ant-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Erik Hatcher <>
Subject Re: New Ant GUI - 'Ant's Nest'
Date Fri, 07 Jan 2005 18:47:29 GMT

On Jan 7, 2005, at 12:20 PM, Chris Clohosy wrote:
> Erik, I'm sorry as it seems I've offended you. No excuses, it appears 
> I haven't dug as deep into Ant as I could have. I promise not to be 
> disheartened by your comments (too many students give up after 
> criticism) but to instead look into the resources you mentioned; when 
> I'm not revising for my exams.

I apologize for coming across as offended.  The bulk of my frustration 
comes from my recent personal experiences of dealing with developers at 
the University that I discoverd are not using version control, have not 
read the book on the framework they are stumbling with and do not read 
partake in the e-mail lists, and also seeing various groups here work 
in isolation without communicating between themselves.

Please take the frustration out of my comments and distill the 
constructive parts of it, as you mentioned.


> Chris
> ----- Original Message ----- From: "Erik Hatcher" 
> <>
> To: "Ant Users List" <>
> Sent: Friday, January 07, 2005 4:29 PM
> Subject: Re: New Ant GUI - 'Ant's Nest'
>> On Jan 7, 2005, at 10:31 AM, Chris Clohosy wrote:
>>> Thank you Rich, Erik, and Dominique for your comments, I'll 
>>> certainly look into them. When I first started looking into this 
>>> around May-time, I could not find any easy way of extracting the 
>>> information needed from the Ant class files. I tried everything, 
>>> from reflection to the Ant API itself and could not find anything. 
>>> The best I could do was to look for the bean conventions like 
>>> getters and setters but I didn't want to rely on that. Alos there 
>>> was no way of knowing if an attribute was required, or if it became 
>>> required if another attribute was set, etc. That's way I was heading 
>>> down the meta-data route with annotations, I didn't have a clue they 
>>> were already marked up with @tags!
>>> This is a university project, and at the moment time is running out 
>>> so I have to prioritise what has to be included in the final 
>>> version, so bear with me :-p But you all have definetely shed some 
>>> light on what lies behind Ant, and I'm very grateful.'
>> *sigh* - I work at a University myself, and it saddens me to see how 
>> many projects are done in isolation without communication with 
>> others.  Ant's source code has had these tags in them, readily 
>> apparent.  If you check Ant out from CVS (surely you did this, no?) 
>> you'd see a proposal/xdocs directory where a pretty sophisticated 
>> XDoclet-based generator exists to pull the metadata tags and 
>> integrate with IntrospectionHelper.
>> This issue has been discussed on this e-mail list and the dev one a 
>> fair bit.  These archives are searchable.
>> So, you did not try "everything", as that would have included looking 
>> at the source code in depth and asking around on these lists.  Sorry 
>> for giving you a hard time, but it is frustrating to see wheels 
>> reinvented due to lack of knowledge.  Wheels reinvented for something 
>> creative and better is a different story.
>> What else could we have done to make the knowledge of what already 
>> exists more apparent and easy for you to have found?
>> And like a lot of University projects, they get abandoned just when 
>> things start to get interesting.  What will happen with your project 
>> when it reaches the "final version"?  Will you continue to work on 
>> it?
>> Erik
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail:
>> For additional commands, e-mail:
> ---------------------------------------------------------------------
> To unsubscribe, e-mail:
> For additional commands, e-mail:

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

View raw message