Love your thought & exploration!! Kudos
However, predicting user behavior is tricky stuff, and it's better to not do it at all than do it wrong.
I've seen a couple executions of similar concepts where sites will try to predict when a visitor is about to leave and throw up a popup as a last attempt to grab conversions (http://www.exitgrabber.com/).
Good in theory but hella annoying when you move your cursor around the page and keep getting popups...
Perhaps this is a band-aid to bad UX? Maybe the targets could just be closer together?
Maybe the targets could just be closer together?
That’s the approach I like to take.
I really like that this is being discussed, and there’s lots of great thoughts in the article.
I prefer just fixing the UI though, rather than trying to predict user actions. A path analysis is a good idea for common tasks. It should probably include:
- Number of events (moves, clicks, drags, etc).
- Distance travelled.
- Size of targets.
- If the task requires switching input methods (high cognitive load for doing so).
When you’re done, you can lay them out in straight lines and compare different UI versions. Makes it very clear which is easier, quicker and with the least cognitive load. I really wish more people would do this.
Interesting approach for designing better UX using mouse position and move!
I'd like to thank you for all contributions on the article Rıza. You've pointed out some nice spots.
Nice idea, it might start to annoy me if unintended menu items kept popping up.
This looks like a solution looking for a problem.
I can see where you're coming from with that but I disagree. There doesn't have to be a problem per se for you to want to make something better.
This is an attempt at speeding up UX. I don't know if this iteration is the right one, but I think its a nice idea for trying to make things better.
Yeah, it has some issues – but for the most part it's damn smart step in solving a problem, the inertia created by the mouse speed is inspired.
The biggest problem I foresee is... us. Designers. An interaction like this would really need to be used in moderation, where it's totally applicable. I fear if we had this power at our finger tips it would be on ALL OF THE THINGs. Making you feel a little bit like a drunk Harry Potter when trying to navigate a website.
I think it would be really interesting to run some user testing with this when it is in a little more solid shape. I wonder if the user would even notice the enhanced functionality, if it would enhance their experience, or if there would be false positives that would negatively impact their experience.
I like the idea of predicting desires and providing the appropriate state changes before the user actually requests them. I could easily see this being used with complex sites to prefetch data so menus and additional articles could load faster.
This is a pretty interesting concept.
A couple of things that came to mind: Would the user be confused as to where the trigger occurred? Would they mistakenly try to re-create the action as the way to find the drop down instead of knowing that this hinting is in place? It would be a bad experience if they unknowingly keep triggering something unintentionally.
Would they continue to move the cursor all the way to the button anyway? If so, is this really worth the trouble?
Always good to see something different though, thanks for sharing!
I think this is a really nice idea. In playing with the demos I don't think it works very well though. It doesn't feel fluid or natural.
Thanks everyone for sharing your ideas.
Predicting where user's going to click is a really challenging work, I'm fully aware of this. This needs a high amount of tracking data of user interaction with an element and requires more detailed formula and algorithm as well. This is the reason I call it an idea. The only purpose of the plugin is to show that is doable somehow. The plugin currently has both false positives and false negatives but I think it will do better in feature.
I'm impressed. The examples look like they could use a little more work, but it's a great idea.
How does this differ from the plugin linked at the bottom in the Amazon example: https://github.com/kamens/jQuery-menu-aim ?
It only works for a drop down which uses a triangle to check if mouse is inside it.