> The results you see in OpenAI searches are not paid placements, but organic results.
Aaaaaaand it’s gone. Please step aside, this chat is for people who will actually buy the sponsored products. Next please!
2,382 karma · joined March 18, 2024
Aaaaaaand it’s gone. Please step aside, this chat is for people who will actually buy the sponsored products. Next please!
Unfortunately, the self-worth, poor compensation and no protections part is key in how a lot of people make it into games, and get their first games shipped. Beyond that, it’s easier.
I’m not saying this is a good or bad way to enter the industry, by the way. I have a strong opinion on it, but it’s outside the scope of what I wanted to share.
I found that AAA studios will pick someone with game shipping experience and some good code examples on GitHub over someone with a university degree 9 times out of 10.
The degree does matter a bit more in AAA, especially R&D roles. These studios actually have some R&D to do. But for other roles and other company sizes… does it truly matter in Sweden? I mean, wouldn’t a decent portfolio (including just a list of titles they worked on) jump anyone to the front of the shortlist, ahead of degrees but weaker experience?
That’s been my experience.
Much has been said and written about the cruelty of immortality, since the myth of Sisyphus and probably before.
“The Mortal Immortal” is a story about prolonging lifespan but not healthspan, and also not emotional fulfilment in life — https://en.wikipedia.org/wiki/The_Mortal_Immortal. There are also TV shows (“Altered Carbon”, “Black Mirror” USS Callister and other episodes, “Twilight Zone”) and video games (“Nobody Wants to Die”) about it.
What your parent comment speaks about (the vanity of endless life) is particularly explored in Bernard Williams’s “The Makropulos Affair”. Endless life could mean no regard for quality of life and endless trivial pursuits.
While fear of death is widely considered universal, many people (for various reasons) fear death significantly less than others[1]. Fear of death levels appear to generally stabilize to low past 60 years of age. Many philosophies reject the fear of death, including some of our very popular religions. Only about half of terminally ill patients feel clinically significant death anxiety[2]. It is normal to not fear death without being depressed, enlightened, or stupid.
I think it’s not so black and white as evolutionary psychology might suggest.
[0] https://peterattiamd.com/outlive/
And I’d say most people probably don’t fear death or non-existence itself, but rather the process of dying (suffering, stress, pain, shame, loss of agency, the grief inflicted on others, etc). In palliative care settings, where the process of dying is well-managed (physically, emotionally, spiritually), people don’t seem to be that afraid. Many make peace. Or at least that’s the image painted by popular science articles on the matter. And speaking for myself, N=1 and all, I really do fear the process only, and not the conclusion.
That’s why I think dignified death laws for these settings are important. And why I say I fear what society will put me through when it’s my time, if such laws aren’t passed.
I am concerned in general about me living past my health-span. It’s a new concern for many, as medicine traditionally focused on prolonging lifespan. But now people are talking about how full of suffering life is when one doesn’t have whatever minimum of health they deem required to live, but are kept alive anyways. Sometimes agains their wishes. It’s just a macabre prolongation of the process of dying — something that really is scary.
Or that’s what I’m afraid of, anyways. But I understand it’s a difficult and somewhat taboo topic in society, so apols if it offends.
It’s good we are starting to develop dignified death laws. With the world population as it is, more people will die in the next decade than any other in history (even the plagues). Just looking at population graphs, 1B might die between 2050 and 2060. Much suffering can be avoided.
Hah, I also noticed that. For some reason, ex-army programmers really excel in their junior years (or sometimes even just months), too. Do you think it's the attitude towards problem-solving?
And, of course, most tech companies will say that tech should be a meritocracy. Then they'll offer bonuses and incentives for anyone who can get a friend to fill an open role. Consider that.
I entered the gaming industry a long time ago as a software engineer without a computer science degree. Later, I thought I'd get one online, just as you are considering. It was reasonably cheap compared to my income, and I thought I had the extra hours for an evening online course. But I found it to be very demoralizing. I'd be micro-optimizing containers in AAA game engines, shaving microseconds off operations by reducing CPU cache misses on target consoles in the morning, and earning a 60% grade on my JavaScript game in the afternoon, because I didn't follow the method shown in class. I also did poorly in CS math, especially with formal proofs, which were a lot of fun for math enthusiasts but grating for me, being an expert in game math and graphics math at the time. When my fourth semester rolled around, I found only one or two modules to be somewhat useful, while others, such as HTML web design, had no bearing on my profession. I dropped out about halfway through my degree, leaving me with a lot of money spent but nothing to show for it. I could have forced myself through it, but it would have been on pure discipline (which is easier for some than it is for me!)
My advice is: don't get a CS degree because it's doable. Get it if you need it for something, like if you wish to enter a specific industry that requires a meaningful academic experience, such as research. Your degree will then serve a purpose. Otherwise, you might feel like you're not getting much out of it, and it won't be easy to justify carrying through. Especially if you're already performing at a level significantly higher than what a degree puts you on. Or maybe the purpose a degree could serve for you is fun - perhaps the academics would be rewarding for you in a way that doesn't have to flow into your work, or you get networking opportunities that a good uni/college affords. But my view is: it's a significant commitment, so there has to be a meaningful, clear reason to do it. Otherwise, it will be all cost and no return.
I think Sturgeons law applies to co-founders, and if you haven't spoken to at least 20, you are likely talking to the crap ones. Not all crap ones demand excessive compensation, there are other types of crap. But there is a huge difference between the ones that will be good for business and the others.
I don’t, I just naturally hang out with the people who are most reciprocating and being the most good vibes.
Ultimately, it’s about me first. I function better and am happier when I have recently socialized. I’m not doing this for others, so it’s all good and I don’t feel like anyone particularly pushes me in a way where I’d need to maintain boundaries for myself in this particular context.
Regarding your three points, I hear them. But you probably have enough free time to text a friend, and sometimes physically meet up with them. I’d say, if you really don’t have time for that (and I was there myself), then probably you’re neglecting your social needs (basic socialising, safety net) for something else. Consider if that something else is worth it. Regarding point 3, I used to expect something reciprocal from friendships, but now I just expect to spend time with/among people. Yeah, most won’t help you in a time of need, and some will speak ill of you behind your back. But you’ll also meet many great people that legitimately will be great friends. So I’d say don’t worry about it — “trust the process”.
Overall, I hear some social anxiety in the 3rd point. Social interactions aren’t always ideal but don’t catastrophize them. Just do what’s good for you yourself first. Make sure your social needs are met. Let all other things and friendships develop or not as they would.
I think many people don’t make the effort anymore, but that doesn’t mean there’s anything wrong with you if you do — nothing to be anxious about, I’d say. Besides, leading and organizing things has its advantages.
I started texting my acquaintances and old friends at random just to ask what’s up every once a week or two. I also started calling some people instead. Almost everyone responded really well to that. When I go out to eat, I often check with a few people if they want to go if I swung by and picked them up.
This is how things used to be for me before the pandemic. But it was difficult to get back to it. What I found though is that most people are lonely but they don’t want to put much effort into building friendships. And that’s ok, I can be the one who initiates the outings and chats.
It’s okay to be the one who initiates. It seems like not everyone can, somehow the level of social anxiety has gone up in the world. In the end, I get my socialization full and so do they. So I’d recommend to whoever feels a bit lonely — reach out to your past friends and current acquaintances.
The only issue with always being the initiator is that no reciprocation is a bit of an… issue to our social brains. Validation, trust, confidence, and friendship itself forms better when there is reciprocation. It’s best not to overthink it, the world is different and what it means to have friends has changed. This is the new normal. It’s better to be the one who leads all the friend groups and activities all the time than to be lonely.
I am so tired of it, too. Toying with the legal boundary of lying in communication is pathological, maybe even sociopathic.
Everyone knows when someone is doing it, too. We just don’t have the means to punish it, even in the courts.
The whole “I won’t get punished so I’m doing all the immoral things” habit is foul to begin with. I don’t know how, but I hope our society can get over it. As things stand, there is no way to outlaw being an asshole.
Just like Facebook, the first-mover advantage has favored many now-established studios and franchises. They exploded game-development costs because they could, and funneled these costs into marketing and moat features indie developers could not build (such as huge open worlds, amazing sweaty character face wrinkle rendering tech, and SOTA systems). But many of these companies did not respect the player's wishes for well play-tested games with interesting stories and mechanics. Still, they captured the top 20-30 franchise part of the annual net bookings, and strongly compete in the top-50 game part. Some even built some black hole games (GTA Online, Rainbow Six: Siege, Fortnite). For a long time, they avoided much of the pressures felt strongly by smaller companies. They were "above" the 99% of games that have to compete for close to 1% of the revenues. Their marketing was so strong (plus, they strengthened it with access journalism) and features so moated, they could do no wrong.
However, over the last 5 years, things have changed. Many AAA industry legends have left their jobs at major studios to start small studios and create games as a form of interactive art, rather than to make publishers rich. Ultimately, in their view, the greed and blind following of what players would consume (trends) in large numbers led to a sterile industry that could no longer create art.
The growth engines got exhausted because players did not actually demand what they were offering, such as season passes, eSports corporate shooters, microtransactions, padded playtimes, user-generated content, and the other things. The new growth engines (AI, targeting kids, etc) are also what the players don't want very much. The industry understands it, and investors are starting to catch on after facing a decade of poor returns, too. The crucial point I am trying to make is that the industry spent a lot of money on these growth engines that the players didn't truly want, led by market metrics that genuinely showed they were consuming it. But now the gig is up, the writing is on the wall, and everyone inside and out of the industry sees it.
As a contrast, many Eastern companies (Nintendo is an especially prominent example) stuck to classic pricing models, did not inflate the cost of their games with their money for moat (most indie developers can make games to compete with Nintendo outside of the IP), and never used the growth engines used in the West. These companies, along with many people in them whom I know personally, are largely unaffected by the industry crisis. They were always making games their users wanted.
Finally, I have to say, the industry is split in two. 8/10 AAA companies are struggling because they cling to the growth engines (old and new) that the players don't want. About 2/10 game developers and publishers genuinely build games that people want, even in the West. And now that the pressure is up, some AAA executives from the 8/10ths are becoming acutely aware of this. Emphasis on "some". So, yes, the industry in some part was, is, and will continue to make games that players want. But the more interesting part for our discussion is the large part of it that wasn't, isn't, and perhaps won't be.
Of course, there's some probability I'm reading this wrong. I'm making my business bets in the industry based on it, but that doesn't mean it's necessarily right.
And thanks for reading the report before engaging in the discussion. That is appreciated.
Calm it with the ad hominem attacks. It's not the place for it.
I didn't mean it's on a continuum, as you assumed. Apologies for phrasing it unclearly. I meant that the weights are public. They are open; there is no debate to be had about it. Generally and broadly, that is already considered open-source.
And we all understand what "open-source" means in the context of Llama - it doesn't mean one of the idealized notions of open source, it means open weights.
It turns out that demand matters when you sell a product or a service. And it is elastic in ways other than price (such as convenience, value, appeal), but not infinitely so. In plain English, you can force anti-social media onto the market by making it appealing/hooking/addictive/convenient/supposedly valuable for a while, but not indefinitely. People do demand proper socializing, especially recently. Many are realizing they've been sold a total bag of goods just because they consumed it, and it's not good enough to displace real human connection.
To me, they sound a bit like “no true Scotsman”. Llama is open source, compared to commercial models with closed weights. Even if it could be more open source.
That’s why I looked at it in a broader sense — what could happen in an open-source world to improve or replace Llama. Much could happen, thanks to Llama’s open nature, actually.
The most capable parts are for-profit organizations that release open-source software for their business imperative, public benefit companies that write open-source software for ideological reasons but still operate as businesses, and a tiny number of public benefit organizations with unstable cash flow. Most other efforts are unorganized and plagued by bickering.
Llama itself is challenging to take over. The weights are public, but the training data and process is not. It could be evolved, but not fully iterated by anyone else. For a full iteration, the training process and inputs would need to be replicated, with improvements there.
But could another open-source model, as capable as Llama, be produced? Yes. Just like Meta, other companies, such as Google and Microsoft, have the incentive to create a moat around their AI business by offering a free model to the public, one that's just barely under their commercial model's capabilities. That way, no competitor can organically emerge. After all, who would pay for their product if it's inferior to the open-source one? It's a classic barrier to entry in the market - a thing highly sought after by monopolistic companies.
Public benefit companies leading in privacy could develop a model to run offline for privacy purposes, to avoid mass consumer data harvesting. A new open-source ideological project without a stable business could also, in theory, pop up in the same pattern as the Linux project. But these are like unicorns - "one in a million years (maybe)."
So, to answer your question, yes, Llama weights could be evolved; no, an entirely new version cannot be made outside of Meta. Yes, someone else could create such a wholly new open-source model from scratch, and different open-source groups have different incentives. The most likely incentive is monopolistic, to my mind.
DeepVariant, Enformer, ParticleNet, DeepTau, etc. are some well-known individual models that are advanced branches of science. And there are the very famous ones, like AlphaFold (Nobel in Chemistry 2024).
We need to think of AI not as a product (chats, agents, etc.), but as neural nets (AlexNet). Unfortunately, large companies are "chat-washing" these tremendously useful technologies.