Lately, many photographers have begun to experiment with HD video, as most new DSLR cameras include video functionality. The quality possible from these cameras is outstanding, in no small part due to the high-quality lenses most photographers already use, which will greatly exceed the quality of the lenses used by typical lower-budget videographers.
I've played around with video a bit, first with a waterproof P&S that does 720p (though not at particularly great quality) and then an E-P2, which is pretty good for video with the Panasonic f/1.7 lens.
I was particularly happy with this video from the P&S (Panasonic DMC-TS1) - it's 45 minutes or so of footage taken with the camera strapped to the front of a motorbike, sped up to fit into a minute and a half.
My main camera is a 40D and I don't feel the need to upgrade for any other reason anytime soon, so the E-P2 is it for video for me at the moment, unless I'm doing something involving a lot of water or the possibility of the camera getting knocked around (such as driving around on a motorbike in the rain). I have done some time-lapse stuff with the 40D, but nothing particularly notable. As an aside, I think that's a good way for photographers to get into experimenting with video - it's much more similar to planning a photograph than planning a complex video, so it might make more sense for photographers.
Video is fun, but I find that to do anything compelling takes a lot more planning and "doing" than taking photographs. You can't take a compelling travel video, for example, unless you spend a *lot* of time taking video and planning shots, whereas you could be more casual taking photos while traveling and still come up with a compelling slideshow (assuming you're a decent photographer in the first place).
I took a bunch of random videos the last two times I went to Thailand with the E-P2. Some neat stuff that I was planning to edit down into a short compilation of different clips that show different aspects of Thailand. You can imagine the type of thing I mean, you've probably seen something similar before.
But it's incredibly difficult to do that. You'd need hours and hours of random footage to be able to edit it into something interesting (even just a couple minutes' worth). So most of my footage is sitting on a hard drive, unseen by anyone.
My most compelling two videos happened because I decided to take the time to try to document things fully so that the narrative of the edited down film would make sense. Doing that involved neglecting my girlfriend (and almost letting her fall into a river while I was recording), so while it yielded my most interesting film, the experience discouraged me from doing it again.
This is the video in question, including the part where she almost falls in :)
This is the other one that I sort-of planned; the subject is more interesting but the execution was more off-the-cuff and didn't work as well.
So, my long-winded point is that it takes more than idle experimentation to do anything interesting with video, while idle experimentation with photography can often yield good results. You need to have a pretty clear vision ahead of time of what you want your video to be, and take a lot of time to plan so that it works properly.
And ultimately, flashy editing can't replace well-planned and nicely-captured footage. My videos are minimally processed (contrast added sometimes, maybe a couple other small tweaks), but they lack interest not because of that :) You'll definitely find that getting the exposure and other settings right in-camera is much, much more important with video than with photography. It's an order of magnitude harder to fix this kind of thing in post for video than it is with a RAW still image. There's also a much larger leeway... for example I wouldn't worry too much about over-exposing and blowing things out, unless you've got a big setup with video lighting where you can better control such things - at which point you're no longer just experimenting :)
