Problems: Self-Directed Learning.
Online Reading Time and Topic Tracker
A browser plugin that tracks my online behaviour and provides visual graphs on what I'm doing.
A plugin that tracks my eye ball movement, time spent reading an article, parts of text I selected, etc. This data is then presented as a visual graph showing topics I read about, etc.
Hey [Justin], what else do you think it's worth tracking to measure what people learn when they browse the web?
[ruta], good idea, which certainly is doable. So many statisticians, computer scientists and data scientists working on computer vision. What you need, is a higher resolution cameras on computers, or existing solutions (tobii) to use your eyeballs as mouse.
[ruta], btw., be more specific in the naming, because this 'Online Behaviour Tracker' is very general, could refer to much more than eye-tracking. Maybe 'Online Reading Time and Topic Tracker'.
-- this is a project then, not an idea anymore. no?
[ruta], originally, the difference between a project and an idea is that a project concerns specific group of people, using specific tools, trying to deliver specific deliverable. Just reducing the generality of an idea does not make it into a project. Adding concrete people and tools and scope does. Idea can be very specific and detailed, like a patent, but it is not a project without specific time and space, people, tools, scope in real world. :)
[Mindey], okay, got it! An Idea becomes a plan when it has timeline, people, tools, etc. Thanks!
[Mindey], how about heatmaps, to get the data on how I interact with the page?
[ruta], heatmaps are only a way to display the click density,... Your problem is to collect data points about the eyeballs first, and infer the actual words and topics read from that data, - the mouse pointer data, and other data.
[Mindey], if we think about how people read online, we'll see that many of us highlight text while reading. My assumption is, when we highlight text, we reflect on what we are reading. I use various web clippers to save content for later (i.e Evernote). I also use Rescuetime to track my web browsing behaviour. Maybe there are simpler ways to solve this problem instead of tracking eye balls?
[ruta], very good observation. Indeed, selections during the reading could be used as data source, and a browser plugin could be made to collect that information on all websites you visit. Statistical models could be constructed to filter that information, but it would be much more accurate to have eyeball trackers, like the mentioned (tobii).