Mailing-List: contact ant-dev-help@jakarta.apache.org; run by ezmlm
Errors-To: <duncan@x180.net>
User-Agent: Microsoft-Entourage/9.0.2509
Date: Mon, 18 Dec 2000 20:51:59 -0800
Subject: Re: Whoa Bessie... Was -- Re: [Proposal] AntFarm
From: James Duncan Davidson <duncan@x180.net>
To: <ant-dev@jakarta.apache.org>
Message-ID: <B664296E.CFA8%duncan@x180.net>
In-Reply-To: <00ae01c068f7$12bed140$80dc1fcb@cognetnt>
Mime-version: 1.0
Content-type: text/plain; charset="US-ASCII"
Content-transfer-encoding: 7bit

On 12/18/00 5:28 AM, "Conor MacNeill" <conor@m64.com> wrote:

> I think that must have been difficult to write :-) (This may sound cruel, but
> as you say, what the hell). While you may have come up with the original
> concept of ant, I think you also neglected it for a long time.

Yep. That's fair. Like I've said, I was off doing good things elsewhere in
Open Source land.

> I think it is fair to say that was somewhat due to the work of Stefan and
> myself (my turn to sound arrogant, I guess, although Stefan deserves the most
> credit).

Yes, you do deserve kudo's for keeping it going. I'm not happy about being
absent. 

> I didn't see it that way. In fact, I thought we brought some order to the
> chaos. Getting release 1.1 out there by itself rather than as an component of
> Tomcat was very important and really kicked along the acceptance of ant as a
> separate entity. .nt 1.2 also seems to have been pretty well received by
> users.

Very well possible. It could be like presidents where the effects of the
past are always off by at least one (ie, the good economy of the Clinton was
probably due to presidents before him -- please no political debates from
here.. :)

>> And features have changed between releases. Sometimes being added, sometimes
>> being removed.
>> 
> Of course features have changed. The automatic find option is a minor issue. A
> bigger one is the immutability of properties. We are trying to find the right
> way to do that and it is only by pushing code out into the community that we
> can really now if it works or not.

But the community has had to deal with that instead of the pain being
brunted by a few people who occupy the role of "architect".

>> Maybe it's because other people are trying to take it where they see it --
>> and its because I wasn't here to help out as much as I should have been. But
>> I still feel a strong sense of ownership of Ant. And a strong desire to make
>> it into what it should have been to begin with.
>> 
> How would you say the current Ant is radically different from what you want it
> to be?

Its more complicated in syntax.. Its less flexible in picking up tasks.. And
it doesn't lend itself to integration well. And it's a pain in the arse to
deal with all those scripts which were intended to go away.

> Is that the model you really want?. I don't know JDOM at all, but I wonder
> what sort of community will build up around it. To me OpenSource is not just
> about having access to the source. It is about the community that builds up
> around that source. In Apache projects, you can achieve recognition for your
> contributions by being voted a committer. It is an incentive to contribute.

JDOM has a pretty active community around it. There are people besides Jas
and Brett that are committers. But Jas and Brett are "Project Leads". If you
want a change, you can argue with them, but when there is disagreement, they
get dibs.

> Isn't that what the concept of a committer is all about. It is the recognition
> of your contributions by peers. How do you want to go further? Some  sort of
> perpetual ownership for the original contribution?

Well, at least not being marginalized out... If you create something you
obviously have a vested interested -- and a vision that may not be easily
communicated via any other means.

>> I'm quite happy with competing against Mymidon -- and for the two of us to
>> steal as many ideas as we can from each other. But if Mymidon succeeds, I
>> really think that it should be as Mymidon. Not Ant.
> 
> So who gets to define what is Ant? Is that you?

Bluntly, yes. With the help of a lot of people. But where there's
disagreement... Who defined Cocoon. Stefano. Who defines Apache 2.0? Ryan.
Who defines Perl? Larry Wall. There's a pattern there. Collaborative
development still needs a lead.

>> To make a point -- look at how we refer to Tomcat 4.0 -- we call it
>> "Catalina" in casual conversation even though it's Tomcat 4. Why do we do it?
>> 
> I thought it was Tomcat = Catalina + Jasper

Now it is.. Before it was Tomcat = Tomcat + Jasper. Tomcat was just a
servlet engine that I wrote a long time ago. Anil V. and a few others wrote
Jasper. Some other people whacked on it a bit before it went public in
Jakarta. Then Craig said "this sucks" and built Catalina. So, the current
equation is T = C + J; but it used to be T = T + J; as confusing as that is.

> Wierd indeed. I wonder if O'Reilly would want a book about AntEater? :-) I'm
> curious, will your book describe tasks and concepts contributed to ant by
> other people. That will put you in an interesting position :-)

Why would it do so? I didn't say that I wanted to write every line of code
in Ant. I think you have taken what I said in a way that I didn't intend it.
Quite frankly if the core of Ant is good, I'd rather not write *any* of the
tasks if I could get away with it. I'd just want them to follow some
conventions. :)

And would O'Reilly want a book about AntEater that wasn't Ant. Who knows. It
would be a hell of a sell job.

BTW. There is a chapter in the outline for external tasks. I don't see how
that would put me in an interesting position. From an authorship
perspective, I think that it's important that Ant act they way it should --
and it doesn't matter how it's coded as long as it works.

-- 
James Duncan Davidson                                        duncan@x180.net
                                                                  !try; do()