This is Project 262c86ec. There is no well-formed statement of the research. At one point in the past I called this kind of thing a “symbolic engine”. Others appear to search for the best knowledge base design or personal computing operating system design, out of a dissatisfaction with the app and program ecosystem. To me, they are different facets of the same problem.
Please talk to me if you are interested; I need to connect to the discourse. Specifically, so I can answer your questions in my writing.
Snippets of Reasoning
The Challenge of Integration
Written 2024-06-20
The term “itemization” is brought from Alexander Obenauer, since his first manifesto of itemization. In my previous literature this had always been considered under the notion of “integration”. This is extremely difficult. Simply put, integration-capable things need to consider a million possibilities, so they are tremendously difficult to design. In fact, anyone knowledgeable would realize that such integration expects huge problems about language. When one actually tries to design it, they will surely confront the hardcore problems in philosophy of language, math, and formal systems—the same problems programming languages have explored.
What am I inventing then? A different line of thinking, where any formalization is post-hoc in light of a history; an information system to truly support cognition with a history.
Keep a History, A LOT OF HISTORY!
Written 2024-06-18, 2024-06-20, 2024-06-22
Perhaps it has been clear to many researchers by now: in creating this futuristic vision of information processing, we at the very least need to seize every chance in the system to record user interactions and their effects, or in other words, history. Everything input immediately becomes a part of history, and things from the history will eventually become useful as current data, because from our research we have learned to blur the line between the history and the current (or in my opinion, fundamentally, the reality and the cognized). Every uninspected step in automation is lost opportunity. Every abstraction will become an obstruction. Abstractions are not any more real than its raw data. This total openness to change is a unique characteristic of the information system of the future.
Presently, I hold the claim that if a system processes any information ever encountered, ever derived and could ever been cognized as they are preemptively “already in the system”, then such a system refers to precisely the notion of history.
Putting Everything to Scrutiny
Written 2024-06-13, 2024-06-16
Thesis 284a40ad. The curse of natural language. Natural language is a curse because it is habitual. You almost cannot control your response to a certain piece of language you care about, although you can partially fix this by evaluating the sources and working 10 minutes on human psyche. And now thanks to generative “AI”, it is easier than ever for anyone to exploit your habit.
Proposal 51879e1b. Assigning UIDs to everything is an attempt to evade the curse of habitual language. It will work in both directions—influx and efflux—with respect to a person. It not only protects users from concepts flooding onto them, but also never presents anything they say to other people as universally qualifying. Obviously, this will have to work as a culture, although induced by an instrument.
What right do I have to show you information in any typeface but Univers, in any form factor but a constrained one, bounded by a frame with meta-operations, linked to my references? I want my information to be helpful to you, not coercive, so I by default assume this completely powerless stance.
In the same way as we shouldn’t irreverently appropriate others’ fine works for own, as we must carefully take over the values in those works. There are writings that do not deserve to be typeset in Garamond, designs that do not deserve an edge-to-edge layout. In fact, all presentations must be supplied with meta-operations, thus the reader always has the space to declare content as incomplete. Thus, these meta-operations enable the user to treat the content as history. Enabling any idea to enter history is a categorical feature of this system.
Agency-attached Language
Written 2024-06-13
An article expressing basic views is here.
Because language is too public and erroneous (previous curse 284a40ad), one might be motivated to explicitly declare a language private, as in UID Everything! (51879e1b) This feature produces what I call agency-attached language (fbfecb8d), where “agency” means the faculty of deliberation. However, this is only effective if unauthorized speakers are unable to use the language. Therefore such UIDs must be backed by an authentication feature such as asymmetric cryptography. I am not sure how this will work.
Does this sound like Web 3? The problem with the current status of Web 3 is that although data is stored decentralized, language is not maintained decentralized. Language as practiced by any user should be subjective and time-dependent, and the knowledge system must be designed to conform.
Knowledge System Architecture
Written 2024-06-14
Department of Defense Adopts a Philosopher’s Applied Ontology
I am not defending the advance of war technology here, but can you analytic philosophers all be a bit more practical? I present you the challenge of creating an implementable knowledge base framework meeting the following requirements, but you are welcome to debate these requirements as well:
- It always works for the user’s cognition. It should be possible for things to be expressed as close to how they are conceived as possible. This is the top priority of a knowledge base. There is not necessarily tried truth to be expressed, let alone formal truth. This defines a knowledge base, rather than a programming language, up front.
- It allows users to formalize those conceptions, enabling the following formal features.
- It can function as a database.
- It can be automated with good expressivity.
- It can be integrated with other systems with little extra complexity.
- A professor and an analyst would want to use it.
Optimization is not required. Now, what would be a philosophical argument leading to a good knowledge base architecture?
Formal systems too hard, help
Written 2024-06-16
Reference to a dynamic reality is a challenge because of how complex it is. I thought it takes very expressive language to make things work, and this might need to be formalized. But do we have the right data in the first place? Obviously but hastily put, the complex system reality is more expressive than any formal system. All these complex situations requiring expressive language may all be unstable to perturbation after all—the abstractions we establish are only post-hoc understandings of how they behave, rather than their essences. However, if better data can be collected, then a lot of complexity can be balanced. For example, labels can be put at a lot more places than we currently are doing.
Necessity to Collect More Information
Written 2024-06-18
Perhaps a great illusion in this project is that, in this futuristic vision of information processing, we will not be requiring more information—mostly about their reasoning process, as well as sources—from the users than what we currently are doing. Thus I have used effort in vain to reason in philosophy and formal systems. This might have been fundamentally my incapability of imagination.
However, there still a worry about the proliferation of information and combinatorial explosion.
GUI
Written 2024-04-17
Though, relying too much on vision might not be a sound approach about information.
Incorporating Zoom
Written 2024-06-18
Permissive navigation has discrete navigation steps. Not all things do well with discrete steps. For example, consider a chat session with your friend. In its miniature form, it could be your friend’s name only. If you allow it to be a little bit larger, it can show one line of message preview, then perhaps a date, then perhaps an input bubble. In all these cases, the chat is a real chat, itself, in the sense of Alexander Obenauer. There is no clear way to specify at which point the user has navigated to the chat, as it is not very discrete.
There is a solution: Zoom and Detach. The user will be able to zoom on an item, creating an easily dismissible bubble on top of it, to reveal more aspects of the item. The bubble can be interactive. If the user wants the full interface to persist, then they will detach the bubble, declaring that view independent.
Ultimately, you can see this as a special version of map zooming. Duplicating a view is the remedy to any non-discrete navigation.
Related Projects
NixOS
Acquainted 2022
The operating system derived from a functional package manager. This is the closest real thing to a malleable system, albeit behind-the-scenes only due to addressing only the package manager—that is, the software supply chain and environment. Unlike the rest of the research, NixOS is a reality, so every malleable system researcher should better try it. Additionally, read the paper.
Plan 9
TODO
EROS
https://en.wikipedia.org/wiki/EROS_(microkernel)
CJDNS
Acquainted 2024-07-09
https://github.com/cjdelisle/cjdns
A step from the current Internet’s trust-oriented routing infrastructure, CJDNS uses asymmetric cryptography and a distributed hash table to derive IPv6 addresses. Sounds like a lot of agency.
Joel Jakubovic
Acquainted 2024-06-28
https://doi.org/10.22024/UniKent/01.02.105537
Ink & Switch
Acquainted 2024-06-14
Most projects are pertinent, especially Cambria, End-user Programming and Alexander Obenauer.
Alexander Obenauer
Acquainted 2024-06-18
OLLOS, WonderOS/Workbench, Embark.
Obenauer uses the phrase “itemized OS” for his vision, featuring “universal data portability”. His lab notes are all important. The first few concrete designs were inspiring, but the later ones slipped to be somewhat ad hoc.
Orion Reed
Acquainted 2024-06-22
A very comprehensive researcher. “system interaction, knowledge representation, cybernetics/self-organisation, programming systems, […] abolishing apps, objects are reference(s), PLs should be views/lenses, digital language is privatised.” (Here)
Liberatory Computing
Acquainted 2024-06-22
This project includes a very good research library with more than 60 works. Surprisingly, I have attended an in-person conference presentation on one of these works.
Malleable Systems Collective
Acquainted 2024-06-22
Well as the project name indicates.
https://forum.malleable.systems/t/taking-alan-kays-vision-to-the-extreme/200
HASH
Acquainted 2024-06-14
A very commercial project for organizational knowledge. Supports types, simulations.
https://hash.ai/blog/self-building-database
https://hash.dev/blog/structured-data-value
Key concept: Distributed schemas. But this might not be the distributed language I wrote about. Not much new GUI.
I have successfully run their test server. Contact me if you have trouble doing the same.
Mercury OS
Acquainted 2022
First (2018) system-wide implementation (that I know) of something resembling permissive navigation, but not malleable.
Laevo
Acquainted 2024-07-12
Xanadu
A reminder that information systems should not imitate paper. Any digital system should strive for better expressivity. https://xanadu.net Demo: https://www.youtube.com/watch?v=1yLNGUeHapA
That Conway Paper
How information systems heavily reflect organizational structure. Conway, How Do Committees Invent? Consider poststructuralism.