DGT Board support?

I have signed to lichess a couple of days ago and I (sadly) don't miss FICS at all.
There's still one thing that prevents me to switch completely to lichess: my DGT board.
With FICS interfaces like babaschess it is possible to integrate the DGT driver and use the board while playing over the server. I wonder if is possible here to integrate somehow the DGT driver or anything similar.
Thanks to those who will answer
Regards
Claudio

I am not a developer here, but I guess it should be doable. Should be some kind of client side app (Java Applet or something) to access serial port.

BTW forgot to paste the link I googled... chessprogramming.wikispaces.com/DGT+Board

Sorry for the trolling! This idea got me intrigued so I cant just stop thinking about it. I did bit more googling... Personally, I'd go with stand alone Java app with native libraries compiled for win32 and linux. Since you would use app just to hook DGT to lichess, it should just be a matter of handling the communication between those two.

It would indeed have to be a native application, as there is no other way to access the driver. It's also worrying how the technology is patented, which might cause legal issues.

Lichess actually has an alternate solution for using a real board, though it does not work particularly well. See github.com/turkotron for details. Basically, it uses a phone camera and QR codes on the back sides of pieces on a transparent glass board.

DGT chessboards are really expensive :-(. If I had one, I'd volunteer to implement client side app for lichess. Whomever would implement it, needs a real board (or some kind of board emulator).

Anyway, if you are using a camera, you need not QR codes. A friend of mine came up with this idea some time ago. You can test putting different signs on top of the pieces and putting camera above the board. Hint: ink needs not be visible for this (IR visible), phone camera should see it.

If that works, then only things one would need is ink (or stickers) and phone mount/holder.

Actually, it was discussed at length well before Thibault and his company went and did it. I came up with a similar idea. I proposed putting the codes on TOP of the pieces, so you wouldn't need a transparent board (and lines can potentially throw off recognition).

I must say, I wasn't aware of this invisible IR ink stuff before!

But for IR ink to appear, wouldn't the phone need to have an IR lens mounted on it? I've seen things like that before, like in this video:
www.youtube.com/watch?v=8Vc-69M-UWk (heatcams pick up infrared concentrations)

If you do indeed need a special lens, it wouldn't meet the conditions that were drawn up for the Turkotron. It should be (a) very cheap and (b) have easily accessible equipment so that even skint teenagers can do it. What would you say about the above concerns then?

Mobile phone camera (most) can see infrared light coming out of a TV remote controller (www.youtube.com/watch?v=qpY6670zhXo, http://www.instructables.com/id/Poor-Mans-Cell-Phone-IR-Filter/).

Next thing is to find what ink to use that camera can see, but naked eye can not. I know a chemical engineer who knows a great deal about this, so I will ask a few questions.

In the meantime I suggest googling, as someone most definitely has tried it.

I am thinking something like this:
1. Test if IR ink works with phone camera.

2. Make IR stickers to be placed on top of chess pieces (using simple shapes).

2. Take what is working already and add support recognizing pieces based on those shapes, instead of QR codes (http://opencv.org/platforms/android.html ?).

3. If that works, then produce enough of those stickers and sell them on lichess for low price, to ensure algorithms would work for everyone.

If the pieces look different enough, the camera has a high-enough resolution, and the scene is well-lit, it may just work with "simple" computer vision tools.

In fact, these students have successfully done this, it seems:

http://people.etf.unsa.ba/~esokic/mypapers/2008ISSPIT.pdf