Well, I’m writing today on the lapcom, not for any particular specific reason, but just because, as of yesterday afternoon when I was preparing to leave the office, the various strengths (or magnitudes) and directions of the vectors in my brain came together and produced enough impetus for me to pack and then carry the lapcom with me.  After that, there’s much less of an energy barrier in the way of actually writing using the lapcom in the morning.  If I have it with me, I prefer to write using the lapcom rather than the smartphone.

As for the subject/topic of this blog post, well…I don’t know.  I have no idea what to write or what to say, other than to say whatever comes out of my (figurative) mouth here.  I don’t know what I’m going to write ahead of the actual writing any more than you know what my next word will be before you read it.

I guess that’s how conversation happens in real life as well.  People (at least, neurotypical people) don’t tend to script their conversations.  They just start talking, and the process of talking is identical with (or at least part of) the process of forming the ideas in the first place.

You don’t think about that to say before you think and say it, nor do your think about what to think about and say before you think about thinking about and saying things.  One cannot in any literal sense fully plan ahead what one says or thinks, because it would imply an infinite regress.

No, most of the time, expressing one’s thoughts is the first circumstance in which one formulates those thoughts (though of course it might well not be the first time one has had the same or similar ones).  Ideas, and even memories, do not exist fully formed within the brain, as if on some SSD drive; they must be reformed every time they are brought up.

This is part of why memory in humans is so malleable.  All memories bring out not just information from the past, but also color it with whatever information is leading to the memory being recalled at a particular moment.  This feeds back on and can alter the memory over time.

The things that happen with neural nets/deep learning systems/LLMs is teaching us at least some things about how thoughts and minds work in non-linear ways, though it’s important not to draw too many or too literal parallels to actual brains.  For one thing, brains are vastly more efficient than any kind of software/hardware we have.

Though your brain is certainly the most energy-hungry portion of your body, using about 20% of all the calories you burn, this still is only about 20 watts or so.  That’s less than your typical lapcom uses, and far, far less than any LLM or similar system uses.

The brain is also more complex on many levels.  Synapses are not simple binary switches, like transistors; they are more or less continuously variable in the amount of neurotransmitter they release into a synapse, and the number of receptors in the receiving synapse, and the degree to which activation (or inhibition) of those receptors affects the functions of the postsynaptic neuron and how that interacts with other inputs and the basic metabolism and chemistry of the next level of nerve cells.

Those functions are also affected by hormones of various kinds, from peptides to steroid hormones to things like histamine and glutamate that are neurotransmitters and hormones and (some of them) even amino acids.

Then, of course, there are the inputs from the many and various “glial” cells, which are not neurons, but which “support” neurons in the brain, and which (among other things) create the myelin sheaths that allow nerve transmission to happen much more quickly in “white matter”, and so on.  These things can alter and tweak the “weights” in any particular neuron “node” in ways that are more complicated than the nodes and weights in neural nets.

And of course, the scale is daunting.  You have about a hundred billion neurons in your brain (give or take), and each one has connections with 1,000 to 10,000 other neurons, making the total number of synaptic connections in the brain on the order of a hundred trillion to a quadrillion.

Just imagine if every star in the Milky Way galaxy had some kind of hyperspace links to 1,000 to 10,000 other stars.  That would be a complex galactic network.  The brain is this scale of network.  And the synapses, again, are not binary switches like transistors, but are more or less continuously variable within a given range.  Likewise, the impulses traveling down an axon, though binary in a sense (each nerve impulse either happens or does not), actually vary continuously, because it’s the rate of impulses arriving that affects the release of neurotransmitters.  But it’s not the only thing that affects it.

Anyway, this is all interesting stuff, I guess.  Learning about neural nets is fascinating, and is yet another use of linear algebra, matrices and the like (which are also good for General Relativity).  And learning about how gradient descent works in machine learning and so on is interesting and thought-provoking (an amusing coincidence, not an irony).

I’m reading some things about this and trying to pick up on it, just for general understanding.  I doubt that I’ll ever do anything with any of it other than maybe “talk” about it here (which is nearly the only place I talk about much of anything).  But at least it’s interesting.

It’s futile, of course.  But nearly everything I do or have done is or has been futile.  And that’s not just on a cosmic scale, but on a human, day-to-day, interpersonal scale.  Almost all my efforts are either wasted or fully counter-productive.  This is the outcome of something like me trying to do and be good as much as he can; it just tends to be pointless or worse.

None of it matters, though.

Anyway, as often happens with the lapcom, I’ve written too much of nothing today, and that’s irritating to me.  I can’t imagine how annoying it must be to my readers.  At least, the rest of your day will probably be less annoying by comparison with this post, right?

Right?

I guess maybe it doesn’t really work that way.  It’s all just irritating, isn’t it?  It’s irritating that I don’t catch a terminal illness, some kind of deadly but not-too-quick infection or cancer or something.  It would be good to be able to have people know that I was dying, so that if anyone out there is mad enough to want to see me or greet me or say goodbye to me before I die, they could.  I think I would appreciate that.

Odds are, though, I’ll just disappear and be gone someday, perhaps very soon, and almost no one in the world will notice.  Those who do notice will only do so vaguely, because there is practically no one to whom my life is integral or even very strongly connected.  I’m just background static for the most part.  When I go, it will just be good riddance to bad rubbish, as they say.  There’s really no sensibly available better alternative of which I am able to avail myself.

As Kurt Cobain sang, “Oh, well, whatever, never mind.”  He had some good ideas, did Cobain-sensei, and he carried them out.