dev@jsr311.java.net

Re: Welcome to JSR 311

From: Dhanji R. Prasanna <dhanji_at_gmail.com>
Date: Tue, 10 Apr 2007 13:26:59 +1000

On 4/10/07, Paul Sandoz <Paul.Sandoz_at_sun.com> wrote:
>
> On Apr 6, 2007, at 1:29 AM, Dhanji R. Prasanna wrote:
>
>
>
> On 4/5/07, Marc Hadley <Marc.Hadley_at_sun.com > wrote:
> >
> >
> > >
>
>
> +1. This should be more intuitive than learning SAAJ for instance. I am
> always for anything that one can learn almost entirely via Javadoc and
> hands-on (which 11 packages may make difficult even if half can be ignored
> by the lay developer).
>
>
> Oh no not SAAJ! If we end up like SAAJ we will have truly failed :-) Given
> that we can limit our interaction with the DOM API i think we might be on
> safer ground :-)
>
> I think the API looks like it is starting achieve one goal, which is to
> help focus discussion.
>

hehe. =)
Yes I suppose my point is that it seems like a quite-lean api on the whole,
so why the package explosion?


> Is there a criteria for choosing what containers get a spi?
>
> Not sure i understand the question, can you give an example?
>

My bad, i misunderstood Marc to mean there were specific containers you
wanted to support in the spi.

We found the container SPI very useful and quick to integrate new
> containers. So we thought it would be useful to present this to the EG and
> see it this is a worth while thing to do i.e. is there a benefit? or is it
> better to leave this type of thing to each implementation?
>

I wasnt able to find a container impl based on the spi in the files on
jsr311.dev.java.net--is there a good example somewhere? I think seeing it in
use would give us a better idea as to its implementability.


Just a couple of trivial qns:
The packages are under javax.ws rather than javax.jws--is this only because
jsr311 is not a WSDL-style webservice?
The packages are named javax.ws.rs, rather than javax.ws.rest as in the
initial proposal--any reason for this change?


Thanks,
Dhanji.