Forum Posts Feed
Organizing lots of thought snippets
< Next Topic | Back to topic list | Previous Topic >
Posted by Prion
Jan 26, 2015 at 03:54 PM
Hi Paul
it’s all good, I wasn’t referring to you personally. Besides, it was just *my* opinion because I sometimes find it surprising how some articles stay around for ages while some others are forgotten almost instantly. BJ recommends splitting everything into chunks of 500 or 1000 words (don’t remember precisely) in order to aid the artificial intelligence inside Devonthink. Even if it was automated, I wonder who would actually do this? To be fair, I always thought it was suggested more like a thought experiment rather than an actual practical advice.
Prion
Posted by Chris Murtland
Jan 26, 2015 at 04:50 PM
I agree that a single outline hierarchy can become unwieldy. However, I still prefer Workflowy for this because of the power of filtering the outline at a granular level. If I have multiple lists of thought snippets, each paragraph or even word (in single-word lists) becomes its own independent item. I find that this does facilitate combinatory thinking if you’re willing to think about the search terms a bit. I’ve gotten in the habit of searching for almost any random word or phrase that comes to mind just to see what thoughts and collected snippets I may have on the topic, however vague the topic may be. I suppose that’s the power of keeping everything in one application, not just Workflowy, but I think the granularity can help prevent overwhelm; plus, I prefer seeing the actual, combined contents based on my search rather than just getting a list of each item with hits that I then have to go through one by one.
Also, tagging at the paragraph level is another way to “collect” seemingly unrelated snippets into some conceptual grouping, regardless of outline location.
And if you hoist before doing either of the above you can then work only within a certain region of your outline if needed.
Posted by Hugh
Jan 26, 2015 at 05:04 PM
I’ve probably evangelised SBJ methods in the past, and I don’t apologise for that. His composing method of a few years ago - in my mind summarised as “First, develop your outline; second, drag carefully selected pieces of evidence/research into your outline at the appropriate nodes; third, write your essay/paper/article around the pieces of evidence/research” - has always seemed to me to be a useful response to the not-infrequent DT or Scrivener forum question “How do I compose a factual article/essay using DevonThink/Scrivener/both?”.
But I must admit that “mini-chunking” has never been part of my credo.
Posted by Dr Andus
Jan 26, 2015 at 09:52 PM
Hugh wrote:
>summarised as “First, develop your outline; second, drag carefully
>selected pieces of evidence/research into your outline at the
>appropriate nodes; third, write your essay/paper/article around the
>pieces of evidence/research”
These steps seem more relevant for the final stage of writing up. However, I think the OP’s question concerns the processes of organisation, analysis and synthesis that precede this final stage. How to “carefully select pieces of evidence/research”? This is what hierarchical outliners are not equipped to do (indeed, they are entirely the wrong tool for the job, at least for large data sets).
Hence the need for a Zettelkasten type tool that allows one to gather stuff (and resist the temptation to start sticking it into a fixed position within a hierarchy too soon), rearrange in many different ways, discover new relationships between data (based on some initial tagging), annotate them further, and then only at the very end start doing the outlining for a piece of writing (well, OK, interim writing can also be helpful, but it’s not outline-driven).
Posted by Prion
Jan 29, 2015 at 04:48 PM
I am correcting one detail of what I previously said: The “find similar notes” feature has returned to Tinderbox in the meantime but is located in a very different place (the “get info” popover).
Prion wrote:
>Or more precisely: Tinderbox used to do that, it seems that the feature
>disappeared during the transition to version 6 and still has not
>reappeared. It will hopefully do so but for anyone looking for this
>feature now, Tbx is not the way to go. Or perhaps I am overlooking it?
>
>