ant-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Alejandro Abdelnur <alejandro.abdel...@sun.com>
Subject Re: on *evil practices*
Date Wed, 12 Sep 2001 01:52:13 GMT
bevan,

i agree with your perception of ms-word, that is why i use it only as a
word processor, the difference here is that it's my choice to do so. i
do not like to be constraint because somebody else thinks is dangerous,
let me shoot myself in my foot if i'm stupid enough to.

i know that there have been long discussions about things of the sort in
the past and i have no intention to start a new one.

the changes [i was asking for in my previous email] are to allow
developers (in this case myself) to use alternate implementations of the
Project, Target and ProjectHelper classes, nothing changes in the
surface of ant, but enables developers (again, in this case myself) to
extend Ant to their own needs without having to modify the ant core thus
having to retrofit changes with every new version of ant (already done
that from 1.2 to 1.4).

thxs and regards.

a

Bevan Arps wrote:

>  At 17:52 11/09/2001 -0700, Alejandro Abdelnur wrote:
>
>
>> i do not think that the way of avoiding *evil practices* should be
>> achieved
>> by trimming the capabilities or the power of expresion of Ant but by
>>
>> provinding guidelines on how to use things. always there are things
>> that
>> look like *evil practices* to purists but are needed by a real task.
>
>
> The key here is "trimming the capabilities or the power of expression"
> of Ant.
>
> There are always decisions to be made in the development of any tool.
> One of the most important areas is to decide what the tool shouldn't
> do.  Any application that tries to be a Swiss-army-knife or a
> jack-of-all-trades inevitably ends up being a poorer tool.
>
> A great example here is MS Word - which tries to be Word Processor,
> Desktop Publisher, Content Reviewer, Document Designer, Mail Client
> (and more) all rolled into one. The first it does extremely well - the
> others less so.
>
>
> The specific functionality that has been rejected from Ant seems, to
> me, to fall into two categories:
>
> -- Shortcuts to functionality already present
>
> Rejecting these doesn't reduce the power of Ant in any way - and does
> serve to keep Ant simple. A good example here is the presence of
> if/unless attributes on the fail task.
>
> -- Dangerous Features
>
> ie Functionality that provides a moderate level of increased ability
> at the expense of great danger.
>
> One common problem common to any tool is the fact that new users tend
> to blame the tool for their own mistakes or misunderstandings. Make a
> tool with too many Dangerous features and new users will find the tool
> too quirky and discard it, looking for something simple.
>
>
> I applaud the Ant Team (take a Bow, all of you) for having the courage
> to "keep the vision".
>
> I might (in fact I do) disagree with some of their decisions - however
> I can't ignore the fact that it is their dedication to "what might be"
> that has given us such a fine tool.
>
> My 3c,
> Bevan.


Mime
View raw message