Perlbal moved to Github
While on news (and starting posts with the word "while"), Perlbal has moved to Github.
I have a lot of issues revolving Perlbal. I wish there was more development on it, I wish it catered to more generic situations (I've found myself needing a few good hacks for it and abandoning ideas because of it), I wish there were more hooks, there was more DOCUMENTATION and a decent website. I wish, I wish, I wish.
I have been trying to do several of these but my tuits have run out (expect a post relating to this).
Hopefully with the move to Github, it will be able to accept changes from the community more easily and perhaps fuel more development into it.
Good luck to it!
Gots to do the bloggingz
Regarding documentation, there is a PerlFoundation grant for that:
http://www.perlfoundation.org/jose_castro_bruno_martins_perlbal_documentation
Its active since August 2010. AFAIK, there hasn't been any reports from them, but I'll ask José when I see him.
I don't know where the documentation is being written to, but maybe we could suggest they use the newly created github repo.
I saw this grant a while ago, and I remember thinking "IT'S ABOUT DAMN TIME!" but it's unfortunate to hear there haven't been any reports on the grant. This doesn't seem like a very difficult task (knowing a bit of Perlbal to write a few plugins, install it and configure some advanced stuff) and the applicants seem way more than "able" to accomplish it.
Thank you for contacting José to see what the status is.
I agree the Github repo would be an excellent place for them to put it since it could be viewed, approved, fixed or reviewed by anyone and merged in faster than.. something very fast!
Hope it works.
Have you sent questions to the mailing list? There's a fair amount of existing documentation as is, but extending Perlbal can require some learning.
I have sent a few questions actually, and received replies for them. People were kind and responsive.
It seems like the problem can be described by a feeling that Perlbal takes care of what Perlbal writers care about. This is not necessarily a bad thing but it isn't the best paradigm for a public project.
For example, a public project usually cares about how it looks, so it creates a website that hopes to be accessible - something Perlbal doesn't have.
A public project usually cares about making the documentation more available. I've learned more using a presentation someone gave rather than the actual docs (which took me a LONG time to find). Presentations, slides, tutorials, aren't really available through the website - you need Google for that. The official documentation isn't even available on the website, which is horrible.
A public project might care about cleaning up code, using more modern tools and provide more hooks, more features and the sorts. Perlbal uses obsolete programming methods and extremely-deprecated behaviors. I understand the desire to support older Perls but even the RedHat/CentOS people (who are pretty much the oldest) have Perl 5.8.8.
There is just so much that needs to be done in Perlbal but it seems like it all revolves around how much you perceive your project as a public project and not a private one. It feels like the authors see it as a private one instead.
I don't completely understand what you mean by it being a 'public project'. Perlbal is open source, has liberal license, and is well supported by the developers. It is by no means a private project - anyone can submit patches, and they are readily accepted if they meet the quality standards.
The developers mostly care about making it bug free, stable, and performant. If you spend the time to grok the event driven plugin architecture, you'll find that extending Perlbal isn't that difficult if you take the time to learn how to do it.
Part of making Perlbal compatible for lots of users is supporting older versions of Perl like 5.6. Cleaning up code is a very broad statement, and often code cleanups that have no specific goal result in a broken codebase. For what it is worth, the Perlbal code is some of the cleanest I've seen - dense, but clean.
You might try posting specific ideas about what you want to do with it to the list; I'm sure you'll get some good direction. You haven't specified exactly what you need it to do - only that you wish it was 'more generic'. Apache might be more what you are looking for there - it has many more modules ready to use, and you don't have to write a line of code.
There's a lot there in Perlbal already - but it sounds like you need to decide exactly what you are trying to accomplish, and give it some effort of your own.
It is by no means a private project [...]
Did I say it was a private project? I apologize. I should have been more accurate. Let me try again:
"it isn't the best paradigm for a public project" - That means that while I know Perlbal is a public project, it seems like the authors have picked (perhaps implicitly rather than on purpose) a paradigm that isn't the best for it, IMHO. This isn't to say that I think Perlbal is not a public project.
"It feels like the authors see it as a private one instead." - the emphasis here should be on feels. It feels like it's being viewed as private and not public - even though it is public.
If you spend the time to grok the event driven plugin architecture, you'll find that extending Perlbal isn't that difficult if you take the time to learn how to do it.
I actually have, for the most part. However, my criticism still remains. The plugin architecture isn't well documented. At least it's not "well documented" for me. That is not to say that it's difficult to do, but that it should just be better documented. To do that, you need to care about doing it and it doesn't feel like the authors care about doing it.
the Perlbal code is some of the cleanest I've seen - dense, but clean.
Really? We have different standards and experience probably. You must have a lot more experience than me, and I'm not being a cynical ass here, I sincerely mean it.
Me, I prefer cleaner code. That is not to say Perlbal code is ugly - it isn't - but it can be cleaner and that's a good goal IMHO.
You might try posting specific ideas about what you want to do with it to the list
I did.
You haven't specified exactly what you need it to do
Not in this post, no. But I did in the list.