I write notes about various stuff. But is text the correct format.
Example: I had some trouble learning http cookies. Now, in my backend nodes, had I just had a diagram with client and server blobs having incoming and outgoing arrows showing req, res and cookie being a part in these.
Then atleast discovery of notes would have been easier.
In other words, text isn't intuitive for browsing stuff. Since thoughts are not linear and text is.
Now, should I create a system where there are two worlds. - the picture/diagram world and the text world. And create some software that can relate the two. In other words if I double click the cookie part in the diagram, it takes me to the notes for it.
But I wish to do this automatically (instead of adding a link to the cookie box in the diagram.
But more importantly, won't a robot need this mapping too? We humans are very good at this mapping of situations - static picture, videos, music into words/text. But a machine isn't.
Make a subsystem for self for picture, note storage and mapping.
Find out if an ML model needs such a system. Or general inference can already successfully converge to meaningful text.
If not point 2, Make a general system or a ML data structure that can be used by AI.
I write notes about various stuff. But is text the correct format.
Example: I had some trouble learning http cookies. Now, in my backend nodes, had I just had a diagram with client and server blobs having incoming and outgoing arrows showing req, res and cookie being a part in these.
Then atleast discovery of notes would have been easier.
In other words, text isn't intuitive for browsing stuff. Since thoughts are not linear and text is.
Now, should I create a system where there are two worlds. - the picture/diagram world and the text world. And create some software that can relate the two. In other words if I double click the cookie part in the diagram, it takes me to the notes for it.
But I wish to do this automatically (instead of adding a link to the cookie box in the diagram.
But more importantly, won't a robot need this mapping too? We humans are very good at this mapping of situations - static picture, videos, music into words/text. But a machine isn't.