Of the many dysfunctional, outdated aspects of the network TV business, the pilot process may be the most broken. Every year, dozens of very expensive pilots are produced in a short, identical window, with everyone fighting over the same tiny pool of actors, decisions being made in a rush based on limited data, often just on the gut instincts of a handful of people. Only a small handful will ever air, and only an even smaller handful of those will make it to a second season. It's an inefficient process in virtually every way.

Why do the network do it this way? Because, like so many other aspects of the business, this is how it's always been done, and it's hard to steer around this particular iceberg. The networks pay lip service to the idea of doing year-round development, for instance, to avoid the casting crunch, but it happens only in isolated cases.

One potential fix in the age of Hulu, iTunes, etc., would be to make all of the pilots available online for viewers to sample and offer feedback on. It's not an ideal solution — it would be a self-selecting sample that, by its very nature, would probably be more likely to watch shows online (where the networks don't make remotely as much money) than on TV — but it would still provide far more feedback than the networks get now, and possibly more useful feedback than the traditional network testing that inevitably give high marks to terrible shows featuring recognizable stars. But the networks can't or won't do that, because there are too many entities involved with too many egos to potentially bruise. Some pilots are so terrible they should never see the light of day, and no executive wants to be second-guessed if one of their pet shows gets lower marks than one they passed on.

Because Amazon hasn't spent decades making shows, it's not bound by tradition or unwritten rules. For some reason, Amazon chose to produce eight comedy pilots (six live-action, two animated, plus another six children's shows I won't be talking about here) at roughly the same time as the networks, and therefore had to draw on the same diluted talent pool. But once they were made, Amazon decided to open up the process to their potential audience, and crowd-source reaction to these pilots. You can watch all of them at AmazonOriginals.com, and then rate them and/or take a more detailed survey about what you liked and didn't like about each, which include:

* "Alpha House," about four Republican senators (John Goodman, Clark Johnson, Mark Consuelos and Matt Malloy) who live together in a large Washington, D.C. home; created by "Doonesbury" author Garry Trudeau;

* "Betas," about a social media start-up in Silicon Valley, starring Joe Dinicol (Nick the comedian from "The L.A. Complex") and Karan Soni ("Safety Not Guaranteed");

* "Browsers," a musical comedy set at a (very) thinly-disguised version of the Huffington Post (with Bebe Neuwirth as the Ariana Huffington stand-in), created by "Daily Show" veteran David Javerbaum, with songs co-written by Adam Schlesinger from Fountains of Wayne;

* "Dark Minions," an animated sci-fi comedy about two low-level functionaries in an evil intergalactic empire, created by and starring Kevin Sussman and John Ross Bowie (aka Stuart and Kripke from "The Big Bang Theory"), with supporting voices by Clancy Brown, Phil LaMarr and others;

* "Onion News Empire," the latest TV spin-off of The Onion, this time a "Police Squad"-esque parody of cable news (and dramas about cable news like "The Newsroom"), starring William Sadler, Jeffrey Tambor, Chris Masterson and Cheyenne Jackson;

* "Supanatural," an animated comedy about two divas (voiced by Jameeliah Garrett and the show's co-creator Lily Sparks) who battle demons and other evil creatures when they're not busy working at the mall;

*  "Those Who Can't," created by and starring Andre Overdahl, Adam Cayton-Holland and Benjamin Roy as three immature high school teachers;

and

* "Zombieland," a continuation of the movie, from the film's writers Rhett Reese and Paul Wernick, and starring Kirk Ward, Tyler Ross, Maiara Walsh and Izabela Vidovic in the roles played, respectively, by Woody Harrelson, Jesse Eisenberg, Emma Stone and Abigail Breslin.

Roy Price, the head of Amazon Studios, told TV Guide that the pilots will be up for about a month, and that judgments will be made on a variety of factors: how many times each one was viewed, how many times they were rated and how highly, the scores and comments from the survey, etc., in addition to the preferences of the Amazon executives. And he said that Amazon decided to stick to comedy for its first time out because it made more sense to specialize out of the gate.

But what's interesting about the choice is that comedy pilots are very hard to do well. I can probably count on one hand the number of genuinely great sitcom pilots in the last 25 years ("Arrested Development," "Malcolm in the Middle," "Modern Family," "NewsRadio" and "Friends," in some order), whereas many of the best comedies of that period had pilots ranging from mediocre ("30 Rock") to terrible ("The Office") and took many episodes (or, in the case of "Seinfeld," several years) to entirely find themselves. So Amazon is asking people to predict which series they might enjoy watching based on 25 minutes that could be wildly misrepresentative of what follows.

I've watched all 8, and the only one I saw very little potential in was "Zombieland," where the new actors mainly serve as reminders of just how important the original actors were to making these characters and this world entertaining. Maybe if the show is recast with the best actors available from the rejects of pilot season, it can become something, but this doesn't work.

But all of the others have either notable flaws or very obvious potential pitfalls should they go to series.

"Onion News Empire" is probably the most fully-formed of the pilots, and definitely the funniest of them, but unrelenting satire is very hard to pull off on a regular basis. ("Police Squad," for instance, lasted for six episodes, and at least two of them are lousy.) The "Browsers" songs are fun — and much more interesting than the paper-thin caricatures each actor is playing — but can Javerbaum, Schlesinger and company really produce multiple comedy songs to fill out 6 or 13 episodes a year? ("Flight of the Conchords" called it quits after two seasons, in large part because Bret and Jemaine had exhausted their entire back catalog, on top of the new songs written expressly for the show.)

"Alpha House" has the biggest star in Goodman (plus an amusing surprise cameo from an even bigger one), but it feels like Trudeau on cruise control, taking easy shots at the most obvious conservative targets.(*) "Dark Minions" has some amusing gags but is basically an extended bit from "Family Guy" or "Robot Chicken."

(*) When we discussed it on this week's podcast, Fienberg wondered why Trudeau hadn't made two of the characters Democrats, which would generate more conflict; "Doonesbury" is an obviously liberal series, but Trudeau has told plenty of jokes at the expense of Democrats in power, or of left-leaning civilians. It's not even about ideological fairness — and I'm reluctant to go much further into this, lest I violate my own No Politics rule — as about comedic wisdom. Opposites in conflict are funnier than a group of like-minded individuals.

What I do like is that most of them have very specific voices. I didn't love "Supanatural," for instance, but I appreciated what it was trying to do by putting these two women at the center of a world filled with evil. And "Those Who Can't" is evocative of "Workaholics" (which I find hit-or-miss, but which hits often enough for me to keep checking back in on it) while having its own sensibility. (The Benjamin Roy character — a tatted-up history teacher whose pretensions of rebellion and coolness couldn't possibly bore his students more — isn't one I've really seen before.)

So while none of them are great, or probably even good (even "Onion News Empire" is spotty, despite having the highest laugh percentage), many have the potential to be good down the road. The question is whether the average Amazon user will be any better at figuring that out than the average network development executive — and also whether Amazon's research team knows what it's looking for in the data.

For instance, at this writing "Zombieland" has more ratings than any of the others: more than 3 times as many as second place "Alpha House," and more than 9 times as many as last place "Supanatural." That's not surprising, as it's the one pre-existing brand name of the bunch. But is that any different from, say, NBC believing there was a lot of interest in "Emeril" (which, we were told over and over, tested very highly) because people loved Emeril Lagasse?

The survey has its limitations, as most of the mandatory multiple choice questions seem to be missing a middle option (in between, say, "I probably will watch" and "I probably will not watch"), but it also offers several open-ended question about what we liked and didn't, and also allows a good scale of ratings for individual components (the songs of "Browsers" vs. the characters, or the humor of "Onion News Network" vs. the premise).

We don't know exactly how much weight the viewer response will be given compared to what the Amazon executives want to do, but the process at least creates the illusion of participation in the process, and may in turn make people more enthusiastic for these shows than they otherwise might. If you're just a passive, traditional viewer, maybe you bail out more quickly on a show that's not entirely clicking. But if you think the show's future rests in your hands, maybe you stick around and look for signs of greatness so that one day you can claim part ownership of its success.

And as a guy who covers this business, I didn't love any of these pilots, but I'm going to pay attention to which ones survive the process, and what they become, simply because this is a method no one's really tried before. Will it be any more successful than the way the networks do it? I have no idea, but I have to think that any other approach is smarter than the traditional one.

What does everybody else think? Were there any Amazon pilots you particularly liked or hated? What kind of feedback, if any, did you provide? And do you care more about the outcome of this than you do about, say, all the network pilots in development that we keep reporting on the old-fashioned way?

Alan Sepinwall may be reached at sepinwall@hitfix.com