I had an interesting time with this and Fedora20 last night. This command
echo "199e 8101" | sudo tee /sys/bus/usb/drivers/uvcvideo/new_id did not work on my system. Not only is there no uvcvideo subdirectory, the system would not let root make one (no idea why). I wound up putting new_id in my home directory for the moment. The rest of the commands did work and my DMK21A04.as is up and running under linux. This does need to be posted prominently somewhere. This camera was an 8101 id. Thank you. This is the first success I have had in this long running quest.
I then went after the relabeled Celestron Neximage 5. This is made by IS, and is really a versatile camera. It has 2.2micron pixels and can go from 640x480 up to 2500x1600 or thereabouts. It does 60fps at the lower resolution, and has good quality color. Its product ID is 8207. When I put that number in through the above process I got confusing results but the camera did not work (It was more known to the system though, in that the error messaged no longer said "no /dev/video0".)
I think I found out that there is already a file on my computer that uses that number and am working on the resulting confusion. If I get this one working it would be great, and it needs to be posted somewhere as it is a reasonably priced high quality color camera.
On has to redo the camera's firmware to get it to work, though it is a pretty foolproof method. It save your old firmware as a file, and gives you directions on how to go back. So far the camera works, but does not have all of its resolutions available, and seems to be a bit slower than under its normal operation. A good day's work, but that is pretty much all I know for now. Both cameras are alive under uvc in linux.
Though since I'm thinking about it: Sometimes it shifts for a frame or two from where it should be, which I'm not sure if it's a v4l2 (I haven't observed it do it wit guvcview), long exposures, an INDI problem, a PHD2 problem (I don't think I've seen it using Kstars/Ekos), or a Mini-PC problem (I don't think I've seen it on my laptop, but I'm not certain of that). Like somewhere between 1/4 and 3/4 of the frame is off/wrapped around (and mostly out of guide star detection range, so the next frame or two it picks it up). Has anyone else seen that sort of issue with other tis cameras, or neximage cameras?