Mailing-List: contact jcp-open-help@apache.org; run by ezmlm
Precedence: bulk
Reply-To: jcp-open@apache.org
Received-SPF: neutral (asf.osuosl.org: local policy)
Message-ID: <44450793.2080100@apache.org>
Date: Tue, 18 Apr 2006 11:36:51 -0400
From: Geir Magnusson Jr <geirm@apache.org>
Reply-To: geirm@apache.org
User-Agent: Thunderbird 1.5 (Windows/20051201)
MIME-Version: 1.0
To: jcp-open@apache.org
Subject: Re: EJB3 vote
References: <4444E164.3030709@pobox.com>
 <CD3E17F7-2027-45C1-9021-AF2619A063AD@iq80.com>
In-Reply-To: <CD3E17F7-2027-45C1-9021-AF2619A063AD@iq80.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

and thx for the reminder re Cayenne

Dain Sundstrom wrote:
> I couldn't agree more.
> 
> Can't we get them to call them EJB3-part1 and EJB3-part2 and give them 
> separate JSR numbers?  I think this is a really bad precedent for the 
> JCP to set.  Maybe we can get the EC to make an official "never again" 
> statement :)
> 
> BTW we also have http://incubator.apache.org/projects/cayenne.html which 
> is working on a JPA implementation.
> 
> -dain
> 
> On Apr 18, 2006, at 5:53 AM, Geir Magnusson Jr wrote:
> 
>> The EJB3 spec is up for final approval vote this week.
>>
>> There's murmurings that there will be not TCK available for the JPA 
>> subspec of the EJB3 spec.
>>
>> I find this completely unacceptable, and unless someone can give me a 
>> darn good reason, I plan to protest EJB3 being up for a vote on simple 
>> process grounds (there is no TCK) and failing that, I wish to vote 
>> against the spec.  I'm fully aware that EJB3 failing a vote prevents 
>> Java EE 5 from completing, but Sun is as well, and seems to be ok w/ 
>> that risk.
>>
>> When this mess was created two years ago, Sun as the spec lead 
>> promised that there would be a separate TCK for JPA.  While I was 
>> against the idea of having JPA buried in EJB3 spec - I had been 
>> strongly lobbying for it to be an independent JSR - having a separate 
>> TCK meant that some of the problematic issues with the situation were 
>> resolved - that we could do nice things w/ JPA in J2SE/Java SE, 
>> independent of J2EE/Java EE.  The terms of the "deal" were clear.
>>
>> This affects us - there is a new podling, Open JPA, that solves a lot 
>> of problems for many projects at the ASF, such as Geronimo for it's 
>> EJB3 implementation and many others that wish to use the O/R mapping 
>> approach to object persistence.  It's a way to solve the "hibernate 
>> problem" for projects like Roller.
>>
>> Anyway, in the event that the murmurings are true and that the TCK 
>> won't be made available, I will act early as I wish to give Sun as 
>> much time as possible to remedy the situation and provide the TCK as 
>> required by the JSPA.
>>
>> geir
> 
>