Tobias, your ending comment that
I think we're not looking at different things at all. You have a "related > list", "train of thoughts", whatever you wish to call it, which you may > want to expand on later / individually, that you seem to also want to sort. > The only little thing I added was to be able to also assign a "type" to > each... so as to be able to quickly and easily further qualify what each > related item is. > ...sums it up. Yes, we're talking two very close steps that really would complement eachothers! However, in discussing here my idea has further evolved in one aspect; to treat each tiddler as a startign point for a train of thoughts. I'll explain below what I mean with this. A key aspect is that it is an automatically generated list that already contains "intelligence". One can, with your ideas, refine it to make it more useful: Your question below serves as a starting point to clarify the idea: In which way is "capturing the train-of-thought sequience" possibly not > "manually creating" one? I prefer not to throw a bunch of stuff on a pile > to only later see if I can sort that bucket full of stuff. I'd preferably > create things at the right place to begin with, if only some plain, sorted, > flat, related-list that I can filter. > Take the sidebar list *Recent*; Its sequence is not random like stuff thrown into a pile. It is incidental, "time ordered", which captures some interesting aspects; for instance, the 'recent' tiddlers are of different importance than the 'past' in this list. And this list lets us reminisce on how we at points in time were dealing with particular areas of concern as they are somewhat grouped togheter in time. The aim in my proposal is to take advantage of this *naturally* occuring and topical grouping. Perhaps the term "area of concern" is more appropriate than topic. Such areas can typically not be *pre*defined when it comes to thinking. The area(s) of a stream of thoughts is not clear until the stream is over, if even then. And the number of areas of concerns is of course unlimited. Compare it perhaps to writing a newspaper. Sure you can have defined sections in the newspapers but you cannot predefine the headlines for the articles. So my proposal is (now) to treat *each* tiddler as a potential starting point, essentially *creating* an area of concern from there on. In a way this is how to *identify* topics. As our brains stream thoughts, a tiddler can serve as a point in time and a point in the stream of thoughts from where to record the steps. Here <https://www.youtube.com/watch?v=wYlVBhwfvL4> (starting at 30 seconds in) is an example of this process captured. My analogy refers to the time period when Sherlock first created this in his head (i.e not actually when he recapitulates it in the car). It is a stream of thoughts and the incidental sequence of the thoughts is crucial, not just the conclusions. One thing lead to another. The documented sequence allows backtracking. It would be meaningless to put a predefined topic on a process that is just being created because it develops in unpredictable directons as new facts or thoughts come in. If you did pre-type it, the type would have to be at a meta level and not really describing the content. ("This will be an *article*" vs "This article is about a cat that got stuck in a tree...") In practice this would mean a local type of *Recent* listing generated for each tiddler. It could sit hidden e.g in a popout, and it could be limited to a certain number of entries. Opening it would allow for the quick categorization you describe. I think one challenge would be to automatically capture the relevant titles in this list. I proposed any internal link clicked on and any child created. This might not be encompassing enough. Maybe one should add the titles for tiddlers separately opened? The propblem is of course that we don't know to which area of concern that it belongs to, out of the ones active in parallel (i.e the problem the idea aims to solve with to begin with). Or maybe, in an active tiddlers Train-of-though list, adding any other tiddler that is opened within the nex X minutes, thus taking advantage of a time proximity as an assumption of topic relation? Regardless, the resulting list should be easy to work in the way your idea describes, kind of resulting in two parallel lists - one, more temporary, illustrating the time sequence (like the Sherlock clip) for an area of concern and which serves as potential food for the other list (list? tree?) which is refined permanent. Mockups needed! :-) <:-) -- You received this message because you are subscribed to the Google Groups "TiddlyWiki" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/tiddlywiki. For more options, visit https://groups.google.com/d/optout.
