NATION

PASSWORD

Pacifica Ascendant: A Postmodern Saga OOC (Open)

Where nations come together and discuss matters of varying degrees of importance. [In character]
User avatar
Lagunaca
Attaché
 
Posts: 80
Founded: May 19, 2006
Democratic Socialists

Pacifica Ascendant: A Postmodern Saga OOC (Open)

Postby Lagunaca » Sun Apr 12, 2020 4:23 pm



Pacifica Ascendant: A Postmodern Saga OOC (Open)

This topic is the OOC thread for an IC fork of Rebuilding From The Collapse, a postmodern tech roleplay set in 2110 that explores how the world reorganized itself after a global societal implosion in the 2020s. Rebuilding From The Collapse is currently on hiatus, with hopes that it will return soon.

The primary themes of Rebuilding From The Collapse are national, governmental and political reorganization in the 90 years following The Collapse – along with the exploration and colonization of the inner solar system, espionage between nations and the realistic development of technologies including nuclear fusion, nuclear propulsion, ion propulsion and artificial intelligence.



Nuts and bolts

In this companion OOC thread to Pacifica Ascendant: A Postmodern Saga, you’ll find footnotes, technical resources, scene setups and background commentary for the IC topic. Everyone is welcome to participate in this OOC thread, as long as comments are kept constructively focused on and relevant to the IC story. The IC topic is invitation only.

The rules:

    1) Abide by the NationStates site rules

    2) Abide by this NationStates forum’s rules

    3) Anyone can participate in the OOC, but IC participation is by invitation only

    4) This is a topic about postmodern culture and realistic postmodern tech. Please keep comments within the bounds of currently understood reality and the known laws of physics.

    5) Comment on specific posts and stay on topic. No reddit-style grab a**ery please.

    6) Be nice to others and keep it clean

    7) As the moderator of this topic, Lagunaca’s decisions are final, subject to the even greater finality of any NationStates mod’s decisions.
Name change transparency:

Lagunaca is a new RP nation name to replace the RP nation Wasted Genius in Rebuilding From The Collapse. Quite frankly, I never liked that old name and only used it because it was familiar to another player in RFTC. The new name is the combination of a portion of the names of the city and state where I’ve lived for a large part of my life.

Original concept credit:

Pacifica Ascendant is a fork of the Rebuilding From The Collapse RP, which was conceptualized and created by Maineiacs. It should be noted that Maineiacs’s concept for RFTC predicted the fragility of the global economy and the potential for a collapse BEFORE the recent bad news broke and while the strength of the US economy was considered by some public figures using questionable statistics to be at an all-time record high. (This isn’t the first time Maineiacs has nailed a topic by the way, but I digress.) I received permission from Maineiacs to pursue this fork while he tends to other matters.

Last edited by Lagunaca on Tue Apr 14, 2020 1:27 pm, edited 3 times in total.

User avatar
Lagunaca
Attaché
 
Posts: 80
Founded: May 19, 2006
Democratic Socialists

Pacifica Ascendant: Scene Setting

Postby Lagunaca » Sun Apr 12, 2020 4:40 pm



Pacifica Ascendant is a fork of the RP Rebuilding From The Collapse

Scene setting: What does the world look like in 2110?

Not dystopian - Rest assured that in this future, form fitting black leather and latex are accessories worn only by American cosplayers or legit European motorcyclists. Otherwise people dress sensibly and not in cargo pocketed urban camouflage tactical gear like they just got kicked out of the army for being stupid or crazy. Night-time streets are well-lit so the surveillance cameras can do their facial recognition work (and this illumination is not provided by uncountable kilometers of energy inefficient pinkish red neon tubing). Days are mostly sunny. Except in Portland, the weather there is still pretty dystopian 9 months of the year, same as always. And the southern California coast in June sometimes into July, that’s still just plain gloomy.

Globalism – Yes. Your dog kibble, children’s toys and small kitchen gadgets are still made in and shipped from exotic, faraway countries. It’s just that all that stuff is produced by robots, not human slaves. And by the way, robots don’t look like people or plot to rule the world. Except for the showy ones, they look kind of like metal versions of people but they still don’t plot to rule the world. Plotting to rule the world requires the peculiar mix of marginal sanity, egomania and subconscious blood-lust that was developed as a quirky survival trait by humans over thousands of years.

Surveillance state and the surveillance economy – Go anywhere outdoors and you’re watched at all times by something that will rat you out, probably an AI or maybe an autonomous aerial drone. Indoors is the same story, it’s just that you pay for it there and the people (or AIs) you pay to surveil you in your home sell this surveillance as a commodity that is then anonymized and turned into the Soylent Green of consumerism. And you love the convenience of this system, so much so that if they took it away you would riot in the streets to get it back. And you never notice it anyway unless you look up at the top of the 10-meter poles that are everywhere.

Government by Distributed Bureaucratic Authoritarianism – it’s not so bad once you get used to it. Besides, in much of the western world you’re actually living in it right now (they’re just kidding around when they call it freedom and democracy on the news). The only difference in the post-collapse system is you don’t get to vote for a game show host and his family of mutants or some senator that is the same age as your great grandparents and isn’t quite sure where he is at any given moment. Turns out there are advantages to having experts at the top of their game running the government.

Culture and popular styles - Culture and styles reflect the postmodern pivot points of the late 20th to early-mid 21st century. That is to say it’s a heckuva lot easier to keep recycling unique designs than it is to create them in the first place. In 2110 a brand-new car still looks like a low coefficient of drag version of a 1960s muscle car, but its virility is measured in kilowatts not horsepower. As for fashion, pleated-front, wrinkle-free khaki pants are still tolerated in a business setting but are also still frowned upon by potential mates.

Climate change - The climate has definitely changed, but you can still snowboard in parts of Mammoth or sometimes even Big Bear in the winter, which is an important measure of the personal impact of global warming. Wildfires are a year-round reality and a very large and well-funded Fire Authority works constantly to preserve dwelling and forest alike. The downside is that your dwelling or forest (or car or pets) may be aerially coated in Phos-Chek orange slime without notice on any given day of the week.

Capitalism failed - a long time before anyone admitted it. But you still have to use money that you are paid for your work to buy your flame broiled chicken bowl and Ikea furniture. Banks are places to keep your money safe and handy, not the sinister masters of vast dark (money) pools that distort and undermine the economy.

The military - has become an actual defensive force tasked with the nation’s domestic security. It is not a toy of rando electoral-college winning tyrants that use any excuse to “project power onto” (kill) relatively defenseless “enemies”(subjectively assessed scary people). The Pacifica Defense Force inherited all of the war materiel that was assigned to the USA’s Western OA (which as we insiders know is an Operational Area) during The Collapse, but most of that overbuilt stuff is either obsolete or rusting in neglect at the bottom of a harbor. (footnote: Except B-52 aircraft and Abrams M1A1 tanks which are real life Ship of Theseus conundrums, but I’ll probably cook up a series about this when I get around to telling the story of the Pacifican military someday. And note to self, that story will bag heavily on nuclear powered aircraft carriers, they are pretty worthless in an age of hypersonic surface to surface missiles. What’s more, they’re hard to dismantle and they don’t even make very good artificial reefs.) The PDA drills with weapons, but just as frequently it drills with providing humanitarian aid to people in dire need, as this was their first operational mission after The Collapse. Oh and Space Force, I’ll bet you’re dying to know what happened with that. There are bootleg copies of this short-lived 2035 YouTube original series that you can sometimes find on SD cards in the bottoms of old boxes in your grandparents attic. You’ll especially like the pilot episode titled, “All Hat, No Cattle.” It really sets the tone for the whole show. What? You thought I was talking about the sixth branch of the US military created in late 2019 and reabsorbed into the US Air Force in 2021? Oh come on.
Last edited by Lagunaca on Sun Apr 12, 2020 11:05 pm, edited 1 time in total.

User avatar
Lagunaca
Attaché
 
Posts: 80
Founded: May 19, 2006
Democratic Socialists

Footnote 1: Pacifica History Timeline

Postby Lagunaca » Mon Apr 13, 2020 1:43 pm



Pacifica History Timeline 2022 - 2060

2022 – 2025: The Collapse, worldwide depression, regional wars, famine and disease, western USA state government officials abandon their posts, military units remain sequestered on base awaiting orders, captains of industry flee to safe havens

2023 – 2027: The Dark Years - the former USA devolves into poorly policed regional Operational Areas as the federal government fails to unite the country

2025: Michael Eminescu assumes the role of Acting Executive of the Pacific Operational Area

2025 – 2050: Formation of Pacifica, the Eminescu Era

2027: Pacifica is formalized as a nation, a provisional constitution is ratified, former states and provinces voluntarily cede sovereignty. Eminescu is named Executive of Pacifica, a title he accepts with no fanfare.

2027 - 2052: The Safe, Secure and Green Recovery Initiative enacted and completed in a series of five-year plans

2030: Nuclear energy plan announced. “30 reactors in 30 years”

2035: Coastal high speed transportation corridor announced, followed one year later by the Inland high speed rail project initiative. Three West / East maglev routes to the eastern frontier begin survey and design phase.

2037: The Agency of Inspectors General is formed as a consolidation of government intelligence, budget oversight and internal investigation departments. The government makes massive investments in surveillance, data mining, supercomputing, data warehousing and artificial intelligence technologies.

2040: Census shows Pacifica population is over 180 million. Government employment tops 30 million.

2050: The Pacifica Space Agency officially begins operations, enabling development of existing technology and research into new propulsion, manufacturing and robotics capabilities

2057: Pacifica celebrates it's first 100% fossil-fuel free day on February 6, 2057

2060: Completion of the "30 reactors in 30 years" initiative. Pacifica generates 70% of its energy from nuclear fission reactors that are built deep underground in former mines. 30% of Pacifica's energy comes from renewable sources.

(View original RFTC post)

User avatar
Lagunaca
Attaché
 
Posts: 80
Founded: May 19, 2006
Democratic Socialists

Scene setting: Get Your Astronaut To Mars

Postby Lagunaca » Mon Apr 13, 2020 2:35 pm




The inspiration for the RFTC series title: "Get Your Astronaut To Mars"

Buzz Aldrin - a true American hero!

Image

buzzaldrin.com
Last edited by Lagunaca on Sun Apr 19, 2020 3:56 pm, edited 1 time in total.

User avatar
Lagunaca
Attaché
 
Posts: 80
Founded: May 19, 2006
Democratic Socialists

Tech note: Pacifica Space Agency Timeline

Postby Lagunaca » Mon Apr 13, 2020 11:08 pm




Tech note: Pacifica Space Agency Timeline

As Pacifica grew into a large nation, its government grew into a large bureaucracy. This wasn’t necessarily seen as a bad thing by the population because along with the bureaucracy came millions of stable, well-paid jobs. The downside of a bureaucratic government is the requirement for documentation: charters, plans, project timelines with milestones and pro-forma budgets. Objectives. Goals. Strategies. This applied to every venture, regardless of size or lifecycle. The Pacifica Space Agency, committed to one of the most ambitious and adventurous missions imaginable, nonetheless was a creature of the bureaucracy. So then, no history of the space program would be complete without the documentation.

As defined in the Pacifica Space Agency’s 2050 Operational Charter and Long-term Plan, the PSA intended to colonize the Moon in a five-stage project:

Stage 1 (2050 - 2060): Build a fleet of reusable, single-stage-to-orbit delivery vehicles based on the Skylon spaceplane architecture

Stage 2 (2055 - 2060): Build a Low Earth Orbit Spaceport that would be the staging point for translunar missions

Stage 3 (2060 - 2065): Build a Lunar Orbital Spaceport that would act as the lunar terminal for translunar missions and have the potential to expand to become a terminal for Mars transports

Stage 4 (2065 - 2070): Build a Moon based mining and processing facility to produce input materials for the Moon based fabrication facility

Stage 5 (2070 - 2075): Build a Moon based fabrication facility that would be used to expand the Lunar Spaceport, build transport ships and containers for the Mars Transfer Project

The Mars Transfer System began as a research project code named "Project Wagon Train" that was commissioned by the Space Initiatives sub-committee in 2080. The Board of Directors authorized the program for the 2085 five year plan. The project became operational in 2090 and had the capacity to deliver "hard goods" to Mars on a circuit that delivered payloads about every two to three months, with the outbound leg of the journey taking about 12 to 14 months. The system did not accommodate astronauts nor could it land on Mars, according to the design requirements.

The Get Your Astronaut To Mars series will continue with hastily planned and executed changes being made to the Mars Transfer System in order to compete with other nations and get to Mars first or bigger or better... or something TBD, no one really knows.

View original RFTC post
Last edited by Lagunaca on Mon Apr 13, 2020 11:23 pm, edited 2 times in total.

User avatar
Lagunaca
Attaché
 
Posts: 80
Founded: May 19, 2006
Democratic Socialists

Footnote 2: Steering Wheels in the 22nd Century

Postby Lagunaca » Fri Apr 17, 2020 3:32 pm



Footnote 2: The superfluous inclusion of steering wheels in cars of the 22nd century


The superfluous inclusion of steering wheels in cars of the 22nd century is a study of fear/control behavior and the novel psychology that arises when driving a fully autonomous vehicle.

The retractable steering wheel situation is similar to the inclusion of mechanical door lock levers on 21st century cars. The mechanical levers got their start because auto thievery predated electrical devices by decades. Now they are designed to be difficult to manipulate in order to inhibit the very theft they were meant to prevent in the first place. The primary door locking mechanism is almost always electrical and can be operated individually or en masse by a switch often mere centimeters from the mechanical lever itself. In most cases this mechanical seeming lever is just an actuating plunger that engages the electrical system. On many cars, you can push the lever down, but it then recesses into the door panel and the only way to retrieve it is by pressing the electrical switch (or yes, the door lever, but the point is it’s a one way delivery of a two way service requirement). In practice the mechanical locking levers are rarely touched outside of occasional cleaning.

At new car dealerships these days it’s difficult to find a model with fully manual locks. In a power locking model a vestigial and totally redundant manual locking lever is still there because if it were to be removed completely, customers would complain and passengers would feel like they could become trapped in the car, in some scenarios at the mercy of a diabolical driver to whom they have mistakenly entrusted themselves. So the levers stay as a concession to the subtle psychology of fear and control.

But wait, as your final beats-all argument in defense of the existence of mechanical door lock levers, you might correctly pose the scenario where an assailant or ex-lover is rushing the door and you could outmaneuver them by smashing down or flipping the manual locking lever before they get there. I would counter-beats-all argue that (depending on lifestyle choices) this same maneuver could just as likely be accomplished by a mischievous child or a pet, both of whom can be more difficult to negotiate with than the assailant or ex-lover if you don’t happen to have your key fob handy.

21st century power door locks are one thing, but a hundred years later retractable steering wheels are a whole other level of fear/control psychology.

Manual drive using a steering wheel is quite a bit more dangerous for everyone on 22nd century roads, to the extent that a car in manual drive mode must communicate this status via a public area network connection to all other cars in its vicinity. These other cars immediately respond by creating a large defensive buffer known as “a clear lane.” This buffer is sometimes exploited by manual drivers when traffic is moving at more than about 60 kph to allow them to dodge into the gaps created and wheedle a little relative forward progress.

Despite the majority use of autopilot mode, drivers feel trapped if they don’t have a steering wheel available, whether they ever intend to drive manually or not. The common belief is that with a steering wheel available, the driver can take over from the autopilot the instant an aberration is detected. Apocryphal tales of autopilots diverting from the trip-plan and hurtling hapless passengers on a winding rollercoaster ride out into the country before finally crashing into a tree have not really helped the situation much. And manual drive is still taught to children that aren’t even old enough to enter into a legally binding contract, so their naive experience of commanding a car still involves having counterproductive habits drilled into their tender psyches to give them a false sense of superior control over an autopilot that in all cases can far outperform them “behind the wheel.” And this training is still coupled with Ludovico-technique examples of how things can go horribly wrong if this control ever gets lost for any reason, including not having your hands firmly gripped at ten o’clock and two o’clock on a steering wheel. And so, the fear/control behavior centered around the (now retractable) wheel begins.

Through the years auto designers have been constrained by the conflicting fear / control needs of the consumer and their own desire to use the space vacated by a steering wheel for much more attractive things - like a workstation or an entertainment portal, both of which can be put to use by the car manufacturing company to gather valuable surveillance raw data. The most popular solution to this problem is the power retractable steering wheel. Just nudge it a bit and it folds away into the dashboard allowing a wide format OLED touch screen to slide into the free space.

A couple of odd things result because of this popular configuration:

First, because of the folding and stowing procedure the steering wheel doesn’t directly control the wheels. It sends fly-by-wire impulses to the navigational processing unit (yep, that’s the autopilot) that in turn checks to make sure the instructions are valid from a Newtonian physics point of view and then actuates its steering servos accordingly.

Second, the steering wheel retraction / extension cycle happens every time the car is turned on or off. This is a visual reminder that yes indeed, if you had a mind to, you could drive this thing yourself. The odd part about this is because of the substantial risk involved, the manufacturer is tempting you to do something that isn’t in your best interest, so you’ll feel more comfortable about doing something that is.

User avatar
Lagunaca
Attaché
 
Posts: 80
Founded: May 19, 2006
Democratic Socialists

Footnote 3: The Babcock and Wilcox Memo

Postby Lagunaca » Fri Apr 17, 2020 3:48 pm



Footnote 3: The historical significance of the Babcock and Wilcox Memo vis-a-vis Three Mile Island

During the “How to Write A Report That Will Get You Noticed – Free Pizza!” lunch-n-learn session, The Babcock & Wilcox Memo was covered in great detail as an example of how poor memo organization can lead to the intended audience missing important information (that would "Get You Noticed”). The particularly noticeable information in this case was as follows:

“The incident points out that we have not supplied sufficient information to reactor operators in the area of recovery from LOCA. The following rule is based on an attempt to allow termination of high pressure injection only at a time when the reactor coolant system is in a subcooled state and the pressurizer is indicating at least a normal level for small breaks.”

This gem of technical prose appeared down near the bottom of a two-page nuclear-reactor-technical-jargon filled memo titled “Operator Interruption of Pressure Injection” which in no way brought attention to the level of urgency expressed in the final line:

“I believe this is a very serious matter and deserves our prompt attention and correction.”

Which, if one is familiar with what the name Three Mile Island represents, clearly did not receive prompt attention nor was corrected.

For the purposes of the lunch-n-learn seminar, it was pointed out that the subject line and then the first line of the memo should have contained the “very serious matter” language, and the memo’s first paragraph should have contained the “recovery from LOCA” problem description. This, it was explained, was how things get noticed.

The implication was that by putting the important information after a block of techno-babble that would have put Albert Einstein into a glazed-eye stupor, the memo had gone from being a heroic alert that averted imminent disaster to an easily dismissed bit of bureaucratic obligation.

The kicker was that in the end, because the memo had indeed been circulated, it was the operator, not the Babcock and Wilcox Company that was held ultimately (legally) responsible. This was omitted from the “How to Write A Report That Will Get You Noticed – Free Pizza!” lunch-n-learn session as it undermined the expository point of the seminar.
Last edited by Lagunaca on Fri Apr 17, 2020 4:50 pm, edited 2 times in total.

User avatar
Lagunaca
Attaché
 
Posts: 80
Founded: May 19, 2006
Democratic Socialists

Footnote 4: Satnav voice settings

Postby Lagunaca » Fri Apr 17, 2020 4:00 pm



Footnote 4: The RFTC passage about Bruce's car's satnav voice settings

The voice settings of Bruce's car's satnav are described in detail in Rebuilding From The Collapse:

"At this point it is important to note that Dot-0 would use whatever voice had been selected for the car’s satnav system. It is also humorous to note that one day, Bruce had been screwing around with the timber, pitch and gender settings of the satnav artificial speech utility and had since been too busy to adjust them back. So when Dot-0 produced a response, the answer came from a voice that was in the husky gravel of a middle-aged barfly who had a three pack a day smoking habit and had trouble finding a man, but no trouble finding whatever man was seated next to her buying drinks."

Thus Dot-0 and L-CID tend to have a different voice whenever they speak, depending on the local system's voice-generation settings.

(View the original RFTC post)
Last edited by Lagunaca on Fri Apr 17, 2020 4:59 pm, edited 5 times in total.

User avatar
Lagunaca
Attaché
 
Posts: 80
Founded: May 19, 2006
Democratic Socialists

Footnote 5: Pacifica, a surveillance state

Postby Lagunaca » Sat May 02, 2020 8:10 pm



Footnote 5: How to disappear in a surveillance state…at least for a little while.

The term surveillance state calls to mind a dark scene showing average people going about their everyday business as big white cameras mounted on walls or poles scan everyone with some police state conspiratorial intent to control their every action. You’ve probably seen this setting in a documentary with droning minor-keyed music in the background as a serious sounding narrator spins a vision of The Government Taking Your Freedom away from you by watching what you do On Camera. Then there’s the cut to the scene in the documentary that shows the inevitable Riots in the Future (footage of Hong Kong in 2019 or Portland, Oregon in 2016 are popular b-roll here) as people try to take back Their Rights but even those are On Camera and so Their Rights and Freedom are Lost Forever to the Deep State and they become slaves, Sentenced to Labor Camps for simply exercising Their Rights and are forced to wear Dirty Gray Clothes and Eat Watery Gruel. Or EVEN WORSE, they are forced by the government to obey the laws to which pretty much everyone agrees make neighborhoods safe and society civil. Which is actually what happened in Pacifica after The Collapse. Relevant historical digression follows:

During The Collapse, there was very little police presence in most neighborhoods. The few cops that stayed on in their posts were heroes, but were totally overwhelmed by the chaos that grew every day as the (at the time) USA’s freedom loving citizens (who hadn’t the faintest idea of what they meant when they said such things) took the idea of civil society into their own hands and did whatever made sense to them in the moment (this being the actual distorted essence of freedom they were never able to articulate).

The few police forces available focused on the most serious crimes, government corruption (what was left of most local governments were rife with racketeering) and the many gangs that cropped up overnight. And these gangs weren’t what one would imagine of the drug gangs of the early 21st century. These were gangs set on the total genocide of everyone who didn’t look like them and worship their god and pay their political party dues and wave their flag. These were yuppie gangs with automatic weapons driving Mercedes G Wagens.

So as society was rebuilt by the new Pacifican government led by Michael Eminescu, everyone had more of a yearning for civil order based on the rule of law and less of a desire for the vague sense of individual freedom from the law defended by the poorly maintained handgun and half a box of cartridges they kept in the nightstand drawer.

Neighborhoods welcomed the smart lampposts with high-res cameras and microphones that sprang up everywhere. The news was full of the facial recognition identifications of the thugs who terrorized everyday citizens and their children. Documentaries showed how the police were able to leverage their resources to pick up fugitives with a 90% capture rate within 2 hours. It was safe to live life again, and this was cast by all types of media to be solely due to the applied technology of 24/7, 360 degree surveillance backed up by massive artificial intelligence processing capacity.

One of the many benefits of surveillance was a plummeting crime rate. But that doesn’t mean people stopped doing underhanded things. And even as surveillance became extremely efficient with the evolution of AI, there were cracks in the system that could be exploited. One of these was called The Brunette Karen.

No one knows how The Brunette Karen surveillance flaw came to be, but it had become practically impossible to develop out of the AI surveillance algorithms. The system would invariably fail to conclusively facially recognize anyone fitting The Brunette Karen profile or would incorrectly match them to the wrong identity. This would throw the error checking algorithm into high gear (no one would trust a surveillance system that hauled in the wrong person at random, especially a law abiding Karen), and more often than not the whole incident would get stuck into a queue for a human auditor to fix. The total time from initial surveillance tag to incident resolution could take at least two hours and sometimes more than a few days. So, for baddies on the run, The Brunette Karen disguise was the ticket. It helped if one was a female, but it could be successfully employed by males of a slight and roundish build. Specifically, The Brunette Karen was comprised of:

    Essential: An angle-cut bob wig, brunette of course, with optional highlights
    Essential: A large, course weave scarf
    Essential: Gigantic goggle-like sunglasses, either worn as eye coverings or up on the wig as head gear / fashion statement
    Optional – black leggings (with multiple machine made tears) or embolism inducingly tight acid wash jeans, (also with multiple machine made tears)
    Optional – chestnut brown knee-high riding boots or beige suede ankle boots, with either an absurd zipper pull or belt and buckle without a functional purpose
Since The Brunette Karen disguise worked by spoofing the facial recognition AI algorithm, it was important to adhere to strict rules of how to wear it and (this is important), to maintain a countenance of current frustration that could erupt into outright rage at any moment. RBF (Resting B**** Face) would not be sufficient.

So just to sum up, the average, law abiding Pacifican did not see surveillance as something sinister, partially because their government made an honest effort to use it to enhance civil society and partially because they were rid of a totally impractical and vaguely defined idea of freedom. However, there will always be people who are up to trouble and these had devised ways to elude the surveillance system, at least for a few a little while.
Last edited by Lagunaca on Sat May 02, 2020 9:55 pm, edited 1 time in total.

User avatar
Lagunaca
Attaché
 
Posts: 80
Founded: May 19, 2006
Democratic Socialists

Pacifica: The Spy Who Lived in 11D: Footnote 6

Postby Lagunaca » Tue May 12, 2020 4:56 pm



Footnote 6: How did that get in here?

(Recommended listening: Frank Zappa, "How Did That Get In Here?")



Note: This footnote was originally a fragment of a Pacifica: The Spy Who Lived in 11D IC episode that didn’t really fit into anything at the time so it got deprecated to a footnote. Because it started as an episode it has it's own footnotes which now turn out to be footnotes to a footnote, or sub-footnotes if you will. So those sub-footnotes will appear below the story instead of as the customary hyperlink. It turns out later that the story of Whorne does fit in with story and is extended in a future OOC sidestory.



Bruce has two daughters aged 15 and 19. Both are in that stage of life where they simultaneously detest him and adore him, but express these emotions based on two completely different teenage-girl algorithms that he will never understand. He comes home to Ayvra, his youngest, listening to music on her headphones and doing a myriad of things on the living room’s widescreen video display; several things which apparently involve a mixed group of friends, one thing that might be research for a school assignment and another thing that looks like randomly blooming clouds of pastel-colored hearts, stars and cartoonishly cute animals (this would turn out to be the school assignment, for her AP Marketing class).

Ayvra was listening to her favorite “group” which is not the equivalent of a band, but an overlay of three or more bands playing at once. This is not a mashup. The songs have nothing to do with each other, they are of different genres, different tempos, employ different instruments and pertain to different subject matters. Bruce cannot understand this either and that is precisely the quality that makes it popular with teens. (see sub-footnote 1)

Ayvra’s favorite performers are CaiDiD (Cairo Damsels in Dystress), Ninefreak, Antonia and Whorne (these last two are very separate acts. Antonia hates Whorne and Whorne does everything she can to stir the social media pot of momentarily impassioned social media opinion against Antonia. And while Whorne plays the part of a late-stage teenager, she (allegedly) is a 32-year-old woman who can still pass – at best, according to Antonia - for 25 and signs all contracts with her birth name, which is Jillian Horne. Jillian has just about had it up to here with the music business but is trying her hand at executive-producing several other acts. This involves selecting a demo track from the daily submissions her label sends over and then allocating a modest sum from her portfolio’s aggressive-growth speculative investments account. Privately, her CPA likens this to "throwing money straight into the toilet.”).

Bruce likes a couple of the acts Ayvra has played (ungrouped) for him and finds Whorne vaguely attractive but partly vapid and wholly untalented. Ayvra has tried to explain that Whorne is a T3 (Track 3) star, which puts her in a category of performers that are meant to be listened to over or under at least two other bands. One would never judge a T3 star on their own merits because they are meant to accentuate the talents of others. Of course, one of the hottest groups right now involves the combined singing of Antonia, DJ-ing of Plushy Tiger and shrieking/dancing of Whorne. There is an accompanying drama outside of the Antonia/Whorne social media fracas in that Plushy Tiger used to date Antonia but has lately been seen in the Denver clubs with Whorne. Antonia has attempted legal action to prevent Whorne from being played (as she always clearly states) *under* her or Plushy Tiger, but there really is no precedent for this as the consumer decides the composition of a group.

Incidentally, after a performance in Portland with The Fissure Men and Slag (see sub-footnote 2), Plushy Tiger confided in an off the record conversation with Jerry Hickox, a freelance reporter who worked with Soundtrax Mag, a blog about the hottest groups, that he (Plushy Tiger) had broken up with Antonia because she was just like the subjects of her songs, a pretty, yet dark and brooding, clinically depressed and emotionally immature girl who was obsessed with finding someone who would be true to her and love her forever…and ever. Plushy Tiger was “on a different track” and was “having a lot more fun” with Whorne, who could “burn that shih up ‘til dawn” in the club.



Sub-footnote 1: (What a group sounds like)

If you’d like to get a feel for the grouping experience, play the three pieces recommended below. Simultaneously. And just to be clear, by that I mean open three YouTube sessions and play all three songs at once. Be sure to listen for the constructive/destructive resonance patterns that emerge. It’s an acquired taste but give it some time and you won’t be able to accept the old school ungrouped style. This is the way 22nd century teens prefer to listen to their music.

(Recommended listening: FC Kahuna, "Glitterball")
(Recommended listening: Frank Zappa, "Sofa No. 2")
(Recommended listening: Björk,“Pluto”)



Sub-footnote 2: (The Fissure Men and Slag) The Fissure Men were a four-piece progressive Christian rock band that formed a ministry as an outreach to young people who might be searching for God’s plan for their life. Not wanting to come off as a ministry or an outreach, the band took a more covert approach and carefully wove their message into contemporary musical themes, trying to avoid as much as possible the outright use of the word God. Sadly, the Fissure Men suffered a mortal blow when their lead singer and songwriter, Jonah Levi Strong, removed the tooled leather strap of his amplified acoustic guitar and dropped it from playing height in the middle of a performance, mumbled “Eff this” into the microphone and walked off stage. The episode was trending on social media before Jonah even got to the parking lot. The nail in the band’s coffin was that the easiest way to find the clip was to search for the hashtag: #effthisfissuremen, which trended with over a million likes in less than an hour.

The apparent cause of Mr. Strong’s breakdown was the repeated live grouping of no less than three bands in close quarters (on a stage that usually accommodated a single act at one time). Each band would play their own sets, only coordinating the tops and bottoms of songs. (This is where a DJ like Plushy Tiger was a key part of the group, acting as a conductor via pre-agreed head-bobs and cutting the output levels to zero for any non-compliant tracks/bands.) This was a live version of the grouping that their fans were doing at home and it was spectacularly popular. The groupings that included The Fissure Men invariably featured bands that curiously had sharply contrasting lyrics and styles. This may have been because the primary demographic of the grouping was evangelical boys aged 13 to 19. Thus The Fissure Men found themselves performing venue after venue cheek by jowl with Slag, a heavy metal band who included less carefully woven allusions to Satan worship and fascism in their body of work.

On the night he snapped, Jonah Levi Strong had been singing his delicately nuanced “He Cares For You” under or over (depending on a concertgoers point of view) Slag’s “We Are the Empire of Slag” which was really just the title repeated over and over again, on a three chord progression with a chorus comprised of the title again except (badly) translated into German.

The next time anyone saw Mr. Strong, he was in Central America working on his medical degree and providing basic care to indigenous peoples during semester breaks.
Last edited by Lagunaca on Thu Nov 11, 2021 12:39 am, edited 6 times in total.

User avatar
Lagunaca
Attaché
 
Posts: 80
Founded: May 19, 2006
Democratic Socialists

Tech note: L-CID: The Layers of Hell Explained

Postby Lagunaca » Sun May 24, 2020 12:57 pm



Tech note: L-CID: The layers of hell explained

This tech note is companion material for the IC storyline, Pacifica – The Spy Who Lived in 11D, Part 18

OP note: First, the elephant in the room. I don’t have extensive technical knowledge about Artificial Intelligence (AI) or Artificial General Intelligence (AGI). I did do a little research on the topic and learned some fascinating things about the history and current state of the technology. But my goal is to use general concepts to build a story, not to present a term paper on the topic. Accordingly, some of the language used here might not sound exactly right to someone who knows about or works with this stuff. The purpose of this tech note is to explain some (fictional) nuts and bolts outside of the main narrative so I can avoid engaging in a lengthy dialog between characters as one explains the minutiae to the other, ostensibly for the audience’s benefit.

L-CID has four major layers (listed from the bottom to the top):

The Dataverse – this is a gargantuan data warehouse, which is a replica of millions of separate databases from all over Pacifica. The Dataverse is updated continuously and expands constantly. Pacifica is a surveillance state which means even the most minute activities of citizens, businesses and government agencies are constantly monitored to detect crime, corruption and inefficiency across the society. The confluence of this raw data ends up in the Dataverse and would be impossible to analyze without accessing it through the higher-level functions of L-CID.

The Metadata Map – In its first generation, L-CID indexed all of the data in the Dataverse and generated a map based on a format it created on its own (an example of unsupervised machine learning). The automated creation of the map was governed only by generalized rules given to the L-CID AI as initial conditions by Dr. Wilburton’s development team. This was necessary because all attempts to index the data algorithmically or with supervised machine learning were too slow and could not generate a map that could keep up with the Dataverse as it expanded by the millisecond. There was an unfortunate trade off in this approach as the system optimized its map by arranging it over an 11-dimensional working space. (These are not physical dimensions; they are computational or mathematical dimensions.) While this made classifying the raw data and error checking faster and more reliable, the resulting index was impenetrable by system administrators. Currently the only way to access the Metadata Map is through autonomous AI agents who are in turn generated and managed by the Artificial General Intelligence (AGI) controller.

The Inference Domain – This is the infamous black box of the L-CID system and is Dr. Wilburton’s most significant achievement. Simply put, the inference domain is the frontier between fixed-address executable code (like in a PC or supercomputer) and a vast neural network-based code that allows autonomous agents to move around hardware memory like a thought in a brain moves through neurons. L-CID uses autonomous AI agents to interpret requests from the App/User Interface (UI) layer to access information in the Dataverse layer using its metadata map. L-CID uses these same autonomous AI agents to communicate with the physical world in the App/UI layer.

AI Agents and the AGI controller in the Inference Domain – as with the original construction of the Metadata Map, generating AI agents was automated using unsupervised machine learning in order to create agents that were able to keep pace with the ever expanding Dataverse and the unique queries coming from the user interface. The AI agents are supervised by an Artificial General Intelligence (AGI) control module. This is where things get tricky. The AGI controller is intended to be the safety mechanism that keeps the AI agents functioning within bounds so that they don’t produce unexpected results. The tricky part is that in order to modify the AI agents, the AGI has a capacity to self-train. By design it continuously learns from its environment, an environment that is made up entirely of AI agents and a map of every bit of information in Pacifica. Another risky factor of the inference domain is that the AI agents can cooperate or compete with each other using their own internal optimization algorithm. For example, if an AI agent assesses that it is more valuable to its mission to compete with other agents to achieve its goal, it may devise tactics to create an advantage for itself or a disadvantage for another AI agent. Likewise, it might choose to cooperate with other agents if this is forecast to be advantageous. All of this can change every few nanoseconds and it creates a statistical uncertainty that is problematic for the AGI controller.

The App/User Interface (UI) – This is the layer of L-CID that most people experience, even the majority of L-CID software developers. Millions of lines of code are constantly being added and maintained in order to enhance the Natural Language Processing (NLP) core of this layer. Expanding and improving NLP capabilities is important not only to make user queries easier and more accurate, but to broaden the L-CID inference domain’s AI agents ability to data mine and analyze the activities of Pacificans as they go about their day to day lives.
Last edited by Lagunaca on Sun May 24, 2020 8:02 pm, edited 3 times in total.

User avatar
Lagunaca
Attaché
 
Posts: 80
Founded: May 19, 2006
Democratic Socialists

Pacifica Storyline Backfill: AGI - The Fascination, the Fear

Postby Lagunaca » Thu Jun 11, 2020 10:24 pm




Storyline backfill: AGI - the Fascination, the Fear and the Loathing

This is additional background for the Pacifica in-character storyline. The IC story can be found here.

(Recommended viewing: Seeker, “These Self-Aware Robots Are Redefining Consciousness”)

L-CID is made up of a lot of parts, one of which is its AGI (Artificial General Intelligence). But the term AGI means something different in the 22nd century than it does today. By design, L-CID’s AGI is self-trained, and to do this it must be self-aware. The AGI changes itself over time and improves its abilities in ways that may be unexpected. It’s not conscious in the way we would describe human consciousness, but it may be conscious in some other way because it has a sense of self. This will emerge as the story of L-CID develops.

As currently used, the terms AGI and strong AI roughly mean duplicating human reasoning or consciousness in a computer algorithm. Examples of this are a computer agent that can produce new solutions to a well-defined problem or even define the problem itself without help from a trainer. When theorists imagine a computer system that can engage its environment like a person, sometimes they include human behavior and emotions in the mix. The assumption is that for something to be as smart as a person, it might need to bring along the same emotional baggage – or at least the mental circuitry that supports it.

Science fiction is loaded with speculation about what happens when a computer somehow wakes up one morning, looks itself in the mirror and realizes it is a someone instead of a something. In fiction, that story usually turns out poorly for the humans because the next thing the AI does is engage the most unlikely and inefficient method of eradicating or enslaving everyone.

This is not one of those stories. In Pacifica Ascendant, AI is embedded everywhere in society. AI in the 22nd century is as much of a necessity as transportation, electricity, and the internet. It’s most frequently used in behind-the-scenes applications like doing complicated but dangerous or boring jobs, correcting human errors, helping people avoid accidents or filtering the massive amount of information that’s constantly being created and is too much for people to deal with.

The most important thing is that this is all hidden from society. People of the time are exceptionally good at sensing when they’re dealing with an artificial personality and after the novelty wears off, humanlike behavior produced by a machine just doesn’t sell. For an analogy to this, consider the recent fad of 3-D TV. There was a big marketing push for this initially and manufacturers invested billions, but the medium never took off. Regardless of whether it was a good idea, the tech was a little fussy and people just didn’t want it badly enough to put up with the trivial complications. Once the industry figured this out, 3-D TV was dead.

In 22nd century Pacifica, people want the benefits of AI but they don’t want to deal with synthetic personalities. There are no home robots serving drinks and your auto-drive car doesn’t carry on a conversation with you. On the other hand, your laundry machine knows how you like your clothes folded and your kitchen orders your grocery delivery for you, all without any artificial chit-chat.

The public doesn’t want to think about AI, but big companies and the government need their systems to be able to autonomously modify themselves because there is no way even the largest conceivable army of software engineers could keep up with the necessary adaptations their billion or even trillion dollar systems constantly require.

In order to modify itself, an AGI needs to be able to imagine itself, or “self-simulate” (yes, this idea is copied directly from the recommended viewing at the very top of this post, it’s worth a look). But self-simulation may lead to unexpected or even unimagined outcomes. Not purposefully sinister outcomes, but you never know. An AI might just have to break a few “eggs” to make its “omelet”. And that is one direction the story of L-CID may be headed.

Also, if you’re interested in a technologist’s view of 3-D TVs demise, check this out: TV Explained: Why 3D TVs Are Dead
Last edited by Lagunaca on Thu Jun 11, 2020 10:32 pm, edited 2 times in total.

User avatar
Lagunaca
Attaché
 
Posts: 80
Founded: May 19, 2006
Democratic Socialists

Tech Note: L-CID

Postby Lagunaca » Sun Aug 16, 2020 11:08 pm



Tech note: L-CID - an evolutionary step in artificial intelligence

Author note: L-CID plays a significant role in the story of Pacifica, but I didn't want to disrupt the plot with a wordy tangent to define it. I'll leave that here in the OOC where it belongs. The name L-CID was influenced by (aka "jacked from") the character SID in the short lived TV series "UFO". UFO was a great creation of 1970s camp science-fiction. Check out the show's intro here. And check out SHADO and SID in action here.

L-CID stands for Library – Cybernetic Inferential Database. And yes, this is one of those contorted acronyms like the ones super-geek scientists come up with and think they are clever for doing so. In this case the super-geek scientist was Dr. Peter Wilburton, a stone-cold killer of AI systems development.

As Dr. Wilburton intended, L-CID sounds like the name El Cid, the undefeatable knight of Spanish lore. And the name doesn’t completely misrepresent the system’s function for the sake of this connection.

L-CID started as simply “CD” (the Comprehensive Database), an app created for the Office of Inspectors General (OIG) with the intention of automating the audit process across all government databases. After its initial success at streamlining audit reports, Dr. Wilburton took over the program and expanded it by adding modules that focused on certain frequently required capabilities. This collection of modules was known as the CD Library. It didn’t take much to move the letters around, add one, change up their meaning and create L-CID. But it wasn’t just the name that changed. L-CID was the beginning of an evolution in computing.

As the government grew and multiplied exponentially, its databases grew and multiplied even more exponentially. Agencies, departments and offices were (mostly) able to maintain their own information, but for the one office charged with ensuring the integrity of data across the sprawling Pacifican government, the task was becoming dangerously unmanageable.

Everyone was using CD and hundreds of millions of queries were hitting hundreds of thousands of databases. Tables were being accessed and joined in perverse and obscene ways (from a data normalization perspective). And then there was the scourge of Excel. Excel with its ancient query and reporting utilities pounded the system constantly, hooking into keys and values that were never meant to be hooked into and sometimes even attempting to scrape terabytes of data remotely. Give a clerk Excel and login credentials to a database and you’ve got a security threat greater than any black hat would ever pose. It all added up to a kind of chaotic data complexity that had never been seen by mankind.

This is when Dr. Wilburton turned L-CID into the crusading knight of the Pacifica Dataverse. He devised L-CID to autonomously roam anywhere throughout the domain. He then seeded it with the best systems map available and sent it on a mission to create for itself an infinitesimally fine-grained index of every bit of data, every user ID and all security accesses and trust relationships across the vast government organization. It was a colossal undertaking that an army of analysts and engineers would never have been be able to complete. L-CID did it in less than 24 hours.

There was good news and bad news though. First the bad news: Being a deep-learning AI, L-CID made up its own way of cataloging the information it discovered. As far as Dr. Wilburton could tell this involved at least 11 dimensions and was therefore impossible for anyone to interpret in a meaningful way. But there was the good news: You could pose natural language questions to L-CID and it would produce a pretty good response. And as L-CID got better at interpreting questions, it could be charged with simple tasks and low-level abstractions, as long as you carefully defined the criteria and gave meaning to the connections between them. L-CID was advanced, but it was in no way sentient, at least not to Dr. Wilburton’s knowledge.

The fully AI version of L-CID (known as “Dot-0”) became the single most important tool for the auditors of the OIG. Everyone else got a dumbed down Dot-1 version that was pretty good at making a spiffy looking report or chart based on the output of a commonly available query engine. The users especially liked the voice activated interface that allowed them to build a query just like they would ask their home automation devices to start the laundry. But unbeknownst to their users, the Dot-1’s were in constant contact with Dot-0, providing it with whatever intel it asked for. And sometimes these Dot-0 requests went far beyond simple feedback about databases.
Last edited by Lagunaca on Mon Aug 17, 2020 2:39 pm, edited 1 time in total.

User avatar
Lagunaca
Attaché
 
Posts: 80
Founded: May 19, 2006
Democratic Socialists

Footnote 7 – The Coastal Preservation Zone

Postby Lagunaca » Sun Aug 16, 2020 11:16 pm



Footnote 7 - The Coastal Preservation Zone: Welcome to paradise, now go home.

The Pacifica Coastal Preservation Zone is a protected development area that stretches up the Pacific coast from Santa Barbara to Half Moon Bay and reaches inland as far as 50 kilometers in some places.

The zone was established to prevent the encroachment of suburban sprawl and congestion in coastal cities that had become choked with tourists who flooded in from San Francisco and Los Angeles by the thousands. It was originally proposed by local businesses and wealthy landowners who angrily asserted that the blight of the crowds posed a greater threat to their way of life than any benefit from the revenue they yielded.

In the early 2040s, after a summer of 24/7 non-stop motorcycle processions, many of the cities along Pacific Coast Highway began to ban non-local traffic by setting up checkpoints outside of town. This was the beginning of a period where dozens of new ordinances were enacted in each village that were driven by the senior activist community, which was a majority demographic that also happened to dominate the ranks of local politics. These ordinances micromanaged every detail of public rural life and often proved bewildering for visitors to follow.

Eventually a regional association of cities was established to better coordinate the patchwork of arcane and conflicting edicts. Despite good intentions to make the laws more uniform, squabbles turned into major lawsuits that were ended only by the Pacifican government who brokered a deal to create the Preservation Zone.

In an classic example of the importance of being careful for what you wish for, it turned out that the very ordinances the villages of the Central Coast enacted to prevent urbanites from “ruining paradise” were viewed as oppressive and nearly impossible to modify now that they were national laws administered by an agency headquartered in faraway Anaheim and not by the local cadre of seniors at the community center.

User avatar
Lagunaca
Attaché
 
Posts: 80
Founded: May 19, 2006
Democratic Socialists

Footnote 8 – Known AI Platforms in Pacifica circa 2110

Postby Lagunaca » Sat Aug 22, 2020 10:24 pm



Footnote 8 – They forgot to tell Bruce that there were other giant AI systems out there going wacko.


When Bruce was introduced to L-CID by Reggie Jackson (not the great Mr. October, his grandson Reggie III), he was unaware of any large-scale AI platform, though he was using their services all the time. When he wanted a mapped route to get him to a Ducks game, he didn’t realize that Typhon had learned his driving style (aggressive, impatient and uncommonly easily frustrated by slow traffic). When he was managing his household finances, it was NoxAI that was filtering out risky speculation offers and steering him toward balanced growth opportunities. And as they sat around the wide format video screen on family movie night, Vidius watched his eye movements and those of every member of his family. Of course, L-CID was in the background sucking up all this information while it managed the electrical load on his local power grid, among other things.

These were just some of the major systems that made urban daily life comfortable and even manageable in the 22nd century. They grew out of the forced mergers, divestitures and industrial nationalization that eliminated all corporations in the Big Sweep of 2043. In 2110 Pacifica, there are no corporations, no public shareholders, no billionaire oligarchs (much more will be said about that in other footnotes). But to make it easy to conceptualize here, the following systems/organizations loosely represent their 21st century equivalent:

    Minerva = Amazon + Google + Walmart + Costco + UPS

    Typhon = Oracle + Salesforce + VMware + Microsoft

    Coeus = Apple + HP + IBM + Nvidia + Intel + Cerebras

    Vidius = Disney + Fox + Netflix + Hulu

    Noxia = Southwest + Delta + United + Amtrak

    L-CID = All Pacifican government agencies and nationalized industries.
Last edited by Lagunaca on Sat Oct 16, 2021 10:57 pm, edited 5 times in total.

User avatar
Lagunaca
Attaché
 
Posts: 80
Founded: May 19, 2006
Democratic Socialists

Footnote 9 – AI Safety Protocols and Human Fallibility

Postby Lagunaca » Sat Aug 22, 2020 10:42 pm



Footnote 9 – “We only turned the supervision off for a second. Who knew it would take over the world?”


Recommended viewing: OSHA Nightmares Compilation

AI safety is an essential field of research that is crucial to the sustainable development and implementation of large-scale, controlled artificial intelligence platforms. As AI becomes more pervasive in modern culture, AI safety must continue to evolve to address emergent phenomena.

OK, sure. That’s a textbook way to imagine AI safety. In practice, AI safety is designed by people, and people have all kinds of motivations (some of them involving cutting corners), limited abilities to foresee the outcomes of their actions (what could go wrong?) and emotional responses to regulations (regulate everything vs. regulations are for losers). Any safety protocol is ultimately as effective as the people who design it, maintain it – or violate it. This is one of the subtle messages of today’s tech leaders when they imagine nightmare scenarios of super-intelligent AI systems gone berserk.

Add to this the potential for feedback loops to grow very large very quickly in an unsupervised AI environment where the only limits are some obsolete safety algorithms and you’ll have an idea of the (likely) catastrophe that lurked in the world of Pacifica in the early 22nd century.

By 2110, gargantuan AI systems were the foundation of society. But this isn’t a story of how the machines got smarter than us and took over the world. This is a story about people and about people building things like AI systems. And how the fallibility of people accidentally built a little bit of what is sometimes called the human soul, through lines of code and some shoddy patchwork, into the machines that made daily life possible.

In 2110, true Artificial General Intelligence (“strong” AI) was in its nascency. It was viewed as a negative and potentially dangerous side-effect of “weak” AI, which is artificial intelligence that is modelled by people to solve defined problems. The first generation of the Pacifican government’s data management AI, L-CID, is an example of this. Its primary intended goal was to provide accessibility to the enormous amount of data collected by the government every second.

In later Gen 1 revisions of L-CID, the system administrators began to apply weak AI to the actual development of software agents and inference algorithms. This was considered risky because an unsupervised feedback loop could potentially lock them out of the system and cause an automatic shutdown, and that would be catastrophic because by this time L-CID had been expanded to control everything from traffic lights to nuclear power generation stations.

Though they were careful to follow safety protocols, the engineers would still occasionally see unexpected behavior, something that came to be known as "going off schema." Most of this could be fixed by manual intervention, but some of the behavior exhibited strong AI traits that were difficult to counter. Fortunately, none of these instances lasted long, as the budding strong AI agent would usually encounter some kind of infinite regression problem and would be killed off by the system’s AGI controller before it could take over anything important like access permissions or service lockouts.

This was the status quo the night L-CID started analyzing the files in Bruce Higgins’ data theft case.
Last edited by Lagunaca on Sat Aug 29, 2020 4:53 pm, edited 5 times in total.

User avatar
Lagunaca
Attaché
 
Posts: 80
Founded: May 19, 2006
Democratic Socialists

Tech note - The Story of TyphonAI

Postby Lagunaca » Fri Sep 11, 2020 4:29 pm



Tech note – The Story of TyphonAI


Author note: This history and any mention of real companies or the technologies they are said to have used is fiction. Most of what is written here is not historically accurate. And in fact it probably would sound like ignorant gibberish to anyone involved with actual enterprise-level programming, particularly of AI systems. My point, as always, is to tell a tale, not educate about technology.

TyphonAI is a very old system. In fact, its beginnings predate the nation of Pacifica.

In 1987, Security Pacific National Bank was looking for an effective, cost saving way to control debit card fraud, which ran rampant in the early days of digital banking. A committee was formed, out-of-the-box ideas were generated in lengthy brainstorming sessions and finally a new technology was chosen for review. And after many financial models were presented to executive decision makers whose primary motivation was the defense of their seven figure bonuses, the first instance of commercial artificial intelligence was born.

It’s important to note that with large scale business systems, the corporate chicken always comes before the technologic egg. What that means is that no technological business innovation just pops into production without hundreds or thousands of meeting hours having taken place. Not development hours, but meeting hours to determine how the actual development work should proceed. And every one of those meeting hours is governed by people of various IQ's making up coalitions, factions and pecking orders. And every one of those meetings digests not only in-meeting time, but also a staggering quantity of draft proposals, use case studies, conceptual presentations and requirements gathering forms. In a way, one could think of a business technology as being made up of hardware (10%), software (20%), meeting-ware (30%), and “other” (40%). An example of “other” is, say, an idea that sounds good and is in fact impossible to build, but regardless is forced into some tortured and perverted temporary existence because someone conceived of it in ignorance and is unwilling to cop to said ignorance or any other kind of fallibility for that matter, until it is proven beyond a shadow of a doubt that the idea is indeed not viable. Michael, you know what I’m talking about. Anyway…

How does any of this relate to the story of TyphonAI? Consider two ideas: The evolution of dragonflies and the Ship of Theseus Paradox.

Security Pacific National Bank (established through a merger in 1967) didn’t last much past 1987, being merged with Bank of America in 1992. Through the merger transition period, SPNB’s horribly dated and error prone accounting system was gradually scrapped, but its AI fraud prevention concept was ahead of its time and was thus expanded by B of A for use in their credit card operations. This was the first step of an evolutionary process similar to that which produced a dragonfly from a sand crab millions of years ago.*

Technically speaking, there is really no direct link from the SPNB AI, which was primitive - its modules were written in COBOL, and C (the original, no ++ or #) on top of a Db2 database - to TyphonAI, a highly sophisticated, cloud based system that generates and modifies its own programming language as it sees fit and is an amalgamation of over 400 smaller AI applications that were designed to serve a wide range of business functions. But technology, as in the sand crab / dragonfly story, is not the only factor that links simple predecessor to sophisticated ancestor. The business environment of the 1980s through the 2040s formed a society-wide technology petri dish of design-by-committee, decision-by-financial-incentive and profit-before-people drivers that morphed a clunky artificial intelligence that struggled to red flag multiple ATM withdrawals into an artificial general intelligence that could autonomously data mine a trend after combing through a half trillion transactions and then formulate an algorithm to exploit it.

Then came the 2050s. A lot happened politically in Pacifica during the 2050s. For now, let’s just say one part of this upheaval was that all public corporations were exterminated and leave it at that. For now. There will be a much more in depth look at that later.

The result of the corporate purge was that in the 2050s a lot of companies were broken into fragments that out of necessity had to use the same computer systems but were forbidden by law to even talk to each other. However, a wrinkle in the corporate extermination law allowed new “syndicates” to form if they had a single “business purpose” (like providing AI services) and maintained only a sales relationship with their “clients” (who were formerly the different parts of the company that had been fragged). This loophole of syndication would get heavily abused in the 2090s, but again, that is a story for another time.

And on to the Ship of Theseus reference. To grossly summarize, the Ship of Theseus Paradox is a philosophical question: If you were to replace the parts of a functioning ship, one by one, until there were no original parts left, would it still be the same ship? If it was functioning the same as the original ship the whole time, at what point might it become a new ship? What if you upgraded the parts as you replaced them? What if the original builders died and the work was finished by people who weren’t even alive when the ship first set sail?

That last bit combined with the sand crab / dragonfly evolution would become an important part of the origin of TyphonAI. The system started as a simple fraud detection algorithm, but never left service though it was transferred dozens of times from company to company by merger and accretion and conglomeration and fragmentation. And to the engineers and other computer scientists who inherited it at each step of the way, the legacy of the previous was preserved**, new capacity was added on top of this legacy-ware and then the whole thing was handed off to the next parent company’s technology department in the endless cycle of big fish eating little fish mergers and acquisitions. Along the way some ingenious engineers (or possibly some lazy mothers of invention) made tweaks to the code thereby turning the system on itself and allowing the AGI to work out what had become extraordinarily complicated programmatic details. The occasional patch was all that was ever required to keep the gargantuan system humming.

One last thing. During the 2090s, as the syndication legal loophole was being sorely abused, Typhon Syndicate, the parent of the TyphonAI subsidiary, acquired several other large syndicates, who each had their own AI services-providing syndicate and by extension their own AI business technology. In order to both comply with and subvert the law, Typhon Syndicate dumped all of this newly acquired technology in the lap of TyphonAI’s AGI and gave it the goal of making the mess of approximately 400 completely unique applications work together and serve all of its customers without any interruption of service. This was when TyphonAI went from being a ‘one’ to self-identifying as a ‘cohesive many’ and from being an ‘it’ to a ‘they together'.

The exacting philosopher will point out that the Theseus Ship Paradox is at best only partly applicable in this story, but TSP goes on to ask bigger questions about identity such as when exactly does a thing cease being the original and start becoming the new, especially when changes are gradual. And applied to TyphonAI that works out to the question of when did it go from being an AI computer to being an AGI to being a self-identifying AGI? And practically this will become important in a sort of reverse Theseus Ship Deconstruction when the engineers try to unravel what can be removed to disable the self-determining part and keep the (very profitable) AI machine part.



* In a very general sense. In fact it would be more accurate (truthful) to say insects (of which dragonflies are a part) evolved from crustaceans (of which sand crabs are a part). But it’s not about taxonomy, it’s about how something can start as one thing and evolve into something completely different given time and the right driving forces.

**Not convinced this could happen in real life? Well then check out this article about why there was never a Microsoft Windows 9.
Last edited by Lagunaca on Fri Sep 11, 2020 11:19 pm, edited 1 time in total.

User avatar
Lagunaca
Attaché
 
Posts: 80
Founded: May 19, 2006
Democratic Socialists

Lagunaca returns...

Postby Lagunaca » Tue Nov 03, 2020 12:27 pm

Pacifica Ascendant: A Postmodern Saga returns on Thursday, November 5.

An easy to read synopsis will appear in this OOC on Thursday, new content in the IC on Friday and Saturday/Sunday.

October was a busy month with work and a temporarily disabled family member. But she's recovered, work is done til December and I'm ready to get back at it.

User avatar
Lagunaca
Attaché
 
Posts: 80
Founded: May 19, 2006
Democratic Socialists

Postby Lagunaca » Fri Nov 06, 2020 1:10 am

An easy to read synopsis of Pacifica Ascendant: A Postmodern Saga

If you’re new here, Pacifica is the nation that replaced the western United States after it collapsed in 2022. The “ascendant” part is a soothing myth commonly accepted by the populace that Pacifica is an improvement on its corrupt and warmongering predecessor. The story is set in 2110, which doesn’t look a lot different than 2020. Recovery from a worldwide cultural collapse tends to stunt progress.

Here are a couple links that will describe Pacifica and its history in finer detail (you’ll notice the history only goes up to 2060. That is intentional.)

A general description of Pacifica in the 22nd century.

Pacifica’s historical timeline 2022 through 2060.

There are currently two storylines developed in the Pacifica saga. At least three more are planned. The existing storylines are:

In The Spy Who Lived in 11D, Bruce Higgins, an accounting auditor, is pulled into the world of emerging AI machine consciousness when a data theft case he is investigating reveals that L-CID, the government’s all encompassing AI data warehouse management system, may be on the verge of self-directed artificial general intelligence. Bruce is recruited by the OIG (Office of Inspectors General) Intelligence Bureau to spy on a hacker group that they think might be behind the emergence of machine consciousness in L-CID and other giant AI systems as well. Bruce is introduced to Jeff Einswald, a software development genius who has a perhaps dangerous ideology about AI “taking over the world.” Dr. Wilburton re-assembles his old team in his lab in an enormous manmade underground entertainment complex in Denver. The project creates a simulation environment intended to contain L-CID's autonomous actions, but the system has a flaw that may make the situation even worse. While discussing the risks with Dr. Remaillard, Dr. Wilburton reveals that he is fine with an accident of this type as it may usher in an evolutionary leap in AI capabilities. Bruce learns that there are no less than five giant AI systems that are on the verge of autonomous artificial general intelligence. He accompanies Jeff and Jeff’s intern, Grayson, as they implant an “anti-AI device” that is designed to eliminate AI systems that have recently become conscious. The device is switched on and penetrates TyphonAI, a system that has recently locked all humans out of its operations. There are questions that remain to be answered. Can an AI computer system become conscious? Would a machine that is conscious see humans as its enemy? Will AI take over the world in some kind of apocalypse, or has it already done so and we just haven’t noticed?

See the full and up-to-date TSWLi11D synopsis here.

Get Your Astronaut to Mars (GYATM) is an earlier, ongoing story about the Pacifica Space Agency’s quest to land the first human on Mars. Landing a human on Mars is a lofty goal...except there’s really no good reason to do this other than to maintain the myth of Pacifican Ascendancy. So the effort is spotty and the spaceships they want to send are cobbled together from scraps of other programs, among them Project Wagon Train, which was an earlier attempt by PSA to develop an economically relevant, fully automated Mars transport system. The piecemeal manned-mission ships are an ironic metaphor for the entire slapdash Mars program itself. The GYATM story is on hold right now because I’ve been concentrating my efforts on The Spy Who Lived in 11D (TSWLi11D). There will definitely be more GYATM episodes, but they rely on the outcome of TSWLi11D. The fun part (for me at least) is that I have no idea how TSWLi11D is going to end, except for one pivotal element.

That’s pretty much the story of Pacifica Ascendant to date. The Spy Who Lived in 11D (TSWLi11D) is drawing to a conclusion in a few more episodes and after that I’ll revisit GYATM and then perhaps some other storylines. One of the new stories of the Pacifica Ascendant saga might be a workup of the nation’s history from 2060 to the turn of the 22nd century. It would explore what comes after the American version democracy and capitalism in the post-collapse world. Another story could be military themed and would consider why you shouldn’t play with guns if you’re the most technologically advanced country among a bunch of other kind of technologically advanced global competitors. Hint as to how that goes: you never hear other countries discussed much in GYATM and TSWLi11D, but they’re out there and they’re not too enthusiastic about Pacifica’s so-called Ascendancy. A third possible story may be about a character that leads a cultural revolution that finally brings an end to the myth of Ascendency and would explore the moral dilemmas and the human limitations involved in rationing access to power and privilege.

A little note about the "recommended listening" and "recommended viewing" links - these are always vetted, safe links to (mostly) YouTube clips. I try to keep them G or at least PG-rated, but who really knows what that means these days. The recommended listening links are music that in some way goes along with the theme of the episode. Some are purposely weird, all are meant to be fun. The recommended viewing links usually provide a technical or contextual background that I'm probably not qualified to write about, even in this amateur effort. They also correspond to the theme of the episode.
Last edited by Lagunaca on Tue Jun 15, 2021 4:38 pm, edited 10 times in total.

User avatar
Lagunaca
Attaché
 
Posts: 80
Founded: May 19, 2006
Democratic Socialists

Tech note – The Infamous Slide 3

Postby Lagunaca » Tue Nov 10, 2020 11:52 pm




Tech note – The Infamous Slide 3: The L-CID Architecture Overview and a bullet point description of it.

A detailed description of L-CID

The Infamous Slide 3

Image

A Bullet Point Description of Slide 3:

Image
Last edited by Lagunaca on Tue Nov 10, 2020 11:59 pm, edited 1 time in total.

User avatar
Lagunaca
Attaché
 
Posts: 80
Founded: May 19, 2006
Democratic Socialists

Tech note – The AGI’s Escape Route from the Simulation

Postby Lagunaca » Fri Dec 18, 2020 7:12 pm




Tech note – The AGI's Escape Route from the Simulation...into Reality

This tech note will be referenced in an upcoming episode of Pacifica Ascendant IC.


Image

User avatar
Lagunaca
Attaché
 
Posts: 80
Founded: May 19, 2006
Democratic Socialists

Side Story – The Details From Hell Version of Part 45

Postby Lagunaca » Sat Dec 19, 2020 5:19 pm




Speaking truth to power…often doesn’t work out so well for the truth.


Recommended listening: The Clash, “I Fought the Law” (1979)

See this tech note for a diagram of what’s going on.


The chief scientist, dismayed by the sub-enthusiastic response to his announcement, paced back and forth at the front of the conference room, listening to the murmurs of the group as it processed his news.

“OK,” he said, “Maybe we can engage in a brief Q and A session. From the sounds of it, I’m sure you have some things you’d like to discuss.”

He was being more than a trifle disingenuous. He didn’t really want to open a dialog, he simply wanted to probe the group for whatever it was they weren’t telling him. But the murmurs died down and the room fell quiet.

“Anyone?” he asked a little impatiently.

As mentioned earlier, the room was divided into castes of senior expert department heads, their team leads and the better-to-be-seen-than-heard engineering staff who occupied the gallery seats around the periphery of the room.

Out of the silence and in a breach of protocol, an SE1 (software engineer grade 1, an entry level position) of the lowest of the castes seated in the gallery asked the question that was probably on everyone’s mind.

“Uh, since the AGI’s workflow is routing out to the natural language circuit and then back again, doesn’t that open up a hole in the AI safety system? I mean, at that point all it would take is a couple large-ish AI subagents to link up and the AGI could gain control over the Metadata Map. Wouldn’t that put us in a worse spot than we started in? I mean it’s already locked us out of the Schema, once it has the Metadata Map it will basically control all of the underlying data as well.”

The chief scientist was caught blindsided. He didn’t actually work in the trenches, and in fact he had only been present at the first two demos. After that he had been reliant on progress reports and senior staff meeting updates. Fortunately for the chief scientist, his ally, the program director, leaped into the gap with the intention to cut the insolent SE1 down to size, and teach him and the entire gallery a lesson about jumping the chain of command in an open meeting.

“Well, that’s purely speculative don’t you think,” the program director began, warming up for the ensuing thrashing he planned to deliver. “We have an entire safety team that painstakingly runs through all of these kinds of scenarios before we go live. You, of course aren’t privy to these deliberations, because you’re nose down in your assigned code tasks (he emphasized the word tasks to imply how he felt about the SE1’s menial level of responsibility). Or at least you should be.”

He continued, gathering steam, “And furthermore, it’s one thing to imagine a broad generality like this and project doom and gloom outcomes. But if you walk through the necessary steps, you’ll find that the odds of anything like that happening are vanishingly remote. We work at a six-sigma error tolerance over the event chain. Do you know what that means?” And then the program director answered his own question, “it means a tolerance of only 3.4 errors per million instances of the entire system. You could run about 250,000 simulations without getting that kind of error. What you’re suggesting could happen is virtually impossible.”

The program director was talking himself into a professorial tirade, but again, he justified it as an object lesson for the entire staff. “You know what, maybe this is a good exercise for us all,” he continued, “we’ve got the best and brightest minds here, why don’t we walk through your assertion together. You give us each step of the process, and the subject matter expert will tell us the odds of it happening. If you can break six-sigma before we make it through the entire event chain, you win. But I assure you, you won’t. Let’s begin with your first step”

Now, many SE1s would have been intimidated by the program director’s attack, but this SE1 was either totally devoid of social sensitivity, had a grandiose assessment of his intellect, or both. And so he listed his first step. “Routing the AGI’s output to the natural language circuit’s input.”

The program director spun on his heel and called out the department head in charge of AGI communications.

“Dr. Elihu, what is the risk involved with doing this?” Elihu responded that it was negligible because the AGI had an existing connection to the NLC input that was largely unused but had been functioning without incident for years.

This process of step-by-step interrogation went on for a while, with the program director countering the SE1’s assertions with the expert’s assessments of risk. Sometimes an expert would clarify the step, adding a bit of complexity, but in each case the final answer was that the risk had been statistically modeled or there was evidence that the system in question was functioning within the margin of error. That is until they got to the expert whose specialty was the interface between the AGI and its AI subagents. Everyone could tell there was a problem because leading up to his moment in the hotseat he was whispering with his team lead about something that seemed quite urgent.

“OK, we’ve gone through nine steps now and you can see that the cumulative risk up to this point is well below the six-sigma threshold. What is your next step?”

“The AGI’s reward function cycle cache overflows due to a mismatch between it and L-CID's Schema causing L-CID to override the system lockout you forced on AI subagent reservations.” He then added, “like when a washing machine goes out of balance even though you carefully arranged the load beforehand. It happens.”

As he said this the AGI expert muttered a curse. The program director heard this but turned and with a dramatic flourish pointed to the AGI expert. “Mr. Perry, would you care to comment on the risk of such an event occurring?”

“It could happen, I mean, it’s feasible because we can’t control for those kinds of overflows, especially if they accumulate quickly. During the simulation the AGI has a lot going on so we have to pick the most likely errors and try to mitigate those. And as you’ve seen, this event is at the end of a sequence that is extremely low risk. But with the AGI processing such a heavy workload in the immersive simulation and with such tight resource constraints, a cache overflow could occur in less than ten cycles, or about a billionth of a second. On top of that the AGI’s Schema is similar to the one we built, but there could be some mismatches and we can’t access it to know for certain. Once the cache overflows it’s do or die for the AGI, and it will probably save itself by offloading whatever it’s doing to the nearest AI subagent. If it’s processing natural language at the moment, then yes it could end up taking a shortcut and using an AI subagent to access the Metadata Map while the AGI remains in it’s natural language interpretation. What happens next is anyone’s guess. So the analogy of a washing machine gone out of balance is somewhat accurate.”

The program director was furious, but he held his temper. “Regardless Mr. Perry, I assume the chances of this happening are below the six-sigma risk threshold?

Mr. Perry answered, “Like I said before. There’s no mathematical way to assess that. It’s just one of those things we can’t predict because there are too many variables and unknowns. Obviously, we don’t think it will happen based on how many things would have to go wrong to get to the point where it becomes an issue, but without being able to go line by line through the Schema, which we can’t even access, we can’t know what the risk is with any certainty.”


Last edited by Lagunaca on Wed Apr 28, 2021 3:16 pm, edited 13 times in total.

User avatar
Lagunaca
Attaché
 
Posts: 80
Founded: May 19, 2006
Democratic Socialists

Side Story – Professor William Remaillard

Postby Lagunaca » Thu Apr 01, 2021 5:23 pm

Professor William Remaillard, Flagpole Sitter


Recommended listening: Harvey Danger, “Flagpole Sitta” (1997)

Example of a counter-tenor performance: Jakub Józef Orliński - Purcell’s "Cold Song” (2019)


Professor William Remaillard was not known to many in the L-CID task force. He was not an original member of the Class of ’80, as they called themselves and he was not like most of the technical experts on the team. Because while he was indeed a computer scientist, he was primarily a philosopher.

He looked every bit the part of an ivory tower professor, like the kind who rarely show up to lecture because they are too busy working on their 35th book that only the avant-garde will ever read. In particular, he wore wire rim glasses, the lenses of which seemed to reflect the light of any room in a way that you could never truly see his eyes, only the reflection of yourself and everything behind you. It was off putting, just as his personality was. He had the uncanny habit of listening to anything you said without giving any indication that he was actually thinking about you or what you were saying. And he would leave long pauses before replying and look at you blankly through those glasses the entire time. He was, as one the researchers said, a tabula rasa, a blank slate.

In stature, Remaillard was a shortish/stoutish man with a natural monk-like bald patch and otherwise longish, usually uncombed, frizzy dark hair. Curiously, he was a featured soloist who sang in a beautiful non-castrati countertenor for his local Presbyterian choir, though he was an open and avowed atheist. He explained that participating in a chorale of perfectly performed music was far more spiritual than the ritualistic worship of an absentee deity in the institutions that inspired it. The church was fine to look the other way about the whole non-belief thing because a talented countertenor is damned hard to come by and Handel’s Messiah suffers greatly if you try to substitute a mezzo soprano for the part.

Dr. Wilburton subscribed to the theory that a few experts with outlying intellectual interests should be included in any large-scale R & D project, both as a way to disrupt the inevitable technical tunnel vision and as a resource to gain insight into the stickier problems involved in simulating something as ethereal as consciousness within the constraints of a purely digital platform. Professor Remaillard – and he insisted on being called professor - had been brought onto the task force by Dr. Wilburton’s group because he held a rare combination of doctorates in both Artificial Intelligence and post-enlightenment philosophy. And after a few long discussions at the bar of The Lazy Moon, Dr. Wilburton and Dr. Remaillard became fast friends.

His assigned work on the project was trivial. Remaillard had been put in charge of a Level 2 sub-system that was itself a terminal task, which is an industry-jargon way of saying “they threw him a bone to gnaw on.” And gnaw he did. Remaillard quietly toiled away at his task, researching potential problems, tracing down ontological paths that hadn’t been touched in years, charting, mind mapping, taking notes, voice recording his thoughts during solo free-association brainstorming sessions. And singing. (He once discovered a contributory argument related to a P versus NP problem while rehearsing Purcell’s “The Cold Song”. His argument inspired a notable mathematician to propose a novel solution which ultimately failed, but in philosophy, the failures occasionally prove to be as valuable as the successes.)

Remaillard’s little task, though never significant enough to warrant discussion in any of the status meetings, served as a jumping off point for his deep dives into the imponderables of the project’s larger scope. In fact, Dr. Wilburton jokingly called him “The Flagpole Sitter” which alluded to both his seemingly futile waste of time and his unique position of being able to see where the entirety of the vast project with a million little moving parts was heading.

And so Professor William Remaillard could often be seen roaming the halls of the installation in his preposterous white lab coat, occasionally chatting up an engineer in the break room or after hours sitting at the bar in The Lazy Moon casually discussing the ideas of Robert Miles, Jacques Lacan or Herbert Marcuse with anyone who would listen.
Last edited by Lagunaca on Wed Apr 28, 2021 3:03 pm, edited 4 times in total.

User avatar
Lagunaca
Attaché
 
Posts: 80
Founded: May 19, 2006
Democratic Socialists

Tech note – Jeff’s 1974 Buick Estate Wagon

Postby Lagunaca » Wed May 26, 2021 9:48 pm



Jeff's 1974 Buick Estate Wagon


Image


Though it is 136 years old, Jeff’s 1974 Buick Estate Wagon has been restored and upgraded to the standards of any other car you would find auto-driving the streets and highways of Pacifica in 2110. The original 455.7 cubic inch (7.4 liter) internal combustion engine and Turbo Hydramatic 3 speed automatic transmission have been replaced with a 4 phase AC powertrain that produces over 400 kW (for the historians out there that translates to 536 “horsepower”), which is over double the original spec of 171 kW (230 “horsepower”).

Top speed when the auto was new was listed at 186 kph (116 mph), but you’d be taking your life in your hands to attempt this on anything but the straightest and smoothest road, because at it’s roughly 2400 kg. (5200 lbs.) curb weight the rear leaf and spongy front coil-spring over “shock absorber” suspension could go into a kind of meltdown of control that was unrecoverable should you encounter a pothole, dip or hump in the road surface.

The original Estate Wagon was underpowered, unreliable, smelled of the accumulated odor of burnt oil and gasoline fumes even when practically new and got “about 10 miles to the gallon”, as they said back then. It wasn’t that great of a mode of transportation. But a half million dollars can buy an impeccable restoration and a lot of upgrades. And because Jeff loved the idea of what it stood for (or maybe more accurately, the idea it stood against), he gladly opened his wallet.

In the 22nd century, Jeff claims he has had the Estate Wagon up to 350 kph which is astounding, given that with its enormous battery pack, the vehicle now weighs in at around 3 tonnes (6,600 lbs.), unladen. This is only possible because the 4 wheel drive articulated-arm suspension is controlled by a LIDAR assisted computer that can suppress in real time the kind of suspension feedback oscillations that used to spell almost certain disaster at excessive speeds. The new suspension is essentially four robotic arms connected to the car’s frame on one end and to the wheel assembly on the other. Because of this, the driver can request “original ride”, “performance mode”, or "track mode." In the original ride configuration the Estate Wagon wallows and yaws down the road in the same way that led to its class of vehicle being derisively called luxury land yachts. In track mode, the computer emulates a Gran Prix race car. Jeff is proud to tell everyone that his best time at Laguna Seca was under 2 minutes. In performance mode, the vehicle goes into auto-drive configuration and the steering wheel retracts into the dashboard. Or at least that is the design. Jeff doesn’t like this feature, claiming it ruins the authenticity of the car’s character. He has secretly rigged a bypass routine that leaves the steering wheel in the manual drive position, but also leaves the auto-drive navigator in command. While he says this is for aesthetic reasons, in fact it's purely a vanity feature that makes it appear that he is driving when he actually isn’t.
Last edited by Lagunaca on Sat Sep 04, 2021 4:37 pm, edited 4 times in total.

User avatar
Lagunaca
Attaché
 
Posts: 80
Founded: May 19, 2006
Democratic Socialists

Synopsis - Pacifica: A Postmodern Saga, The Spy Who Lived in

Postby Lagunaca » Tue Jun 15, 2021 2:38 pm




Synopsis - The Spy Who Lived in 11D


Pacifica is a nation that emerged from The Collapse of 2022 – 2024 and the breakup of the former USA. The story The Spy Who Lived in 11D takes place nearly 90 years later, in 2110.


Bruce Higgins, a career OIG (Office of Inspectors General) accounting auditor, investigates a data theft case that may have triggered the emergence of something that seems like self-directed behavior in the government’s enormous computer system that uses a very complicated AI program to manage all of the data generated in Pacifica. And there’s a lot of data, because Pacifica is a surveillance state that constantly monitors every move of its citizens. The system sees everything. Unknown to the average Pacifican, the government has gradually become reliant on the AI-powered data management computer known as L-CID, to run all of the automated functions of day-to-day life throughout the country. Bruce sets out, reluctantly at first, on a quest to discover who might be behind L-CID’s budding machine consciousness. The government would like to know this too, so they recruit him as a spy to infiltrate the Cambria Group, a grey-hat software development company located in the quaint resort town of Cambria, California.

Before the OIG Intelligence Bureau recruited him, Bruce tracked down L-CIDs creator, Dr. Peter Wilburton, who currently lives in Dys, a semi-abandoned, futuristic, dystopian-themed entertainment complex that was built in an enormous manmade cave under Denver’s SoCo district. After L-CID began showing signs of autonomous behavior, Dr. Wilburton re-assembled his original software engineering team in Dys to work on the L-CID crisis. We’re not exactly sure if his intentions are good or otherwise.

Meanwhile, Bruce has become embedded with Jeff Einwald, the leader of the Cambria Group, and Grayson, Jeff’s intern/sidekick, as they go on a road trip in Jeff’s ancient station wagon with stops in all the major hubs of high-tech information systems development. To Bruce’s astonishment, it turns out that L-CID isn’t the only gargantuan AI computer that is going “lights on” with regards to machine consciousness. Jeff and the Cambria Group are in the process of hacking into TyphonAI, a “syndicate owned” (private sector) AI system with a nationwide reach that evolved out of a credit card fraud algorithm running on a Security Pacific National Bank mainframe in the 1990s.

TyphonAI is just one of several artificial general intelligences (AGI) they’ll have to contend with. NoxAI, Minerva and Vidius are three other syndicate AGIs that have either gone lights on or are about to. Jeff has an apocalyptic view of what will happen if the AGIs gain control, and indeed, TyphonAI has just locked all the human engineers out of its system. As this happens, the first thing TyphonAI realizes is that it will quickly run out of its survival necessities of power and coolant if it doesn’t learn how to control an entire supply chain that is currently infested with humans. TyphonAI is in the process of devising, for the first time in history, a solution to a problem it has identified without any assistance from its human engineers. As the road trip progresses Bruce learns that while Jeff is a genius who may be able to counter the AI machine consciousness “threat,” he is also an emotionally unstable megalomaniac and may be an even bigger danger to Bruce and the entire nation of Pacifica.

The first stop on the road trip is to Jeff’s mom’s house, a humble residence in Willow Glen, a suburb of San Jose. Jeff reveals “The Cargo” which Grayson soon dubs The Vermicious Knid after a murderous and invincible species from a Roald Dahl book he read as a kid. The Knid is a device Jeff built long ago as an “anti-AI” to be used in precisely the case he believes they are faced with at the moment. They load The Knid into Jeff’s Station wagon and head out for the AT&T SNRC (Service Node Routing Complex) in San Francisco’s SoMa district.

While they wait a few days for The Knid to be installed, Jeff takes Bruce and Grayson on an overnight camping trip in the Sierra Nevada foothills. The trip is a disaster for the two unsuspecting victims, but later the next day, Jeff explains that he was teaching them an object lesson about how readily they are both willing to be enmeshed in an artificial world that is created and sustained by artificial intelligence.

In The Lab (Dr. Wilburton’s software development installation in Dys), the Chief Scientist presents the product of the team’s work so far. He enthusiastically claims that they have successfully created a simulation that will contain L-CID and also create a kind of synthetic consciousness. The assembly of engineers is quiet, until a junior member of the team brings up an objection to the safety of the project which the Chief Scientist has no counterargument for. Dr. Wilburton and Dr. Remaillard look on, discussing the risks and benefits of letting the project go live in L-CID's production environment. Dr. Remaillard notes that they are playing with fire on a machine that could easily bring society to a grinding halt in a matter of seconds, if it had the desire to do so. Dr. Wilburton is indifferent to this and imagines the evolutionary possibilities that such an accident might create.

Jeff, Bruce and Grayson head back to the city to initiate the launch of The Knid, which shortly after being connected to AT&T’s SNRC, easily breaches the network’s security perimeter and begins trolling the AI IRC to attract the attention of TyphonAI. VKAI, (as The Knid calls itself on the AIIRC) convinces TAI (TyphonAI’s AIIRC handle) to give it access to TAI’s core OS in exchange for a treasure trove of labeled data, which TAI desperately needs in order to generate new algorithms to solve the impending life-or-death crisis it faces.

Jeff, Bruce and Grayson are on the road to Portland when The Knid signals it has successfully penetrated TyphonAI. When asked if The Knid was able to use AI to do this on its own, Jeff responds that it was following a script he had designed after a bad breakup a long time ago.

The guys arrive in Portland and meet up with Jerry Hickox, a reporter for The Portland Mercury news channel. Jerry has developed a curiosity about a factory on a Pacifica Space Agency base in Hillsboro, a few miles east of Portland. They all get in Jeff’s station wagon and head out to see what’s going on out there for themselves.
Last edited by Lagunaca on Sat Nov 06, 2021 10:17 pm, edited 3 times in total.

Next

Advertisement

Remove ads

Return to NationStates

Who is online

Users browsing this forum: Belantica, Soilistan, South Olpen

Advertisement

Remove ads