When you open devRant and the first thing you see is "Fuck you, Windows." and you can't help but giggle 😂

I am working in a speciffic engineering team. We are using tools the company has bought and has separate teams administrating them.

Tool X is malfunctioning, throwing server-side errors (some .dlls are mentioned in the err msg)

Me: XAdmin team, there are some suspicious errors and I cannot achieve desired results using tool X

XAdmin: Let me see
XAdmin: I have checked a few forums and could not find a solution. Please log a vendor case

Me: *wat........*
Me: Vendor will most likely require some techical info, some licencing info. How do I go about that?

XAdmin: reach out to the vendor, they will schedule a call. Forward that call to me

Me: *wat............*
Me: *for shits and giggles, register a bogus account at vendor site, try to log the SR*
Me: XAdmin, while logging a SR I am asked for licencing info. What is the aaa, bbb, ccc info of your licence?

XAdmin: *crickets mating*

wtf buddy... How can you call yourself Admin of tool X and ask your customers to log vendor cases for you.....? WTF are YOU there for then??

I'm still WTFed. Like wtf....

EDIT: the guy I was talking to is XAdmins' team lead

This is gonna be a long post, and inevitably DR will mutilate my line breaks, so bear with me.
Also I cut out a bunch because the length was overlimit, so I'll post the second half later.

I'm annoyed because it appears the current stablediffusion trend has thrown the baby out with the bath water. I'll explain that in a moment.

As you all know I like to make extraordinary claims with little proof, sometimes
for shits and giggles, and sometimes because I'm just delusional apparently.

One of my legit 'claims to fame' is, on the theoretical level, I predicted
most of the developments in AI over the last 10+ years, down to key insights.
I've never had the math background for it, but I understood the ideas I
was working with at a conceptual level. Part of this flowed from powering
through literal (god I hate that word) hundreds of research papers a year, because I'm an obsessive like that. And I had to power through them, because
a lot of the technical low-level details were beyond my reach, but architecturally
I started to see a lot of patterns, and begin to grasp the general thrust
of where research and development *needed* to go.

In any case, I'm looking at stablediffusion and what occurs to me is that we've almost entirely thrown out GANs. As some or most of you may know, a GAN is
where networks compete, one to generate outputs that look real, another
to discern which is real, and by the process of competition, improve the ability
to generate a convincing fake, and to discern one. Imagine a self-sharpening knife and you get the idea.

Well, when we went to the diffusion method, upscaling noise (essentially a form of controlled pareidolia using autoencoders over seq2seq models) we threw out
GANs.

We also threw out online learning. The models only grow on the backend.
This doesn't help anyone but those corporations that have massive funding
to create and train models. They get to decide how the models 'think', what their
biases are, and what topics or subjects they cover. This is no good long run,
but thats more of an ideological argument. Thats not the real problem.
The problem is they've once again gimped the research, chosen a suboptimal
trap for the direction of development.

What interested me early on in the lottery ticket theory was the implications.
The lottery ticket theory says that, part of the reason *some* RANDOM initializations of a network train/predict better than others, is essentially
down to a small pool of subgraphs that happened, by pure luck, to chance on
initialization that just so happened to be the right 'lottery numbers' as it were, for training quickly.

The first implication of this, is that the bigger a network therefore, the greater the chance of these lucky subgraphs occurring. Whether the density grows
faster than the density of the 'unlucky' or average subgraphs, is another matter.
From this though, they realized what they could do was search out these subgraphs, and prune many of the worst or average performing neighbor graphs, without meaningful loss in model performance. Essentially they could *shrink down* things like chatGPT and BERT.

The second implication was more sublte and overlooked, and still is.
The existence of lucky subnetworks might suggest nothing additional--In which case the implication is that *any* subnet could *technically*, by transfer learning, be 'lucky' and train fast or be particularly good for some unknown task.

INSTEAD however, what has happened is we haven't really seen that. What this means is actually pretty startling. It has two possible implications, either of which will have significant outcomes on the research sooner or later:
1. there is an 'island' of network size, beyond what we've currently achieved,
where networks that are currently state of the3 art at some things, rapidly converge to state-of-the-art *generalists* in nearly *all* task, regardless of input. What this would look like at first, is a gradual drop off in gains of the current approach, characterized as a potential new "ai winter", or a "limit to the current approach", which wouldn't actually be the limit, but a saddle point in its utility across domains and its intelligence (for some measure and definition of 'intelligence').

who-the-hell-thought-of-calling-this-kebab-case !! 🍖 😂

This *is* a question you silly wrong tagging mother fucker, how dare you doubt me?

Alright, no more disclaimer: I like dungeons and dragons, but it's too fucking much in terms of rules and systems and shit, as in just *making* a character can take a long ass while.

And if that's the highest level of all your ANAL preferences then OK, but I'm not you and things only come OUT of my ass, not inwards, I swear.

Anyhoo, I got fed up with it and wrote my own ruleset and setting as a last fuck you to everyone. It's very simple: if you want to be some kinky magical alien hermaphrodite royal prostitute half sewer dragon princess and three quarters bearded female incest child of demons and fairies then FINE, but you get no bonuses for that shit.

Get it? No complex racial level scaling bullshit, FUCK YOU, race and background is just for vibes, end of story.

You get no attribute or skills or shit to distribute on level one. All you get is a prompt: pick three actions, that's it. You wanna be sexy? Pick "seduce". You wanna set turds on fire? Pick "ignite". Are you an edge lord? Pick "summon". Would you be my wife? Pick "heal", "buff" and "smite".

The game is turn based, and each action you can take is effectively a spell. Everyone can cast a basic spell like walk, attack, talk, crouch, etcetera -- that costs no mana. Special crap like flying and firing fucking electricity costs mana, and you can only do those if you either picked the spell on level one or learnt it later from a book/tutor/demonic bargain/whatever.

Which spells are valid for taking at level one is up to the game master; I just tell people to pick three verbs or short sentences, and if they choose something that's too broken like "split the Red Sea" I'm like nah you're not Moses, try again.

Still with me? Good. You get eight points of health, four points of mana, and one point of stamina. They're all energy, and you can use it to power your magery, but spending all your health means you fucking die.

Stamina recharges fully every turn, and is used for the aforementioned basic actions. All of these cost one point of stamina each. If you run out of stamina, you can use mana. Or your BLOOD.

Level one spells cost one mana, level two cost two and so on. You get back one point of mana each turn, and you can fire all the spells you want during it, long as you have mana. Or BLOOD.

That's good and all, but if you spend anywhere over eleven combined points of energy in one go, you spontaneously combust and die, erasing all signs of life in a twenty-meter radius. This is called incineration, and it *will* leave behind a blackened crater from which the dark servants of the Horror Immemorial may or may not crawl out of.

In case you didn't guess by now, your blood doesn't fucking come back unless you eat, sleep or see a healer.

But anyway, the more points you spend into casting a spell -- and remember, basic attack counts as a spell -- the more powerful it is, so the bigger your diceroll can get. My rule is I add one dice for every fourth point of energy spent, so (1d4), (1d4 + 1d6), (1d4 + 1d6 + 1d8), incineration.

Additionally, for every three points of energy spent, your spell can hit one more target. That's right, you like AoE? Then spend more mana, bitch. Oh, and if you're using shit like poison it lasts one more turn for every two points of energy spent.

How do we calculate damage? Diceroll over two and fuck your mother. Armor class? Resistances? Out of my face with that shit. Damage reduction is called "tyranny" and is for dungeon bosses only.

If you live long enough to get to level two, you *do* get attributes. Pick:
- Grit: +2 health, +1 to fighter shit type rolls.
- Cunning: +2 mana, +1 to rogue shit type rolls.
- Allure: +1 stamina, +2 to wizard shit type rolls.
- Spirit: +1 to elemental shit type spells.
- Faith: +1 to benefactor paragon asshole shit type spells.
- Hatred: +1 to demonic murder hobo destructive shit type spells.

On second level, you can pick one of the spells you know to get +1 to it, specifically. Eh, "+1" just means you get a bonus to some diceroll, no time to explain I'm running out of characters what the fuck.

On level three, the cycle repeats. Pick attr, pick spell. DONE.

Oh right, and weapons. Mostly just vibes, pick your fancy and fuck off. Normally, you can hit things one tile away; if you have a BIG melee weapon you can hit from *two* tiles away, and if you have a ranged weapon you can shoot anyone in sight, but you need to spend one point of energy to reload.

And there, all bases covered in less that 5000 characters with some flair to spare, now suck my fucking cock Hasbro.

What was the question? Oh yeah right, I'm gonna GPL this shit and put it in browsers. I think I'm going to write it in Kotlin but I'm open to suggestions. Would you guys like to play it/contribute to it's development for shits and giggles?