www-jcp-open mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Davanum Srinivas" <dava...@gmail.com>
Subject Re: [DISCUSS] Fear, Uncertainty and Doubt - or, What can I do with a TCK?
Date Tue, 10 Jul 2007 19:40:45 GMT
FWIW, i was definitely puzzled when i got to know about #7 quite recently.

thanks,
dims

On 7/10/07, Dain Sundstrom <dain@iq80.com> wrote:
> 6. When a compliance bug is found via the TCK, a community can be
> notified using spec references without any further interpretations
> that may be in the TCK documentation or comments.  It is possible to
> find the spec references using a lookup table but the text is rarely
> as clear as the code, which leads to the "changed for tck" comments
> that people hate.  This is very problematic when integrating
> libraries form external communities as discussions typically need to
> enter the NDA arena.
>
> 7. We only share the test harness code with other organizations that
> have a TCK license and want to verify our work.  Since there tends to
> be years or work in the harness code, a third party needs the harness
> code to verify compliance of derivative works (e.g., WasCE).
>
> -dain
>
> On Jul 10, 2007, at 10:33 AM, Matt Hogstrom wrote:
>
> > I've been working on Apache Geronimo for a couple of years now and
> > have heard several different stories on what one can do and can't
> > do once an NDA has been signed.  My objective on this thread is to
> > coalesce our communal intelligence on this matter and document the
> > rules of engagement so the projects at the ASF, where necessary,
> > operate on the same guidelines with the confidential material we
> > have received.
> >
> > I have not seen any formal rules of engagement so here is how we
> > operate in Geronimo.  Our means of operation may be paranoid and if
> > we are doing too much then I'd like to be3 able to let some of it
> > go.  Here is a set of rules we use (passed on from committer to
> > committer so these may not all be actual requirements):
> >
> > 1. No XSDs provided by Sun can be reshipped because the copyright
> > statement in them (as provided from the Sun Website) has a clause
> > that prohibits redistribution.
> >     a. We believe this only applies to the annotations / comments
> > as being copyrightable.  The other programtic elements are not
> > copyrightable as they are part of the programming specification.
> >     b. As such, we type the XSDs in manually and omit the
> > annotations and comments and then insert an Apache License.  This
> > is considered redistributable.
> >
> > 2. One cannot openly talk about TCK status other than a binary
> > we're certified or we're not certified.  No statements like we've
> > passed 93% of the EJB 3.0 tests.
> >
> > 3. One cannot disclose the contents of the TCK which includes the
> > names of test cases or code from test cases.  In some instances if
> > a problem needs to be discussed on the dev list a new test case is
> > created so the problem can be discussed in that context.
> >
> > 4. One cannot disclose information about the organization or
> > internals of the TCK as provided by the licensee.
> >
> > 5. One cannot talk to other folks (outside of Apache) that are
> > licensed for the same TCK about the results, contents or share
> > experiences as that would violate the NDA.
> >
> > Does anyone else have any folklore tied to TCKs that they would
> > like to get answers to?  Add them on this thread.
> >
> > It would be really useful if Geir and others familiar with our
> > agreements can clear up any misconceptions about TCKs.  I am
> > looking to take the contents of this discussion and add this to the
> > JCP page or other relevant web page so we can have a common point
> > of reference for the actual requirements and not stories passed
> > from father to son :)
> >
> > Thanks
> >
> > Matt
>
>


-- 
Davanum Srinivas :: http://davanum.wordpress.com

Mime
View raw message