Hi, I agree with Gunther, that a documentation project such as he describes would be very interesting and useful.
To the extent this has interested me, I have documented Date/Time handling, Exception programming, Perl style guide (consisten with "perlstyle" but more detailed, and API documentation standards using POD. http://www.officevision.com/pub/p5ee/software/htdocs/api/ So I agree that this is another great way for people to contribute. To a large extent, that is what I had originally envisioned. Various people simply write up specs on how they did enterprise development in perl and what CPAN modules they used. http://mathforum.org/epigone/modperl/premangdoo/3.0.6.32.20010811121011.0087 [EMAIL PROTECTED] http://archive.develooper.com/p5ee%40perl.org/msg00058.html http://archive.develooper.com/p5ee%40perl.org/msg00101.html If anyone sees anything they wish to do in terms of documenting Enterprise Programming practices or experiences in Perl, that is another great way to contribute. Stephen P.S. see more comments below. At 08:32 PM 2/15/2002 +0800, Gunther Birznieks wrote: >I would like to offer my original suggestion again that I posted when this >list first started. > >That of not making Phase 1 of P5EE into a means of coming up with common >interfaces that loosely mirror J2EE. > >I still believe the quickest route to P5EE acceptance is if it is first and >foremost a *documentation project* that basically provides a 1-stop place >to go for people who intend to do "Enterprise" programming in Perl and want >to know where to go when they want to solve certain problems. > >To a large degree I believe MOST of the P5EE problems have been solved by >existing frameworks. What's needed is not something that will be yet >another framework (YAF) but rather a document to help people wade through >what exists similar to what the mod_perl guide did for mod_perl. > >But in this case it would be a P5EE guide to CPAN. What exists on CPAN (and >elsewhere) written in Perl that can be used to solve things that people >might do in J2EE and what are the pros and cons to these items. > >It is at this point that this document is written that we can then truly >begin to write code. > >Now, I do believe there is some code that will be useful for the future >such as serialization, jndi equivalent etc... but really, I still believe >so much exists out there that it's quite tough. > >I also think the nature of open source development precludes poeple from >not wanting to participate in a project of this size. The reason is that >open source people want to maintain their own bits and bobs of what they >develop. But if they give it "up" to P5EE, then they merely become a more >anonymous cog in a larger machine just as the J2EE programmer who came up >with JNDI is who?? This is a good point. I would like to make the P5EE as thin as possible: a software "backplane" or "spine" into which many separately developed packages can plug. >I guess what I am saying is that making P5EE into a documentation project >rather than a Borg-project of assimilating technological distinctiveness of >individual projects into a P5EE-collective seems like it would leverage the >culture of Perl and open source development a lot more and would not be as >daunting a task. > >I am also open to keeping the same track as P5EE is going now. > >But seeing that there is a lack of enthusiasm with coding for P5EE, I can >only think there must be SOME reason behind it other than that people are >"busy". The lack of definitions cannot be the only reason. There's a ton of >discussion on this in the archives of this mailing list which I think ad >nauseum defined what P5EE was. > >I can only imagine that the task must seem so daunting, so huge, that by >its nature it just doesn't seem doable or worth individual efforts and this >is why few people are contributing. I think that is likely the reason. >I have a small excuse. I program Java almost exclusively these days. :) But >I do know that there are people on this list who definitely contribute to >open source Perl on CPAN and that in the last 6 months they've been quiet >on here yet I've seen them release stuff for mod_perl and other projects. >So there must be a reason for that as well. Somehow those projects invite >contribution in a way that P5EE isn't. > >Of course, when you write a module for mod_perl, it doesn't get assimilated >(to use the Borg term) into a mod_perl collective. It is an individual >module that gets *associated* with mod_perl as opposed to assimilated into >a larger project. But an individual wrote that module, can be proud of it, >and is the maintainer. It's not assimilated but associated. > >I guess I can't help but think the tact of P5EE has always felt to me like >it was about donating code to P5EE if you were a programmer for it and >although your name might go on the P5EE byline, you would lose control over >your own babies too fast as there is no encouragement of individual >ownership of contributions. For example, I recall posts on here encouraging >people to even state that they would give up copyright to some father >organization such as YAS if they contributed code to P5EE. Good thoughts to keep in mind. We'll see what happens. >Later, > Gunther Stephen
