+1000! The only code that should be submitted is that which is necessary to describe the more complex points of the implementation/architecture. At a minimum, the requirements should be clearly stated and substantiated with use cases. All design points should be clearly linked back to the requirement(s) that precipitated them. Also, integration points (e.g. Ant/Antidote) need to be clearly defined and adhered to religously so that development of dependent projects are facilitated.
jh -----Original Message----- From: Ken Wood [mailto:[EMAIL PROTECTED] Sent: Monday, December 18, 2000 10:11 AM To: [EMAIL PROTECTED] Subject: Re: AW: Whoa Bessie... Was -- Re: [Proposal] AntFarm How about the good old fashioned way: 1. A requirements specification that spells out WHAT the software must do, without ANY indication of how it should be done. 2. A design document that specifies HOW the requirements will be met via a proposed implementation. I suspect the reason we have so many proposals in the form of code is that code is fun to write. Specifications and design documents are NOT fun to write. But, they are a better way communicate a software effort than trying to compare multiple proposals in the form of different codebases. Christoph Wilhelms wrote: > > Sorry, but I have to give a short statement from the view of an Ant user and > "extender"! > > I think it is very simple what we all want and what we dont want. Please > correct me if I am wrong! > > We want: > > * Just ONE Ant! > * A reliable running, fast, innovative Ant (just like Ant 1.2 is now) > * An Ant beeing supported, discussed and improved (just like it has been up > to now) > * Many Ant-Developers that work TOGETHER (on the current and future > versions) > * A clear line! We want to see in wich direction Ant is going! This is > mandatory > for everyone who wants to EXTEND core ant like ANTIDOTE or VAJ-Integration > or > JBuilder-Integration etc. > > We don't want: > > * Competetive Ant-Developers ( ;-) ) > * Many different Ants (Myrmidons, Antfarms, Anteaters, etc.) parallel! Don't > misunderstand > me: the proposals are on one hand good for improving Ant - on the other > hand there is > NOBODY who can compare the proposals with each other (lack of time). IMHO > at the moment > the proposals INTERUPT Ant development-process! > --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
