On Thu, Aug 29, 2002 at 05:02:39PM +0100, James Duncan wrote: > On Thu, 2002-08-29 at 21:40, Nigel Hamilton wrote: > > What do you mean by 'touch it and die'? > > > > Is it some sort of 'boogie man'? > > Yes, it is a boogie man. As yet I've not seen any general consensus on > what p5ee actually should do; it seems a little premature for voting on > a code-base for the *enterprise* edition of Perl. Hell, every time > having a Perl SDK is brought up p5p grins to a halt and the > flamethrowers roar into action over what is or isn't a useful module.
+1 to James on this. Choosing a codebase to evolve into an "enterprise perl" at the moment seems way premature: what problems is this supposed to solve? what kinds of applications is this supposed to enable? what are the other user expectations of what an enterprise foundation should provide? is this codebase on track to solve those problems and meet those expectations? These are key issues that revolve around a central question: what unmet needs exist that require p5ee be developed? Here's the p5ee mission statement (http://www.officevision.com/pub/p5ee/): The mission of the P5EE project is to promote the development, deployment, and acceptance of Enterprise Systems written in Perl. >From that page, I don't see any concrete details on what features are found in an "enterprise system", which of those features are lacking in Perl today (if any), and why this lack of features needs to be addressed with P5EE. Thus, I cannot vote on renaming P5EEx::Blue to P5EE because I'm not entirely sure what P5EE is supposed to be, nor whether P5EEx::Blue is on track to meeting the goals for P5EE. It seems to me that the easiest (and laziest) way to accomplish the P5EE mission (as stated above) is to sit down and determine a series of features (possibly disjoint) that characterize enterprise applications. From there, figure out what exists on CPAN *today*, and document that as the current "enterprise foundation" in its varied, postmodern splendor. More requirements will come out of that process (such as making component integration simpler and easier, logging APIs, XML processing, etc.). In the end, P5EE will not be determined by the discussion of a few dozen people on [EMAIL PROTECTED]; it will be determined by the adoption of the products of this project -- both code and docs, but not email and votes. This is the same pattern that has lead to adoption and widespread use of DBI, mod_perl, Template Toolkit, and SpamAssassin among many other [Perl] projects. Z.
