbox/an existing
project?

> If any PMC is importing code to our repository without going through the
> Incubator, then they are in violation of ASF policy.  -- justin

The message is starting to get across and we'll get there. 

Hen

---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org

Re: Software grants Was: a few steps before approving a project

[Justin Erenkrantz <ju...@erenkrantz.com>]

--On September 1, 2005 10:50:53 PM -0400 Henri Yandell <fl...@gmail.com> 
wrote:

> Figuring out how the sandbox(es) and the incubator work together is
> increasingly flagging on my attention list. Sam's list of rules is the
> most concrete thing I've seen so far:
>
> * If the SVN tree was always on ASF infrastructure,
> * the code was always under the Apache License,
> * and the committers were all ASF committers,
> => then no trip through the incubator is necessary.
>
> <= If any ONE of these things are not true (example: code on CodeHaus
>  created by ASF committers with Apache license), then incubator needs to
>  be involved to ensure that there is a proper audit trail.

Yes.  (The constraint we use in httpd of being developed on-list reduces to 
the same as being in our SVN tree, IMHO.)

> This is pretty painful, write a class outside the ASF that you own the
> copyright to, then go through pain to get it in. Still, it's the
> system and we need to try the system before we try to break it :)

There's a sense of proportion here too: one class that was written in an 
hour may not rise to the level of needing a software grant.  This is more 
about whether the code can be considered 'useful' by itself.

For example, over in httpd, we had a module (mod_arm4) that was written by 
a PMC member in his spare time in private and he wanted to donate it, but 
we asked for the software grant.

The entire process was ensuring that the grant was sent in and 
received...pretty low key.

>> The Incubator (with its delegated responsibilities from the board)
>> currently specifies that all external code imports that is going to be
>> maintained at the ASF must have a 'software grant' on file:
>>
>> <http://www.apache.org/licenses/software-grant.txt>
>
> How is this submitted? Does it have to be fax/postal (ie very slow
> feedback loop)? Do we send to Jim or the Incubator?

The software grant is sent to Jim (usually via fax).  But, like our other 
documents, any officer can receive the documents and pass it along to Jim 
through fax or postal too.

> Didn't get anywhere with the Commons CSV stuff earlier, but I probably
> didn't understand the right questions to ask. Currently I'm preparing
> an Incubator proposal and hoping it'll just amount to a software grant
> and then we can be out of the Incubator.

You probably don't need a full fledged proposal - it's just ensuring that 
the IP clearance form is completed and associated papers are on-file with 
the Incubator.

> So another question here, is the de-facto new project template the
> correct way to bring component/code into the sandbox/an existing
> project?

Depends.  If there's code behind it, yes.  But, if it's just adding a new 
directory that has no code and it will all be developed on-list, then the 
Incubator probably doesn't need to get involved.

> The message is starting to get across and we'll get there.

That's all we can ask.  =)  -- justin

---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org