It sounds like your app has some special needs. I also understand "history" influencing the design that exists today; it's something we fight here and the improvement never comes as fast as we'd like.

> I would like you consider the hubris necessary to believe that home-rolled code is going to be better, on any level, than what two decades of iterative development, deployment, and testing across several languages by the best developers in each can produce.

You obviously have enough experience to know that the answer to "best" is "it depends on the situation". It also sounds like you probably have the experience, like I do, to have scars from frameworks that promise so much and almost do what you want yet don't quite fit your needs.

Perhaps the system I work on is "simplistic" enough that I don't have the same needs you do. The vast majority of our code sits in PM and SQL files that understand how to do the business logic that drives our system (the Model). The "GUI handling" parts are actually quite small in comparison, so the "Controller" aspects are simple and don't require all that much effort. Auth/Sessions are the only hard parts for us and they aren't hard given a few modules from CPAN.

It may be a misunderstanding on my part, so feel free to educate me and anyone else who reads this, but how does a PSGI framework help me when the vast majority of our code is to support the Model part that that would have to be the same no matter what framework I use? The majority of my time in bringing up a new page is spent on (largest time sinks first) the Model part, then the Javascript for the dynamic pieces, then a page template, and finally the Controller (CGI and calls to the Model). The first 3 are framework independent (meaning I'd have to do them regardless of the chosen framework). For us, the last piece which is framework dependent is quite small.

Please note, I'm not saying the PSGI framework couldn't be used or even helpful to us, but when you have a fully working system, then any large architectural change requires a pretty high "wow factor" to convince us that we need to learn a new framework and change. That's where I'm coming from. Perhaps we're in a niche, but the "CGI framework" we have works well for us. Performance isn't an issue either.

If I was doing a brand new system from scratch, I'd seriously consider a good framework that made my life easier, but I've yet to find an article/something that communicates to me concrete whats & whys that the various frameworks have which will make my life significantly easier. I fully recognize that's most likely a failure on my part and why I ask for pointers to places that have good explanations. I also fully recognize that what I probably really need to see is not some simple tutorials, but a fully working medium-sized system based on a PSGI framework. One of these days I'll go look for such a beast on Github and see what I find and if that helps me understand.