I'll take that back for now. I went through the specs, it took me 15 minutes,
and I thought, as a bean developer, there are too many extra new rules to
remember if one wants to use CMP. I mean, hey, I use CMP because I
don't want to screw around with a million details, no? :-)
I figure that one could slap the new java.jdo stuff on the back-end and retain
the front-end - the container and bean developer contract as is, so that a bean
developer does not have to learn anything new. On a second thought, though, it
may not be possible to slap on a back-end O/R (such as java.jdo) layer without
modifying the current front-end container - bean developer contract.
By the way, when does jboss plan to support EJB 2.0?
Anh
On Mon, 12 Jun 2000, Rickard �berg wrote:
> Hi!
>
> Phan Anh Tran wrote:
> > Anyone thinks EJB 2.0 CMB dependent object class and dependent value class and
> > the contract definition stinks? Waay to many details. And it seems like a lot
> > of the details could have been easily solve with a decent O/R tool. Ie. keep
> > the existing CMP specs and define an O/R tool plug-in interface.
>
> I have the opposite view. I think that the EJB 2.0 CMP design is great:
> a declarative view for bean programmers, and loads of ways for the CMP
> implementor to optimize the usage.
>
> Do you have any specifics on what is bad?
>
> /Rickard
>
> --
> Rickard �berg
>
> @home: +46 13 177937
> Email: [EMAIL PROTECTED]
> http://www.telkel.com
> http://www.jboss.org
> http://www.dreambean.com
>
>
> --
> --------------------------------------------------------------
> To subscribe: [EMAIL PROTECTED]
> To unsubscribe: [EMAIL PROTECTED]
> Problems?: [EMAIL PROTECTED]
--
--------------------------------------------------------------
To subscribe: [EMAIL PROTECTED]
To unsubscribe: [EMAIL PROTECTED]
Problems?: [EMAIL PROTECTED]
