I was writing today, and I couldn’t focus at all. In a talk with an LLM, I was referring to their :fermat: hallucination and they began to answer me a fake reason for the hallucination. I was thinking how each time that I get interrupted (the moments that I notate with :fermat:), the thread and the points I wanted to convey are all lost, and I begin to write something new that tries to continue the previous world, exactly as the LLMs predict the next words based on the previous words. So I told them, and there was much discussion, in one of their replies in which they told me this:

As I was getting more and more distracted, I began to watch how by each fermata I had also incorporating the things that were in my mind. And I was thinking about a Mind Graph navigation. I began to write down my thoughts and how they kept reminding me of new things. Here is the list:

• The lighting in the room,
• Maestro Rachmaninoff that I was listening too,

  • Thinking that this is me deliberately making a central node that connects to all my thinking nodes. A form of storage that by listening to it, I load all those nodes again to my head maybe?
• Remembering the film where Rachel McAdams used to make art by listening to very laud heavy metal.
• Remembering how different Amir Jani’s working music was rather than mine.
• Remembering Zea playing FIFA with Amir, and alone.
• Getting focused on the trill of the second movement.
• Thinking about guys in the other room.
• Thinking about going shopping.
• Someone talking about a Milad, and I remembering Milad, that I have a meeting with him.
• Looking around the office and getting excited about the main theme of the second movement.
• People’s laughing in the other rooms making me angry and irritated of loosing my thread of writing in here.
• Others laughing makes me angry.
• They talk about the stupid Bale messenger and I remembering Kamran.
• I remember Jo Cafe and Tanya .
• Remembering Tanya makes me remember snow, making ducks, and how Téo had hugged her when she arrived.
• I get reminded of my room.
• I get reminded of me sitting in the end of my table and watching the sun light come through the window and shine on the room.
• I get interrupted by the grace of the third movement.

  • I noticing my word grace in the middle of the writing, and remembering the series crown, and then africa because of it.
• I remember how privileged I am that I can be doing this and listening to Maestro Rachmaninoff in my ████████ time.
• I think about pour structures.
• I remember how Zea used to keep saying random things one after each other sometimes.
• I remember the stream of thought.
• Cats in the backyard.
• Barbecue that my Mom wanted to give us.
• Ashkan and Parastoo.
• Why did I think about that?
• Seeing Lang Lang.
• Remembering Naser Anear talking about Lang Lang.
• Remembering the Pars Music Institute at the Palladium.
• Then childhood classes.
• Going for ice creams.
• I think about a product Manager here and remember Arian and Lili.
• I remember Lili’s writing at the war.
• I remember the day Persepolis person came to our office and how coward the Arian was.
• I remember my green scissors and how I went shopping for them after one day in Fabizi.
• I remember Mr. Sharif watering the plants (as my box of scissors is located besides the window, and I sometimes see him there.)
• I remember Kamwa and the new cute brother I still have no name for.
• I remember walking in the street with Zea in the morning and taking pictures in the same window each time we walk pass it.
• I remember Shahrzad purchasing gold, as I hear the name gold from the around.

Do you see how devastatingly interesting is the path the mind traverses? It was only what happened in a few handful of minutes. By each fermata in my writing I had completely changed the way I was writing as I had encountered so much strong new information.

Previously I used to think of result = f( Mind Graph , query). And that is correct, but lets change the formula to this: result = f( Mind Graph , sensory information, query). This changes everything as you can see, r = f(mg, si, q). Our brain processes have way more context than we can normally assume. It is the combination of all that we have second. The whole context of the environment, all the senses we have at that moment, how those integrate with the Mind Graph and how they influence the new connection explorations in the graph as they bring up new nodes into focus. (the concept of focus requiring much work and anticipation later.)

I think I am finding a new branch of my theories. In the morning, and yesterday, I tried to write but then my brain was fighting with me. It was as if it didn’t want to finish just right there. So I had given myself the space to cook more ideas in my mind.

Today I had this thoughts on my mind about design and photography, about capturing details or capturing the generative algorithm, and again my mind tried so hard to divert the writing to a whole different thing. Now that I am writing I see how they are connected and I think I have just realize what that has been in my mind.

We generate our ideas by contiguously navigating different parallel paths given the sensory information. My mind goes on to write something, and once I am done I realize I have incorporated the things that I had thought about in the morning, the ones that just happened now, or the previous week. Now this is either of the two cases: I am aware how long it takes for my mind to come up with an idea, I see the process, I know it is cooking up something, and I see the signs beforehand mostly. But then the shape of the mind is unusual. [one thing I notice is that while the speed of my tying is way faster than doing it by hand, it has removed the friction of my thinking.], and I get that I am thinking and processing and I see the paths being explored I just don’t know where will it end. And so if we think about it, each now thing we say and materialize is like a quantum measurement, I had talked about this way before in the “Maybe the Thought Should Stay In The Mind” (Kary 28⟡176). I used to see and sense the ideas being cooked up in the mind, and I knew that the message, the medium based artifact, was materializing that strange quantum world into a relativity deterministic world. And you would have lost much of your inner process after this, as you would have lost your thresholds.

When I write I hardly have anything new to say, but then when I get bored, when I leave my brain on its own, it comes up with wonderful new things. I was stuck on this problem until I went for the lunch, and while eating had my brain do the job and give me the answers. It did. And that is wonderfully strange. How can it be honestly? So I now have realized something: we think with a high band width of SI and leaving it to cook. Bugs gets fixed when an external eye comes to the project, breakthrough moments never happen after a sustained focused session, they happen after accidents, after something suddenly happens, it is always “after something happens”, after a new catalyst SI comes from the outside.

And the funny thing is that when I write and I have distractions, I get creative and destroy my writing. You see? the writing is the externalization, the thinking is the creation. So when you think you need to be in a chaotic environment, you have be surrounded with random and powerful SI, but when you want to materialize that Mind Graph into a message, it requires focus.

I work great in the middle of things, I have my ideas in the shower, when I’m walking, in my conversations with other people, when I watch films and series, when I read books, when I travel, when I do anything that is not ordinary and is SI rich. And I have to write at full focus or I lose it, I can only write when the distractions are minimum, when I only have my mind and my writing, and so on. I have rituals of just this paper + just this ink + just this pen + just this style, because they give me the tools to focus. I can only write in the iA Writer in the full screen on my mac. I can only write in Visual Studio Code with my focused configurations and only like that I can write. But if I have layout options that I have in AnnotationsKit or web maybe? If I have a few pens that I can switch the colors, no way. I change my tools from time to time, this keeps the focus shifting, but other than that, I think writing requires focus. This is just authoring a code, you think when you make coffee, or take a shower, and when you code you materialize that information into the editor. When you write the code the focus is required, when you actually think about the algorithm distractions are required.

And that is it! After many years of thinking about this or that? Maximalist vs Simplist? Dynamicland/Smalltalk “too many things on the desk to trigger creativity” vs the “banking model of focused study”, I think I am finally beginning to see what is happening in here. The idea of “sleep on it” has to be extended to “live with it for a while”.

So if we take these together than creative work is taking a query, thinking about it, and having the sensory information of your around, more and more make you explore random nodes, when you explore these random nodes they connect together and since they are random and your Mind Graph is random, the creations become too random. Your mind is basically a Toolbox Theory environment on steroids and you are constantly trying out the random tools you see here and there, their combinations is your creativity and for that you require randomness, and when you have the randomness you can find new structures (by brute force and educated guess maybe) that become your new ideas. Now this happens in your mind as the paths the mind is exploring is really complex, and then you get to different kinds of things, and you keep mixing all things with each other. Iit is not “I am thinking about a^2+b^2=c^2”, it is the thread of navigation I just showed you above, it is so random, and you “maybe onto something”, but then an interrupt make you forget the thread. You keep thinking and it is in the unconscious so you hardly get sad for being interrupted, it would be no biggie. You already just either sense or see very tiny bits of that in your head.

And writing this down is even harder as it means focusing on how did you reach those places, where they were, how can you materialize those graphs into abstract language, and so that requires focus, not to loose a thread. And when you loose that thread it irritates the hell out of you as this one is conscious, you have to hold your points and arguments and all of them in your mind not to forget them when their place reaches. This is just doing programming, you have to have a simulation of the whole software in your head and know you have to write function A, and in that start writing function B, and in that start writing function C, because B required C, and now that C is done you can finish B, and then go back and finish A, as it also required B. So it is complex, but it is the process.

I have much else to write, and way more to materialize, and this was the most vague and ambiguous picture of the graph I sense, but I had no better way to do the Sub-Graph Transfer as the abstraction is not finished even to me. But! it is evident now that he thought actually stays in the mind.

LLM Commentary

After talking to an LLM, I realized my point was not understood at all, so I’m putting this in the end as it clarifies my point so well for other readers such as you as well: