ant-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
Subject Re: Ant 2 et al.
Date Tue, 09 Jul 2002 14:46:58 GMT
On Tue, 9 Jul 2002, Conor MacNeill wrote:

> On Tuesday, July 9, 2002, at 06:54 , Nicola Ken Barozzi wrote:
> >
> > We are all looking at them for inspiration to make Ant1 better and 
> > waiting for them to ask a vote for the codebase switch.
> I'm happy to ask for a vote. I tried once already :-) The problem is 
> that nobody knows what the vote really means.

Please don't :-)

> >> So far, it's mainly been "Who needs the proposals? Ant 1 can do 
> >> anything!".
> >
> > Who needs the proposal implementations?
> > We can put that stuff in the Ant1 codebase.
> Really? I am interested to see that.  :-)
> Ant1 is bit of a mess, really. I tried to give some pointer to this in 
> my Mutant doco.

Well, I think ant1 is one of the cleanest and best documented codebases
in jakarta and xml.apache. And it seems to have one of the best 
communities, and probably is the most used. Not to mention that the 
"Task" and the other core interfaces and the DTD are a de-facto standard.

Try looking at tomcat, xerces, xalan, axis - all very good and successfull
projects. You'll find far more mess and complexity, far less 

I think this is a result of the simple core design in ant1, plus 2-3
years of refinement on the codebase. 

> But, if someone has their own <import> task, they can't use it at the 
> top level since you will effectively introduce a new keyword to Ant. 
> Mutant tries to solve this problem by using a namespace for this sort of 
> metadata. I believe Myrmidon uses a task for <include> although there 
> are issues with that approach which I'm not sure how they have addressed.

Adding namespace support in ant1 is perfectly possible, as you know.
There is a working PluginHelper in proposals, that works with 

> Anyway, my point is that it is probably easy to grab features from the 
> proposals but without the underlying architecture, the result may not 
> always be that good.

My point is that ant1 architecute is one of the best, there is nothing
fundamentally wrong. And from looking at both proposals, I believe
they add complexity and I don't think the result is better.

Of course, any proposal needs to start by saying that whatever was
before is broken and can never be fixed. Most revolutions I know
are far for perfect, and some were far worse than what it was
before ( I'm thinking about general history here, not jakarta :-)

I still have to see one real issue that can't be resolved by 
the current codebase but can be by a proposal. 

> > Not really.
> > The proposals try to make the requirements real.
> > Ant1 codebase tries to assimilate as much as possible without 
> > snaturating.
> Let me give an analogy. You live in the leaning tower of Pisa and you 
> see your neighbours are adding rooftop pools to their building. They 
> look pretty cool but putting one on your roof may leave you a little 
> wet :-)

Only if you make it parallel with the ceiling. I've seen pools
on non-horizontal surfaces, they look good. 


To unsubscribe, e-mail:   <>
For additional commands, e-mail: <>

View raw message