This is awesome. I had a quite fun time allowing my eyes to play over the pages as if I was looking at them to book a flight and took the care to make a summary

This is awesome. I had a quite fun time allowing my eyes to play over the pages as if I was looking at them to book a flight and took the care to make a summary retrospective note of where my eyes had travelled. I'm normally completely unaware of where my eyes move on a page. Interestingly, I found that when I switched my view from the normal to the eye-quant diagram, it was immediately apparent that they'd got it harrowingly right. They really do seem to have the play of my attention pretty much down. Almost scary. This gets me thinking. I don't know what eyequant's business model is, but it occurs to me that they could do an amusing sideline in training web-designers accurately to anticipate where people will look when they see web-pages. I feel that if I did 300 normal/eyequant pairs in sequence, I could gradually become aware both of my eye-movements, and also of those that people would normally make. I'm reminded of the research that shows that the feedback loops available to doctors who predict disease from x-rays is too slow, so that often their skills deteriorate without them knowing as their career progresses. It was shown that if these doctors are given x-rays from the past to judge where the results are already known, so that they can then be given immediate feedback, their skills quickly recover and then improve. Feedback's obviously vital to learning. Graphic designers, even if their web-pages are being A/B tested the whole time, just don't get sufficiently speedy and dense feedback to ever gain anything but the most rudimentary skill in anticipating where people's attention will go. Eyequant seems to have this nailed. Guess that's what you get with Christoph Koch on your team. But it would be pretty awesome if the technology could be used to help make graphic designers (or anyone indeed) to become aware of the forces on their attention.