See "A Nice Morning Drive", by Richard Foster.

I've just had a chat with someone else working in my lab who pointed out that beyond my problems with this, the projected image itself would appear to jump erratically around. This would be for the exact same reason that we usually don't notice our eye movements (i.e. stabilisation in the brain factoring them out).

He's talking about tracking the movement to simulate it not being attached to the eyeball.

You're right, you'd get fading for any perfectly fixed image. You could always modulate it to avoid that though.

The problem I find with doing as you say (simulating that the text display is at a position in space) is that next you might want some way to turn it on and off. Maybe a hand gesture? And then a way to manipulate the text? More hand gestures? Speech recognition? If it has to do significant processing you're going to need some external hardware. At what point are you basically simulating picking up a smartphone (for no benefit)?

Once you've done that it's functionally no different to installing the tech into a wearable headset. Putting it into a pair of glasses would actually be a lot simpler because you never have to factor out the eye movements, comfort and safety are less problematic, and you have more space to work with. It's possible that the lenses for focusing the image at a close distance might not work when they're not fixed to the eye's position though.

Actually, the problem with the method you lay out is that I'm pretty sure you would have to have the lenses for focusing the display over the centre of your vision (though I don't know much about ocular optics, I study the brain side of things). In that case you would need the technology to be tiny if it wasn't going to obscure the most informative area of your visual field when you weren't actively using it.

You can't read text in your peripheral vision. The best they could hope for would be sticking rapid serial visual presentation (RSVP) text in the fovea (i.e. flashing up a rapid sequence of words right in the centre of the visual field). This could work, but it's hard to see why anybody would want it. You wouldn't be able to multi-task, because the text would be in the way. You wouldn't be able to access the text in a non-serial fashion either, which removes any advantage over having it presented in audio form.

Searle's Chinese room argument applies.

Scientists have been after a larger e-ink display for reading journal articles on for a while. It doesn't have to be colour, it just needs to be big enough for an A4 .pdf to be readable on it (and the figures too!).

I'd recommend that the OP join this list:
http://visionscience.com/mail/cvnet/cvnet.info.html

You could try getting a message posted out to people who work in perceptual psychology and visual neuroscience to see what they have to say about your condition. Some of them might even be local. Be careful not to come across as too crazy though!

How would you test the Wi-Fi sensitivity that you describe with fMRI?

Would you say that about people with idiopathic Parkinson's disease? That word doesn't mean what you think it does (even though you link to an article explaining it).

I guess for Best Buy the important thing would be that knocking that delay down to 200ms makes the PC "just feel that little bit faster". For a customer who's just bought a new computer, that's probably reassuring.

Also: to refer to the sections that mention gaming as only being a small proportion of the whole book is a bit disingenuous given that most of the rest of the manifesto is copied from elsewhere (Wikipedia, other books, the Unabomber's manifesto...).

How does this square with the argument the writer of this piece is making? They quote that section and then avoid addressing it.

There are two meanings of "neural network" here. There is the biological one that the brain satisfies and the computer science one that it has not been proven to. You can't treat them as if they are equivalent.