@ozzyxs1a Awesome, but how does this work? Is it one toolcall for every brushstroke and does it have to process an entire screenshot after every stroke to plan the next one?
@cremieuxrecueil But isn't face recognition a solved problem by now? Don't we just need more of the right kind of data to solve mobility? It seems to me that whether humans do x subconsciously or consciously doesn't matter at all.
@nicol3a It is an advanced feature, yes. The cool thing about storing the data in text files is that you could easily use it in third party software, so there is no need to implement it here, it can still stay as “lightweight“ as possible.
@nicol3a For example, lets say I start spending time on something and type "start *activity*" and then "end *activity*" when I'm done, having a graphic (like a daily calendar view) where there is a line between these two points in time would be super useful.
@nicol3a I would love to try this out as soon as possible. I think it could be great especially for tracking purposes. The more you jot down the more confusing the text file will get, so I would assume that adding an option to untangle all that data visually will make a huge difference.