“I’m not writing shit today”
And that was the plan. But my head is a fucking mess and I only know one way to sort that out, so here we are.
Some development project brainstorming I think. We’ll see.
Every few months I pull down some mind-mapping software and try to use it for a brainstorming session to dump the ideas out of my head and on to a page of some kind.
It’s always a disaster. Mindmaps, especially digital ones are fundamentally tree structures; nodes extending out from a single central point that have children, grandchildren, ad infinitum. It’s REALLY got the same topology as a simple outline, just without any implied sequencing within a tier.
I don’t know, maybe people think that way. But after the first two layers I’ve got this problem where I want to join nodes across the child branches of the tree.
I want a piece of software that can generate and navigate a more generic digraph. Problem is…I’m not sure something like that even exists. Sure I could go to yEd (something like visio) and just create nodes to my heart’s content. But then I’ve got to manage positioning and linkage on my own. Plus, yEd requires keyboard and mouse interactivity.
That doesn’t sound like much but when I’m blasting out text, even in bullet points, it’s a serious pain in the ass to move my hand back and forth off the keyboard. I’m typing THAT fast and it’s a damned disruption to which I won’t condescend.
About the best piece of software for keyboard-only diagramming like that I’ve found is FreeMind, which is a generic mind-mapping software and brings me back to the topology problem.
I really really really don’t want to write something from scratch. Dealing with UI coding is a headache enough at the best of times. But actually going so far as to manage blank canvas drawing, object representations, linkage drawing (bezier curves, anyone?) diagram fitting and such is the kind of thing that just gives me a fucking migraine.
I’ve got it in my head that GraphViz might be a useful way to go. But the diagrams it generates are still a bit stodgy and clinical. They’re useful for displaying small relationships between few nodes OR large clustering on a zoomed-out diagram. But there’s no interactivity so I’d have to regenerate the diagram every time there was a change to the underlying source, much like a recompilation process. AND it can only really be “given hints” about layout as its algos determine actual placement of all nodes.
So that’s really out for anything other than a documentation type process.
I’ve been banging my head against this wall for years. Shit, for decades if I’m being honest. Something like yEd is just SO close to what I want.
What would be about perfect is if yEd had a plug-in architecture or configurability that I could use to drive it completely from the keyboard. Creating nodes and ‘linked nodes’ without having to reach back and forth. I’d be willing to condescend to the mouse for some of the weirder functions, certainly.
Other than that I’m not sure I’ve even seen something that would fit the bill.
If I can let go of the graphic aspect of organizing my thoughts (I can’t, but let’s play a game) then what I’m left with is exactly a wiki structure: Hypertext linkable to arbitrary depth and breadth. I could even have different TYPES of linkages, categories, tags, just straight content linkage. All kinds of things.
In fact this is how I’d prefer to write, though my own software development to this effect is proceeding only in fits and starts.
And because I apparently can’t fucking help myself…off I go into the stratosphere…
I just can’t shake the feeling that if I had The Right Tool that I’d be able to dump the contents of my skull out into something that would help me organize and deal with it so I wouldn’t have to try and keep all the moving parts in the weird 4d diagram in my head all at once. There’s just too much there to juggle without externalizing it.
There’s this overwhelming sense in my head that there’s something big in there that I can’t see for no reason other than I can’t juggle all the fragments and pieces at once so that I can put them in to the puzzle, at whatever dimensionality it exists in.
Now that could be an illusion, just a complete fabrication. But I have no way to tell unless I keep pushing towards that frontier and fail forever. Even then I won’t be sure.
It could just be a damned emotional state drummed up by some inner working that I’m trying to attach higher level conceptual structure to in order to fit it into a more intellectually comfortable framework.
The fucking compulsion to solve this problem is driving me insane. Hell it may have done so quite thoroughly. After all I really am quite convinced I’m Mad.
Here’d be a fun thing:
Check out some of the VR painting systems. Imagine one of those that had gestures that would let you draw links between objects and create objects. Then a voice system that would let you put names on them.
Christ that would be amazing beyond compare. Of course once you had a digital representation you could process it with any piece of software since there’d be no reason not to have it saved in some easily machine manageable text format. Fucking json would work.
I think that’s about the closest thing I can envision to a completely functional piece of software.
Then I could sit at a keyboard and expand on each titled node, giving it a content body and whatever other attributes and tags I couldn’t manage through the VR interface itself.
It doesn’t….SEEM like it would be that hard. But I’ve spent years watching Minority Report and Tony Stark manipulating things in CGI, so my understanding of what’s possible and reasonable may be “a bit askew.”
Dammit there’s something here.