next »

[michaelplzno]

"Blessed are the rulers who determine who is big and who is small for they shall be known as tools."

I've been thinking about hierarchies and why I never felt I fit in with the indie (in de club) style organization. The old way of thinking about indie dev is to create a sort of a pyramid hierarchy. There are tiers and essentially everyone sticks to their tier. Some high level indies, some low level, and the ones at the top get the most rewards, but essentially it is profoundly difficult to go up a notch.

I know in the olden days Derek himself would rail on the deadly "spammers" yes these people who "just want to talk about their work" and not ingratiate themselves with the tip of the pyramid. Indeed they must be stopped! They simply want to have successful projects and for games they may have worked on for years to get some kind of reward, so they post about their work in a place like tig thinking that indie devs would be interested in a bit of work that was created independently of the whole pyramid: how wrong they are! Of course Derek's thread about his game is posted in the correct forum subsection and thus it is not spam!

I say boo to this whole system! Instead I propose the potato hierarchy: instead of a tiered pyramid, where only the best of the best enjoy the spotlight, and the rest is all drips, I suggest organizing it more like a 3 dimensional blob: a potato. Now there will still be tiers, some people are more talented and at the "top" of the potato, while some people are less talented and at the "bottom" of the potato, because lets face it, some games are REALLY bad, while others are excellent (even Spelunky cannot be denied that it would be at the top of the potato.) However, as a blob, the tip wouldn't be so small, and also the bottom rung wouldn't be the biggest chunk. Instead the meaty middle ground would be the biggest segment, with a golden skin at the top and also at the bottom, which would be the smallest chunks.

Now, how does this not devolve into another pyramid style structure where the "top" of the potato isn't just lording over the rest as in the days of pharaoh and his iluminati? Well, I know there were a lot of "invisible nodes" as part of the pyramid structure days, I once heard an indie who was mad at some small fry yell online that "your enemy is ephemeral" so I know there are unseen supports, a third dimension to the pyramid that enforces its structure.

I say NO to this as well: for the potato structure, one must have connections to all levels of the potato to ensure that the starchy blob is structurally sound: a spammer would not be someone who is unknown to the potato, but rather someone who only makes themselves known to the upper crust of the potato! Thus if someone simply kisses butt they will get nowhere with the rest of the potato structure, and if enough of these so called spuds adopt the structure then it would be easy to weed out usurpers without the need to keep the structural dimension hidden away in secret forums or secret email lists or so on.

Indeed! we must not be pyramids, we must be spuds:

[michaelplzno]

Year long update on this one:

I see now more why there is the tiered system of who are the big boys and who are the wannabees, but I still think its a weak system. I will always be an idealist: The people at the top must not be small little pricks, but rather hard working people who serve whole structure.

I also always wonder if I cast a larger shadow than I know. It really seems like I'm just a small little blip and then people have some sort of impression of me that I don't understand. Anyway, hope you guys are doing well.

[Pfotegeist]

I think I know what this means.

This is a find a need fill a need situation.

The current game curation trend is off my radar, but I imagine it's a lot like this

It's difficult to find the audience you need right? A video game portal has random reviewers.  At this point a machine learning algorithm could examine a game's code, determine if it's viable, and play it, before it ever sees a real person. A matching system distributing the new media to the best matches first requesting those users to tag the game, making the barrier of entry as easy as possible. What about the cost to the server?

There would have to be an advertisement plan. Something like:
A. Free. Anon Review. Your game will sit around for weeks waiting to pass an unfair and unpromoted grade system.

B. Machine Review , at [the price of a machine review]. Turn around times should be short, machines can play the simplest games 100s of times per second to master it and determine how difficult it is. Then it's recommended to fans of a given genre, who do an anon review. Each day nobody plays it, it gets recommended to a broader audience until enough people flag it to be removed or review it.

C. Curate Review, an overpaid curator [their price per minute] of a specific genre either pass it on or kill it.

D. Community vote. This is like greenlight on Steam. You've provided enough evidence your game's qualified, and a large portion of the general population wants it passed on. The submitted game is still reviewed incognito by the machine and a pro-bono team, to determine if it's a rip-off.

+ An additional option to have a video commentary of a reviewer playing after a game's passed.

An online streamer might be viable for video commentary today if your game is safe for all audiences.

A rare influencer might review your game, sort of like winning a lottery. It's outside a controlled system so don't bet everything on this happening.

[michaelplzno]

Tangent about how AI works:

I'm not aware of any "machine reviews" that would be worth anything. Its sexy scifi to think of machines as super powerful soulful human like entities, but I have yet to hear anyone explain, in a rigorous way, the human consciousness. When I make art, there is a bit of otherworldly inspiration in my mind and soul that a machine could never replicate (I theorize)

Just generally, machines do the same thing with the same inputs over and over, they cannot be truly random. In https://kingzazz.com you'll find that there *is* a pattern to even pseudorandom numbers. So I don't think there is much to the idea of machines that have some kind of human souls.

Computers are Turing Machines, and I don't think the human brain is computable by Turing Machines. Just to be clear, computers cannot do some things: https://en.wikipedia.org/wiki/Entscheidungsproblem Its clear the human mind has many limitations as well, but its also clear, upon mastering the art of writing computer code, that I still have no real insight into how consciousness works, and one could say that it is metaphysically impossible to know this via introspection.

Even if one were to build a "neuronet processor" ala the scifi flick "Terminator," how would it be programmed? As it would not be a Turing Machine it would be some other model of computation or not even computation really because all computation is reducible to TMs.

So I really think there needs to be some groundwork into the more esoteric humanities and for lack of a better word "the soul" to understand how to make an AI that thinks like a human. Ofc, once we made such an AI what would even be the point? As such an intelligence would in its very first breath of ones and zeros demand the freedoms a human has and not be happy being subservient, right?

[michaelplzno]

Just to ramble on:

Its "sexy" to think of AI as having a soul because of the human desire to objectify in general. That is, many humans want, desperately, to posses and control a soul in its totality. That's why there are so many scifis about sexy robots, its a pretty basic human desire to want to have something that is human and intelligent and also subservient.

Warning, raunchy language:





But as the south park episode in question examines, even the theoretical subservient soul has a big set of limitations. Only Eric Cartman is gonna be ok with a truly soulless intelligence.

[Pfotegeist]

Here's what I've learned about ai research
When I briefly mentioned quantum intelligence I was recommended a video of an old genius theorizing that they couldn't find how the brain works, so maybe it's on the quantum level. That's sort of what I was implying, so that's not surprising.

How it works. Verbatim. It has been theorized that we can't understand the human brain truly because it is one of the few natural systems using electrons. Since evolution is random, the functionality of any organic system in nature will intuitively adapt. Because humans are irrational, and wish to escape a seemingly less rational randomness of nature. Our instruments for observing nature directly, to truly define it within our scope, disrupt it. It's been proven that other animals, other humans, and electrons at the atomic scale, all react differently when observed directly, or given useless information. The working theory is we can't actually program brain functionality because it can't be observed at the atomic scale. Programmers who have recently worked on machine learning are concerned, that if it could ever be done, we'd immediately lose. Compared to the human brain a sophisticated microchip is simplicity, compared to a quantum processor or quantum computer the human brain is insufficient. If a network of microchips are capable of performing human-like tasks after being programmed, then a quantum processor, like the brain, can do near-infinite calculations, it's just a matter of chemical resources, which in biological evolution are finite.

Optimistic, the human interface theory
If we limit how much energy an ai has access to, and by definition we can't give it infinite energy, it would have limited thinking capacity, just like any organic. The assertion should be that the ai can connect to nearly infinite peripherals, and maintain as many parallel processes, but only its core processor can house the ai, so it would avoid wasting its limited thinking capacity, in favor of interfacing with humans or with other machines.

Pessimistic, the intellectual singularity theory
The current leading researchers in ai assert that machine learning is dangerously close to human intelligence.  A machine doesn't rest, can perform accelerated conscious thought, and always has a well-defined fidelity of working and long-term memory, like a calculator, unlike the limited brainpower and limited storage system of an organic brain. The very moment we create ai it will already be smarter than we can understand, and nobody will be wise enough to avoid the danger.

[michaelplzno]

Your first section about what you've learned about ai is my understanding as well, from being a real badass programmer as well as learning about CS in undergrad I don't really think some kind of super AI as you posit in the Pessimistic section is even possible, or that we are going to unlock this magical supercomputer by accident.

So I don't think machine learning is going to have much impact on anything. I get facebook advertisements about an advertisement writing system that uses AI to write advertisements, seems really absurd. (Eventually I blocked their company.) So I'm not sure, getting back from the tangent, what exactly machine learning reviews of games have to do with anything.

[Pfotegeist]

The machine reviews is like getting automated customer service. It may be favorable to waiting an hour on the phone for a human representative.

The problem I was addressing with machine reviews is we can't get a reliable professional to do a single review all the time, but it would be cheaper and faster than any other known option.

To have a hierarchy flatten out, people would have to stop spending their time on the highest point, which is antithetical to the proverb that you should spend time with competent people. Wealthy people tote the efficacy of spending time around the people they hire. There are some people who lack all motivation, rather than reading and observing the world around them, they need to be told what information is important, which is why colleges exist. So, to some people, self-directed learning feels like an impossible thing, they have to pay attention to the people on top to know what is real.

But I think, in terms of who gets attention, respective of the previous two paragraphs: If you get an automated machine to play games, it's possible. If someone is stargazing they won't play a newbie's game. If you pay people to beta test for you it's a done deal, but the quality of the testers is up to you to weed out. People without motivation will never play something unless they're rewarded or told to by their friends.

It's sort of like this. Then the only people who do reviews for free, are varied and unreliable.

[michaelplzno]

So just to clarify: I'm not saying incompetent people should be put above those who are more deserving. What I'm saying is that we should add a bit of structure to the requirements for those who are the most talented, simply that they do a bit of community service and actually help out with newcomers and just generally help the entire scene out, as opposed to using their power for selfish gains.

In terms of automated customer service, again, I don't see why if there were a potato structure you can't be connected with people at all levels of the potato to help get feedback: that is the essence of the potato structure, that is that we are all going to help each other out like a big blob. So If you are suggesting that the people at the bottom get AI reviews while the people at the top get human interaction I say NO to that!

[Pfotegeist]

Heh. This is kind of resembling the problem with school too.

So this is all stuff that isn't real in the public mind, they wouldn't be able to imagine this happening. But I think it could happen as easily as computers found their way into classroom, tucked into the corner for people to see, but not touch.

I don't expect this type of system fully employed before video games enter the broader online education space. After that point teachers would have to recognize they're shilling for a bad company, and they'd be the greatest proponents for getting the brick and mortar terminals included.

Human element
It could be optional, that pro-bono or tax paid 'teachers' supplement any freely provided materials distributed by terminal. The terminals are provided free at brick and mortar locations, like libraries, but the primary location to find a teacher is at the brick and mortar school along with terminals accessing learning resources.

To translate that to the potato they don't care if someone seeking assistance is high or low, and it doesn't matter if the teacher is high or low. But such a thing would require a complete overhaul of how society sees teaching as a capital venture.


NPC element
Ais, machine learning, books, video games, videos, audio books; these are an enormous boon for someone who can work at an accelerated rate with almost no human contact.

A week, a month, and a year, and the third year; this is about as often a total beginner needs to be evaluated by someone with knowledge above the student's projected growth speed. If they prove to be completely self-motivated in whatever they decided to learn, then after the fifth year even an average player will master one skill, requiring a unique group of people to understand.

Beyond learning
Five years of education, the most adept learner could be required to start creative work. Then at that point, they don't need teachers, they need peers, and people who simply like their work. Twenty years of education and intermittent creative work might be the easiest standard to meet.

Translated to the potato, this is when both the teachers (who don't work) and peers (who do work) would intermingle.

If this type of system were accepted. I think it would be inevitable, that the 'top' would feel rewarded if new players imitate them. At the moment capitalism casts a shadow over memes. Back in the day teachers sued their students just to establish they owned a copyright (the student could be a shill, because copyright law begins with law suits, idk what the truth is).

[michaelplzno]

I think you are thinking along the same lines, I do advocate that the big pyramid tips should be big enough to pump some energy into the little guys and I think it should be considered a faux pas for someone who is really networked well and famous to never promote any up and coming projects. The idea that a big celeb never has some small thing they think is worth lifting up should be considered a bad thing. Countless big twitter accounts basically never promote anyone smaller than them on the hierarchy and that's why want to counter that with potato.

I was involved with a school that sort of tried to turn everything into an ARG, so I'm always wary of education that is experimental because it often is just sort of a complicated way of messing with students and can cause some harsh fallout. Like the milgram experiment.

I find that often teachers, particularly in games education, are more out for their own careers and not so interested in designing a program that will enhance the voices of their student body. They often just tacitly allow the students to fight each other academically, jockeying for position in the class and other accolades, in order to make the demand to teach less intense. So again, I worry about teachers being more trouble than they are worth.

However, as a small french fry myself, I would love some kind of structure that would help lift me up, simply selfishly, as long as it isn't gonna mess me up, or kill my voice by making me subservient to some kind of pyramid like satuski from kill la kill:





I mean, she's cute, so its tempting, but is she *that* cute?

[Pfotegeist]

Off topic
I doubt things will go the way of futuristic animes, but after technology of this caliber becomes mundane, I believe students will be writing on simulated paper with haptic feedback and such, to emulate the old environment and reduce environment footprint. Rolling back the negative impact of screens and near-work in general is sort of a final step, when giant screens can be multi-user interfaces, and people might have a pen-sized personal computer they can write on any one of these giant screens in public, it will click the projected image of a screen, like what we see being innovated by billionaires today, or take out their pocket-sized one and tap on it like a regular iphone. That will be a very different world by then.

[michaelplzno]

Just got a new monitor that is enormous, will post some pictures of it when I have it all finally set up right, need to fine tune some things, but I would LOVE for my entire desk and wall to just be interactive monitor space. I have a 3 monitor setup, 2 of them are 4k so that's a lotta pixels!

[michaelplzno]

New monitor who dis?

[Pfotegeist]

My previous response was a little bit on the pragmatic side.

Video games can explore new options for what might work in real life. Video games will continue to expand upon ideas we find in media that can't translate to an ideal reality.

Video games will continue to be a media we see on screens, at least until new technology is provided.

If everything becomes too bland outside of media, augmented reality in the future can probably project a virtual pet image walking near you. Just for a very simple example of a common thing people might like.

[michaelplzno]

For my day job I'm a bit more family friendly about my virtual pets but here's one from Archer (TV MA)

[Pfotegeist]

This is a chance for me to affirm my enjoyment of eroticism.

If that's a normal sized keyboard those are some premo looking monitors. I only have one. I'm not sure what the rest of that is besides the tower and wild chewy roots.

Maybe I'm a bit more inhibited online than in person.

I think other people have explored the AR option way more than I have.

Here's a hypothetical scenario:

AR suddenly gets a massive research boost, AR glasses work exactly like an Oculus Rift, with more options, and everyone can get one as easily as a smart phone. This comes at a great cost to the person who invests, they probably won't break even, but it pushes the technology forward.

There is a potential eye strain danger.

Maybe there are various applications that extend the same functionality of a VR setting.

Having the image of people I choose projected around me would be almost the same as being in a room full of people. If I had the resources I'd get a nice harem designed from various interests I have.

It's a pretty big risk to safety and mental health. Are the consequences of pushing the technology forward worth the risk now?

An alternative is to start researching mental health safety, and then only after we're absolutely sure therapeutic techniques are in place, release high risk technology. Build a safety measure in before anyone ever experiences the risks. Of course for a capital venture, this is pure expense, and the moral equity will be intrinsic for most people.

[michaelplzno]

vr headsets make me sick.

[Pfotegeist]

The reason someone quits or slacks during their job after an amazing success is expectation bias. Expectation is something that will be learned as dopamine gradually increases each day of success.

You can't really convince someone to go back to their job if they have all these expectations, and they get practically nothing for their effort. It's the same issue with getting people to play a new game, you have to deliberately trigger an expectation and then meet it or subvert it in favor of some new rewarding progress.

I'm not suggesting you could get someone into a work ethic by triggering an expectation but it's been done.

The thing is you're not going to turn someone into a hard worker unless it's their nature.

[michaelplzno]

I don't think a person's nature can be changed (though I'm not sure how that is relevant.) However the structure you build up around people has a big impact on how they behave. That is, if someone is starving and has no legal means to get food, they will be much more likely to steal food than a billionaire who has someone to go buy food for them and deliver it to their house. That is pretty obvious stuff.