[a5] is the time really now?

[Peter Royal <pr...@apache.org>]

There has been a *LOT* of traffic recently discussing dreams of what v5 
should look like. There is obviously lots of time and brainpower going into 
these discussions. My question is, is now really the time to be embarking 
upon this redesign? 

v4 is barely a year old. There are a fair number of projects built on top of 
it already. IMHO our energy might be better spent enhancing what we have now, 
making it more attractive to other developers w/ solid component 
dependencies, metainfo that is portable across multiple container 
implementations, and everything else that makes a project "mature".

We all seem to love the design process and making something "just right", but 
is it not "good enough" right now? Maybe I've just been hardened by the real 
world, there's never enough time to achieve api nirvana.....

just some thoughts. 
-pete

-- 
peter royal -> proyal@apache.org

--
To unsubscribe, e-mail:   <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>

Re: [a5] is the time really now?

[Leo Simons <le...@apache.org>]

On Mon, 2002-06-17 at 22:35, Peter Royal wrote:
> There has been a *LOT* of traffic recently discussing dreams of what v5 
> should look like. There is obviously lots of time and brainpower going into 
> these discussions. My question is, is now really the time to be embarking 
> upon this redesign? 

I think the good part here is that starting of with the notion "as long
as we provide migration and compatibility tools, it is okay to remove
deprecated stuff" makes the discussion a bit easier. It seems to me
that, once we have a picture of what we would do in that case, what will
become 4.2, 4.3, 4.4 etc will be less hassle and a better product.

The direction we are going in so far (looking at Berin's summary and
Steve's comments) indicates that those 4.2, 4.3, 4.4 releases will
indeed fill the needs we have, making A5 uneccessary for a long time
yet.

> v4 is barely a year old. There are a fair number of projects built on top of 
> it already. IMHO our energy might be better spent enhancing what we have now, 
> making it more attractive to other developers w/ solid component 
> dependencies, metainfo that is portable across multiple container 
> implementations, and everything else that makes a project "mature".
> 
> We all seem to love the design process and making something "just right", but 
> is it not "good enough" right now? Maybe I've just been hardened by the real 
> world, there's never enough time to achieve api nirvana.....

=)

If you look at the whole A5 discussion as centering around "what should
we refactor?", and then the answer mostly being like "this is cool, that
is cool, let's tweak this, let's formalize that, let's add some
functionality here" (as opposed to "dump this, completely revamp that"),
that seems to indicate we're pretty close to nirvana, and little
refactoring is required.

The thing that gets a lot of attention is the ComponentManager/
ComponentSelector/ Component / Container bit, which has been out on the
table since I got here and is about the #1 point of misuse. We're slowly
making progress (since we've figured out that we should have metainfo in
a separate place).

If it takes a version 5 to settle this issue in the end, so be it. It is
probably worth it.

regards,

- LSD



--
To unsubscribe, e-mail:   <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>

Re: [a5] is the time really now?

[Stephen McConnell <mc...@osm.net>]

Peter Royal wrote:

>There has been a *LOT* of traffic recently discussing dreams of what v5 
>should look like. There is obviously lots of time and brainpower going into 
>these discussions. My question is, is now really the time to be embarking 
>upon this redesign? 
>
>v4 is barely a year old. There are a fair number of projects built on top of 
>it already. IMHO our energy might be better spent enhancing what we have now, 
>making it more attractive to other developers w/ solid component 
>dependencies, metainfo that is portable across multiple container 
>implementations, and everything else that makes a project "mature".
>
>We all seem to love the design process and making something "just right", but 
>is it not "good enough" right now? Maybe I've just been hardened by the real 
>world, there's never enough time to achieve api nirvana.....
>
>just some thoughts. 
>-pete
>
>  
>

I am real happy with A4.
I think there is a lot we can do to strength A4 along the lines you
noted above without breaking anything in A4.  

Cheers, Steve.

-- 

Stephen J. McConnell

OSM SARL
digital products for a global economy
mailto:mcconnell@osm.net
http://www.osm.net




--
To unsubscribe, e-mail:   <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>

RE: [a5] is the time really now?

[Stephen Haberman <st...@chase3000.com>]

I really like the idea of filling out A4 a bit more before embarking on
A5. I've been reading the Avalon docs off and on over the last week and
have been really impressed.

However, I want to use Avalon on a project coming up and my boss would
be much more open to using an A4 that has matured and easier for us to
develop with than using an A5 that is API-perfect but less mature.

Just my opinion; I'm still new enough that I don't have a clue as to
what a "mature" A4 would constitute, other than, from a user's
standpoint, I like it.

- Stephen

> -----Original Message-----
> From: Peter Royal [mailto:proyal@apache.org]
> Sent: Monday, June 17, 2002 3:35 PM
> To: avalon-dev@jakarta.apache.org
> Subject: [a5] is the time really now?
> 
> There has been a *LOT* of traffic recently discussing dreams of what
v5
> should look like. There is obviously lots of time and brainpower going
into
> these discussions. My question is, is now really the time to be
embarking
> upon this redesign?
> 
> v4 is barely a year old. There are a fair number of projects built on
top of
> it already. IMHO our energy might be better spent enhancing what we
have now,
> making it more attractive to other developers w/ solid component
> dependencies, metainfo that is portable across multiple container
> implementations, and everything else that makes a project "mature".
> 
> We all seem to love the design process and making something "just
right", but
> is it not "good enough" right now? Maybe I've just been hardened by
the real
> world, there's never enough time to achieve api nirvana.....
> 
> just some thoughts.
> -pete
> 
> --
> peter royal -> proyal@apache.org
> 
> --
> To unsubscribe, e-mail:
<ma...@jakarta.apache.org>
> For additional commands, e-mail:
<ma...@jakarta.apache.org>




--
To unsubscribe, e-mail:   <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>