Hi R

Gotta hand it to you: This is fun!

Theo

For anyone who's interested, I just posted v0.2 of LeapCursor.js - which adds mouse-enter, mouse-leave, and click events using LeapTrainer.

Thanks!
R

Hi!

In this version the mouse click events are triggered with really any sudden movement of the hand - so it's actually a swipe right kind of movement, but the recognition threshold is very low (in fact, I've come to the conclusion that the threshold too low by default in LeapTrainer.js). It's a fair point though - for the next version I'll add a video showing the gestures. The mouse enter and leave events are just when the cursor moves over elements.

In fact, I've been upgrading this for use in the 3D graphing stuff (including having 2 cursors on screen at the same time, which is quite interestingly expressive, I think) - so I'll post an improved version once I port the changes back into the main leapcursor repository.

The other thing I'm working on with this is moving between 2D (floating around above the apparent surface of the page, like in those demo videos) and 3D, seamlessly. So you're moving along the page, and then into a 3D area - combined with a zoom motion, to pull you right inside the 3D space. This would be used to move between the kind of 2D mosaic of tiles in the semantic search system inside a 3D graph tile.

At least, that's the theory - right now what I actually have is a mechanism to generate WebGL errors at enormous speeds : )

Thanks!
R

Well, that's certainly true - there's currently no competing with the near pixel-perfection of a mouse pointer. Though, equally, all our software was built with existing input devices as an a priori assumption - which, inevitably, has limited the range of designs.

If ubiquitous computing is indeed becoming ever more a reality (and it seems to me that it is - though I may be wrong!) - then being shackled to a desk for most serious computing needs (with all respect to those people who can manage to do serious work on a phone or tablet - I certainly can't) - is not really inevitable or convenient.

Ever higher resolution displays and their falling unit costs - OLED technologies, flexible displays, printable diodes - combined with voice and motion technologies - seem to promise an environment in which mice are still great, sure, when I have one on me and I'm sitting at a desk - but not the only way to go to work.

This is why I'm - personally - interested in motion, and also in new approaches to interface layouts and new kinds of interfaces - since it seems (to me) like the days of the strictly rectangular, relatively low resolution, window into the web are limited - sooner or later we may want to summon up content on all sorts of surfaces - very large or very small - in any number of configurations and combinations - and perhaps not so much in fully packaged applications - but rather collections of micro-functions grouped together in sets that are useful for me, right now.

Technology on the web seems to be going this way - from the enormous, creaking J2EE-style super-frameworks of the late 90s that wanted to cover all our possible needs and also serve us breakfast, to the endless pluggable micro-frameworks these days that we can pick up and slot into place in whatever - and then drop and replace tomorrow when the next thing comes along.. So why shouldn't the applications themselves follow this pattern (in fact, this kind of competition surely breeds better systems).

Man.. I shouldn't even start writing - next thing I know it's 20 minutes later and my coffee is cold - apologies for the verbosity :)