cocoon-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Stefano Mazzocchi <stef...@apache.org>
Subject Re: Cocoon 2.0 Salutation!!!
Date Mon, 14 Feb 2000 22:12:44 GMT
Pierpaolo Fumagalli wrote:
> 
> Stefano Mazzocchi wrote:
> >
> > Well, to be _brutally_ honest (as always), I think that Pier's branch is
> > an alpha0 and it fully deserves it. I mean, there are some interfaces
> > that, man, made me jump off the chair and scream and yell and ...
> 
> I know that... And I keep repeating it.. But noone yet came out and made
> improvements... I just want FEEDBACK...

It's coming... unlike you, I do have a life :)
 
> > anyway... I'm pleased that Niclas is pleased and even if I think that
> > _most_ of the interfaces that Pier designed must be changed or improved,
> > I do believe that we have a _clear_ and _global_ vision of how an XML
> > framework on the server side should look like.
> 
> I totally agree w/ you... Right now the thing is a hack to permit some
> kind of work... I'm using it, right now, and that pleases me... But I
> totally agree that a bigger redesign of all the factorying method is a
> MUST and is nedeed NOW.

I'm in the process of doing a major redesign of the Cocoon internals
because there is something that keeps on bugging me... see my next mail.
 
> > Add SAX2, the proposed API for XML Transformations, XSP and XSLT
> > extentions, sitemap capabilities, avalon-like modularity, a powerful
> > build system, some authoring tools... and man, next year, cocoon will be
> > so far ahead of its times that it will be hard for us to continue
> 
> SAX2 is already there (cvs update, stefano, cvs update!), regarding the
> XSLT API, I don't care much, since IMVHO XSLT is just another filter...

I was speaking marketing-ish :)

> Regarding XSP, I'm looking for an XSLT engine supporting taglibs, and
> operating in SAX, but it seems that nobody cares about those small
> details of implementation :(

What do you mean with "taglibs", XSLT-extentions?
 
> > Yes, Niclas, your prayers were heard, but we were _already_ aware of
> > those problems... it seemed to me that you were simply blindly not
> > paying attentions to our discussions.
> 
> Hehehehe :) When did we start talking about "Cocoon, to servlet or not
> to servlet, this is the question" ????? I believe it was around august,
> when I was at IBM...
>
> > > I didn't layered it out... Hey, I just CODED it... The code is not my
> > > (C) not my idea...
> >
> > And I can tell.... look at AbstractComponent
> >  setCocoonInstance(Cocoon);
> > Aghhhhhhhhh!!!!!!!!!!!!
> 
> Hey, I hate that thing too... So shut up and find something better...
> Give me a hint and I'll code it...

Sheees, what's the english for "permaloso" :)
 
> > Yes. The Cocoon2 design started in May last year when we firstly
> > discussed the use of Cocoon on big environments at the
> > Exoffice-sponsored exolab. Almost all the people that got into that
> > adventure and were subscribed to this list were hired by Exoffice
> > (Arkin, Keith, Pier, Ricardo, Federico) ... Clark and I are the only one
> > that resisted the attack :)
> 
> Hahahaha :) The thing is that one have to decide what to do, and I love
> working with my friends... So, things come easy for them, you get one,
> you get all of us...
> And you will resist not for so long :) :) :)

I'm serious, when I'm finished with my college I'll put myself on Ebay,
then we'll see :)
 
> > Anyway, this is to show that we already knew the problems but we didn't
> > want to do another "patch and go".. doing a revolution in a open source
> > project is a _very_ dangerous approach. You must understand where to go
> > way before doing or the project will be forked.
> 
> IMVHO, Cocoon 2.0 is not a "revolution", but the natural evolution of
> the 1.x... 

Oh, sure, from a management point of view there is no "revolution" when
all the parties agree in doing it.

But if you rewrite all the interfaces, this is a revolution in a open
source sense. This is what I meant.

> I'm thinking about how to re-enable all the stuff done in the
> 1.x thread, and it's possible. I still have to give some thought about
> it, but I believe we found a way (last thursday, with Keith)...

Please, don't. Most of the thing from the 1.x tree either come
automatically (with the avalon-like componentized framework I'm adding),
come with code cut and paste or should not come at all!

In my next mail I will explain why I think the XSPProcessor should be
totally rewritten... in fact, Cocoon2 will be the XSPProcessor.... well,
sort of..
 
> > No, seriously, Pier sucks in creating interfaces and architectures, but
> > he's an _awesome_ coder. that's why we work together so well, we are
> > complementary (I hate writing code!)
> 
> Kiddo, I did my homework... I gave you something you can see and refine,
> revolt, or redo from scratch... NOW STOP TALKING AND GET TO WORK... I
> NEED THOSE F***ING INTERFACES! :)

It's so easy to pick on Pier... and also a lot of fun...

> And anyway I'm not an awesome coder too... Just look at my "packeted"
> coding style, and you'll see...
> 
> And, (sidenote for Stefano's last week flame), when I say something, I
> DO IT and I perfectly know how to evaluate time :) :) :)

Hmmm, again, what's the english for "coda di paglia" :)
 
>         Pier
> 
> BTW, I love the new Cocoon logo, the old one sucks! :)

Agreed....

BTW, I'm planning to make "Cocoon" T-shirts for ApacheCon2000 with the
new logo, who wants one? :)

-- 
Stefano Mazzocchi      One must still have chaos in oneself to be
                          able to give birth to a dancing star.
<stefano@apache.org>                             Friedrich Nietzsche
--------------------------------------------------------------------
 Come to the first official Apache Software Foundation Conference!  
------------------------- http://ApacheCon.Com ---------------------



Mime
View raw message