Mailing-List: contact repository-help@apache.org; run by ezmlm
Precedence: bulk
Reply-To: repository@apache.org
Received-SPF: pass (asf.osuosl.org: domain of steve.loughran@gmail.com
 designates 64.233.184.200 as permitted sender)
DomainKey-Signature: a=rsa-sha1; q=dns; c=nofws;
        s=beta; d=gmail.com;
        h=received:message-id:date:from:to:subject:in-reply-to:mime-version:content-type:content-transfer-encoding:content-disposition:references;
        b=PYak3X03GO46fHjjxMZzxg1tzwqx71rliVNAM1H1+VnOKGiFQbtqaO+20rUTwNcAlbxVhYzInj+DkF5uUj5Y1L38uKGv+UgGmZB2KRYu2VSz/aMFJ812tNPFXkYxHsbpfEHAKAWCiqoDWuHH0bj2y7+J6sWTSFq4rH/dWcSCebM=
Message-ID: <c84a50780601021234w2fab47d4q7770a72d371e4b8b@mail.gmail.com>
Date: Mon, 2 Jan 2006 20:34:03 +0000
From: Steve Loughran <steve.loughran@gmail.com>
To: repository@apache.org
Subject: Re: Making a redist of all the javax.* packages
In-Reply-To: <43B97B95.7020609@maven.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline
References: <NBBBJGEAGJAKLIDBKJOPOENPFJAC.noel@devtech.com>
	 <43B97B95.7020609@maven.org>

On 1/2/06, Jason van Zyl <jason@maven.org> wrote:
> Noel J. Bergman wrote:
> > Jason van Zyl wrote:
> >
> >> 1. Sun has never changed the license for things like JavaMail because
> >> although it is technically illegal to make Sun Binary License artifact=
s
> >> available outside a distribtion, Sun is not going to go after anyone.
> >
> > If we want a change in BCL, we should bring it through Geir and Cliff *=
now*.
> > We spoke with Simon Phipps about all of this at ApacheCon.
>
> Geir knows and has been unsuccessful in the last 4 years to do anything.
> It just seems to be a non-issue at Sun.

Their OSS projects hit the problem. For example netbeans has some
thing that you have to click through to download some JARs, for
automated builds you need to talk to their lawyers and you get a key
to bypass the click through.

So, they have solved the problem not by coming up with sensible T&Cs,
but a convoluted access process.

>
> > Furthermore, the Glassfish versions of the packages are under what so f=
ar
> > looks to be an acceptable license.
>
> I think taking those if possible or scraping would be more productive
> then trying to talk to anyone at Sun. I would be elated if Cliff could
> do anything but nothing has happened in the last four years since the
> first time I broached the topic Geir.
>
> >> There is not a single case, that I'm aware of, where Sun has pressured
> >> anyone to remove SBL artifacts from a site or SCM. We did this at Apac=
he
> >> because it is technically not allowed.
> >
> > And there are some people at the ASF who don't believe that we should u=
se
> > BCL licensed artifacts AT ALL.
>
> That's fine and dandy but not very pragmatic in a lot of cases.


Well, if we cant do any convoluted workarounds, so be it.

I am still doing an example application for the sequel to Java
Development with Ant, one that includes a hibernate based,
tomcat-hosted application, and the ant-based build process to go with
it, a process that can do sourceforge-ftp and email redistributions as
part of the build.

I think I could legitimately include all the servlet, mail, ejb JAR
files as part of the redist, because they really are needed for the
application (and the meta app, the build process) to work. I'd just
include the whole thing in an installer that put them in the m2
layout, but popped up a sun license first.

You know what really annoys me? All this inconvenience, its all there
because Sun's lawyers didnt lock down their agreement with ms enough,
and got stuffed when Visual J++ added new keywords and stuff in the
java.* packages, then left out rmi. The current paperwork is intended
to stop MS doing that again. But they arent going to? Why would MS
want to copy JAX-WS, or the servlet API. They've abandoned Java for
C#, and its own APIs.