Thoughts on Context and Conversation for Ideas
How to browser, ingest content and then to then reproduce something useful is a thing I find myself pondering an enormous amount. There is certainly the mindless browsing like activity and on the other side of the spectrum there is the focussed note-taking + research of trying to answer a limited specific question.
But there exists a massive spectrum in between. Spectrum that includes many forms of knowledge building not easily explained, but only in retrospective. Connecting backwards the dots of what you were ingesting.
This semi-tacit approach of reading a bit of this book, following certain twitter topics, watching certain YouTube videos, asking a few questions on Reddit and then checking out code on GitHub seems to on the face of it fall into the bracket of mindless browsing. But there feels a qualitative difference. For one, there is, once one thinks about it, usually a many through lines. Perhaps it's material on deep learning techniques coupled with watching of something on metal fatigue and manufacturing difficulties with ideas on learning curves and cost decreases.
It looks like a big jumble, and in part it maybe, but there are, in my experience, ideas, notions and new creations embedded in this mental latent space of interests.
Right now the best means of exploring that latent space, that I have found. Has been conversation, in conversation, with an equally interested partner, one can flow from one topic to the next. Whereby at a certain point certain ideas bubble up. Still nascent, and much work to do, but the spark has been lit. You can connect two or three different topics, you have new solution to your problem, new paper, new article and new tool to build. And what it took was the exploratory give and take of conversation.
So my intuition tells me there is something here, unexplored and not yet captured by our current crop of tools, and it's not managed extensively either. Like some complex note-taking system. It emerges from conversation without any prior demarcations or structure.
If tool could be this partner, a sort of computational conversation buddy, constantly connecting and remixing in the background for you. Tool then in which it can bring to the front seemingly hidden concepts, but not so hidden to be unknown to you. Since you need a base level familiarity, then you would have something rather powerful.
Something that would push the speed at which many ideas are formed and push the rate at which new ideas are discovered. We shouldn't all have to wait for the most opportune conversation, for many the wait maybe too long, and the time between too infrequent.
