I've only got a few minutes so I'll try and answer these quickly. I apologize in advance if my answers appear brief:

• Perl's never been about that -- I have several hundred dollars worth of Perl books on my bookshelf that disagree with you. I took a class with a very popular Perl training outfit for $100 (yes, a single class) at a Perl conference. I would personally gladly pay more if I could learn things like Perl internals, better faster ways of profiling/debugging; etc. etc. Maybe it is not what you are used to, but, like anything in computing, there is some money involved. Remember, in free software/open source money is not the evil -- dousing people's freedoms is.
• "...I think I'd also expect some sort of certification and ranking of some sort ..." -- I agree with you, and this proposal takes that into consideration. I'm proposing a system that allows for every class to be audited by anyone in the same way you can audit normal brick-and-mortar university classes except no one will boot you after a certain number of weeks. Here you can complete the course for the same low price as the actual participant (perhaps less) but your assignments will not be graded and quesitons will probably go down in the priority order. 'Grades' should look more like pass/fail with each class grading on a point scale which accumulates points for various aspects of things we equate to good Perl programming like style, terseness, creativity, readability, and, of course, functionality of the class project. Grades should be conferred for projects and not for question/answers since being good at answering questions does not necessarily translate into requisite skill.
• "...to verify that I'm getting the materials and staying current..." - I would audit the course before I take it or ask a friend. The courses could have publicized student rankings as well.
• "...How do you add value from a profitability perspective..." - On the part of the participant they get value by learning a new skill; from the university's perspective it gets value by producing students that are proficient and produce quality things for others to use which leads more people back to the university. it is all kind of like a cycle.
• "...profiting off of modules you didn't create..." - I can't imagine a CPAN author would be unhappy to learn that one of his/her modules was being used in a class somewhere in a university or referenced in a book. In both cases the author will likely not be compensated but will benefit because the popularity of their work will be enhanced. When I hear the name 'Lincoln Stein' I am inclined to want to know if he will be speaking somewhere or if he has written a book/article/course. I don't known Lincoln from anywhere other than the modules he has written.
• "...ignored by the "academics" of this university..." - There's always that risk of a school being pompous and only exhibiting certain works, this is true and happens quite a bit. Think of H.S. English classes -- you were not reading authors who were interesting but rather authors that the academics found interesting. No wonder so many people hate to read. I think here, thanks to crowdsourcing and other aspects of the web, we can be much more proactive and weed out classes and modules that are by a chosen few in exchange for those that meet current and future needs of actual students. It all comes down to transparency in the course selection process as well as honest feedback from participants. If there is a better module out there to utilize in a class than students should bring it up, discuss it, and get the attention of the instructors and those who oversee them. My plan is to have a tiered evaluation system for grading anyway so this would undoubtedly come up.
• "...Also who's your target audience..." - I'll have to outline that in a different post but thus far it is looking like 2nd year H.S. students to the adult education crowd which can be roughly segmented between those with a technical education and those without. Those with advanced education/experience in programming and CS topics would have to be advanced to a graduate-style, semi-independent research type of program which I am also envisioning.
• "...It'd be fun to go at Perl from those angles too, but in many cases those courses don't focus on the language as much as the computer science concept..." -- Oddly enough I spent some time today thinking of ways to provide classes to high schoolers who have at least completed some basic computing skills course, are saavy at working with social networking tools, and want to expand upon this. Injecting Perl along with necessary computing topics is necessary for a well-rounded class unless you already have this knowledge gained from elsewhere. Younger students are easily impressed and you only get so many chances to educate them before they become discouraged and give up math/science altogether for a career elsewhere. I was cooking up a course specifically geared for younger students today that incorporates social networking along with elementary data structures, file storage, and other necessary goodies. I'll try to post that when it is ready.

Thanks for your reponses!