More than meets the eye.
About six months ago my indie game company Lunar Giant wrapped up its latest project with nerdcore rapper Mega Ran. The final product was a 2D platformer much like Mega Man, but with Metroid elements peppered in, rap music, and characters from the "Mega Ran" universe. The game was released in conjunction with an album by the artist.
Nevertheless, after three months of development, we had a final product. Here are some things we learned during the process, by way of a brief pros/cons list.
Selling the game
We ended up approaching our business model (if you can call it that) for the game by way or promoting donations, merchandise, etc. This didn't pan out at all, and Lunar Giant actually ended up taking a net loss on the game's production. Part of the problem with building HTML5 games right now is that the marketplaces, while they exist, are so ridiculously new and have such a small reach that it's really difficult to sell your product.
Technical limitations (audio, frame rate)
Other tools out there like Unity or Flash have huge communities built up around them. When you run into a problem in ImpactJS (or any other HTML5 engine, for that matter), yes, there's a community, but it's much smaller and the answers are much more sparse. Anybody who's ever tried to teach themselves how to code, or how to work with a new game engine, knows that this makes life much more difficult.
Extremely easy to develop for
Lunar Giant made MRTG in about three months. That's a pretty impressive turnaround time for a game with the kind of production value that I think we achieved. Why was it so easy? Part of the reason is that, because of the limitations we faced, we ended up cutting a lot of features that probably didn't belong in the game anyway.
One of the really great things that HTML5 gamedev promises is code portability. There's much less fragmentation of a game's codebase. Many developers end up hiring contractors to port their game from its original platform to a different market, and the hope behind HTML5 games is that a developer would have the ability to reuse something like 90 percent of their codebase when shifting to other platforms.
Everyone has access to a browser and a computer these days. Not everyone has an iPhone, or an Android device, or an Xbox. HTML5 games can be played by just about anybody with a computer.
Does the future of HTML5 game development look bright? I'm not sure. Many people think that it's a dead end not worth pursuing. Me? I think just like any new technology, it'll take some time for people to truly understand what it's meant to be used for, but once they do, it has a bright future.