We're actually planning/hoping to release our next game,
Liferaft, as shareware in flash. We're not thrilled with the sponsorship model so far mostly because it breeds such a disconnect with us and the customer/players. We'd rather have a direct relationship with the people that play our game for a number of reasons...
The main idea behind shareware for us is to leverage the power of the portals by offering up a portion of the game for free with a link back to our site to purchase the rest of the game, probably for like $5 or something. There's been some concern as far as the community on these portals rejecting attempts like: "OMG I have to pay?!!?! 0/5 this suxxxz!!!!!" and that may be the vocal minority but I don't think it will effect the success of the game _that_ much. We're not sure though, it's certainly an experiment. Frankly, flash is a great way for people to hear about our game and play it with little to no barrier to entry. If we get 100,000 plays across all portals and our conversion rate is something like 1% who buy the game then we'll make 5K on the game. We might have been able to make more on a sponsorship but 100K plays is hopefully on the low end... Or maybe the conversion rate is too low, or perhaps... too high. :-/
Other games like Fantastic Contraption have had a great deal of success with a modified shareware model in that they offer the level editor as a paid option rather than more content in the game. These sandbox games tend to be more of a catchall in terms of audience so having a paid level editor that can save levels for the hard-core side of it makes sense and ends up being mostly hidden from the majority of casual players who probably never really care anyway.
For Liferaft that won't be the case. You'll get to a certain point in the game and then we'll show a "pay $5 to get the rest of the game" has a much greater chance of frustrating a large portion of our players. But oh well, we're going to try it and we'll be sharing our experience on our blog and so on. It comes down to the fact that we are making the game we want to make and while a sandboxy type game with level editing/sharing may be the better, more tested route, it's not what we want to do.
On Flash...You see the big problem with Flash in our eyes is that it doesn't make any monetary sense to spend more than a week on a flash game given all of the current models. Sure there are exceptions but this is generally true. Easily my most successful Flash game was made in like 4 days with Jiggmin, Effing Hail. It was fun to work on and I love the game and am proud of it and all, but sometimes we get the itch to sink into a game for awhile and make something special. When you start spending 2-3 months on a Flash game it really ends up losing you money really quickly. Though... working on a game for a couple months is fun and allows you to make a much better, deeper game. It's fun to make quickies but we want to make something bigger and the current Flash models don't support this kind of thing. So... this whole Liferaft experiment is really our last venture out into the world of Flash and what it can do for this size game. If shareware can give us a decent return on our months of investment then it makes sense, but I think for that to happen you've gotta think outside the box with Flash. What the portals and FGL have setup for you is a very narrow-minded business model that works really well for churning out games and even better for the portals themselves. The devs get the short end of the stick.
Of course there's plenty changing. Microtransactions and all that jazz can really benefit your game and are "thinking outside the box" as far as Flash is concerned. Again though, like with Fantastic Contraption, MicroTX games are very specific and either your game can fit really well into that model or you'll need to modify serious portions of your game to accommodate that system.