2020-05-12 | Subject | Freaky³
It is exactly one year after I shifted my personal focus from data at rest and private, solo journaling, to modeling system knowledge via ontologies. I wrote an article to kick off my effort last year here. My opening sentences called out the situation of impending collapse, unreasonable economic models, and negative externalities. I concluded that the methodology that I envisioned would help respond to various crises that we faced during collapse. And yes, this essay gets a bit on the freaky side of things with my religious archetypes and other associations, not the least freaky of which is the term collapse. But, well, it is a freaky world, my formative years were freaky from a nurture and information perspective, and my life has been freaky. Don't get me wrong, I cherish Freaky³, I wouldn't trade it, but it taints what I write and think. Still, I assumed collapse, as there appeared to be no way that humans would stop their forward motion, their all-consuming waste of the ecosystems that humans depended on.
My long-term strategy for the focus was that the modeling would be useful in my career, both for broader analysis like organization-wide data flow, but also useful for tactical efforts. I knew that my past methods using interviews and manually drawn diagrams of systems were too time consuming. My premise was that leveraging existing ontological/graph analysis tools to form agile capture and visualization as well as automated inference would resolve this problem.
At first, my employer was interested in how my ideas could help them. I focused on tracing data flow through their extremely complicated systems, and highlighting dependencies on an expensive third party cloud service that they wanted to eliminate. I made good progress. I generated deep data flow diagrams that showed the dependencies using about 50/50 inferred relations. They ended up laying me off, though, and eliminated my role, saying they had no need for my expertise.
I then spent most of my waking hours since late last September creating the entire system in Python, using a simple data store of a filesystem with symbolic links. The advantage of using a filesystem is that it can be managed with filesystem version control software like git. It was also simpler than a database, which had been my focus with MCJ for the last couple of years. I decided to combine this with MCJ to layer documentation on top of the graphs, and design a robust versioning and replication scheme to allow collaboration. Finally, I wanted to get the system to the point that it could handle simple inference, a form of machine learning/AI.
I finished this at the end of April, up to the point that I had inference working on a single dimension across all layers of an explodable graph. From that point forward my main focus is capturing the software, which I did here, and filling out the other domain documentation. Documentation is slow going, and a bit more boring than coming up with my own agile graph that can be queried and has a rudimentary form of artificial intelligence. No effort is finished until the documentation is complete, though, as painful as that might be. While I would like to move on to more sophisticated ontologies, the documentation needs to come next.
About this time, global focus and anxiety about c~19 reached a fever pitch, and we started shutting down the economy by staying home. I figured I would use the supply chain as an example of my modeling method, as it seemed immediately relevant. I wrote about that experience in this article, but in summary I realized just how extensive and encompassing the global supply chain was and how it worked.
Now, as this paper shows, the extent and problems are not that much of a surprise. This is all tied together. My big aha moment was that what is happening now was simplifying the supply chain. This is the answer to the ouroboros problems. We normally can't break out. For instance, we can't move from fossil fuels because it is part of everything from food to medical to solar panels to electric cars.
There are some attempts to come up with different supply chains, but the whole thing is rooted in oil right now. Normally, just the resistance from the fact that all of our jobs and much of our personal time are part of the supply chain keeps us in the supply chain. It is not possible to extricate ourselves from it.
But it is what we need to do, because we are destroying all good things with this web of the global supply chain. We need to simplify it to stop the destruction.
My dad told me once that computers didn't save anybody money. What they did was allow companies to scale because of their precision, ability to crank numbers, and process information consistently. I never completely understood how this could be until relatively recently. The global supply chain in its current complexity could not exist without computers. And, while it is true that it might cost a single company more money to run their business with a computer than without one (at least when my father was selling computers in the 60s and 70s) the more interesting bit is that by scaling the company it allows other companies to grow along the resulting supply chain. This growth provides other livelihoods, which also become customers, and it grows and grows and grows. The wider and deeper it grows, the harder it is to control negative externalities. But the economy grows, and that is how we view what is good for us.
By staying home, we have done the impossible. We have facilitated the simplification of the supply chain. Yes, we have also cut out an enormous amount of jobs, jobs that people need to buy food. Humanity has been given a gift, though, overall, as far as having a habitable planet.
So I have ended up in a different place a year later, than I thought would. This happens frequently with me. I can imagine that we could unwind the stranglehold the web of the global supply chain has on the world. I don't know which way it will go from here. Certainly it is becoming simpler as I write. Will it have a more direct-farm-to-consumer focus? Will it be a single global corporation that routes needed goods around the world, leaving most of the global supply chain branches dead? I can't answer that. Now, we still have big issues with climate change already baked in, so to speak. We will also come out of this crisis with cheap oil and hungry humans. How many hungry humans, and how rapidly we recreate business as usual, is likely up to how c~19 mutates and its success at putting itself forward in the future. But there is hope, a small hope that our new-found symbiotic relationship with c~19 will save us from ourselves.
I wrote the Dark Angel meme based on an idea that came to me in the morning. I used the same rules as dreams, trying to stay true to the first words that came to mind. Afterwards I considered what Dark Angel meant, and remembered that there was a show by the same name that I watched when it came out. I figured it would be interesting to watch again, so I ordered it off of ebay. I finally got around to watching it during my workout today. In the pilot episode, the Batman-guy says that the Dark Angel is a soldier created using recombinant DNA, a chimera. I really don't know what the guys name is yet, but he sure is acting like Batman.
Season 2 of Dark Angel is even more related to our current troubles.
Articles tagged with dad on Aggie Codrust:
Articles tagged with dad on O.R.N.G.:
2018-12-03: The Cost of Computers
2018-08-07: Rock and Roll
2018-07-15: Great Grandpa's Log Splitter
2017-08-27: Imagination and Cripple
2017-03-02: can't be fixed
2013-03-21: You are Not OK
2012-11-21: The Hostess Entries
2012-06-02: Rainier Beer and Chat
2012-04-22: Forgot Hatred
2012-01-15: Colored Seaweed
2012-01-09: Old Pontiac
2011-12-13: RUST SW street
2010-08-30: The Wrong Ladder
2009-06-28: An Innocent Computer Move - Sunday
2009-06-27: An Innocent Computer Move - Saturday
2009-06-26: An Innocent Computer Move - Friday
2008-08-09: Old House and Fire Pit Dream
1993-02-05: Letter to Grandma and Grandpa
1980-07-08: Painting Houses, Digging Ditches, EOL
1971-01-05: Honeywell Computer Chip Animals