Keep in mind that this allows, and protects, the academic channel to do their own thing. Right now they are pulled by those that don't care about academia because they want a job and know having the credential is how they get it.
If we begin doing this, and as a result, none of the academics can get a job as they are all filled by the 'embedded' students, I don't mind seeing how that shifts everything.
This aligns the incentives. Education provides positive externalities that the market will not provide.
> allow companies to direct the research towards their ends
So long as they hit the requirements to reward the prize, I don't much care what else the students are driven to do. One might ask how does this differ from when a company hires a fresh graduate? In fact, I wouldn't mind also applying this system to those that have degrees, and are trying to advance in their careers.
My only concern is that the students would then be able to be hire-able by other companies in their field. If that answer is mostly yes, the system is working as intended. If it's not, the requirements would be raised until they are.
It's probably cleaner to directly couple those that want it, so that those that want to decouple can be left to the academic path. It's pretty normal to hear from people that when they entered the workforce they learned way more than what they learned at college. Let's incentivize this. Governments should directly subsidize education.
Instead of getting student loans and going to college for 4 years where you don't accomplish much, companies should directly offer high school students 4 year full paid internships. At the end the student needs to go through a test from an expert in that field, possibly a professor lead examination, and to the degree that they pass the company and student share in a large prize. If they pass that test, they officially get the degree.
You're not trying to get individuals to complete work that requires thousands of participants. You're trying to get it so that those one level below currently working on such projects can make it pencil out.
You're trying to give individual researchers, who believe they have a chance at a new strategy, to take their new strategy and sell equity to the prize they might win to a new firm.
Governments can always offer prizes, and any model that sufficiently meets the criteria of the prize would win it and claim a large cash prize. Once claimed, the model would then be free to all forever.
Yes, businesses have the same flaw with a bit of better incentives.
With a prize based approach, what you want is for researchers to make small targetted gambles early in their career, earn their prize profits from those gambles, and work their way up into bigger gambles. There's nothing crazy about this scheme, it's how businesses have formed for hundreds of years. If a company wants to buy equity in the prizes of an independent researcher, I see no issue with this.
> then why would this model not be possible in a government office
I see no issue with governmental workers, or those from a university medial research center, or any other persons who can achieve the results from receiving this prize.
> someone needs to pay for all of the trials that don't work
The researchers. They are the one taking the risk. They are the one who would get the rewards. Skin in the game.
> With a public funding scheme, the benefit analysis could be based on public health interests.
Yes, we should directly pay the organization, or persons, based on the value they generate from their own. The bigger the public health interest, the more they should be paid. Right now we give a tiny piece to these people, and thus harm our incentives. Fully fund the results. It's risk-free.
Prizes pay for working results. Grants pay for possibilities. Grants are therefore riskier, and thus we allocate less resources to them.
Since the people with the money don't understand the science, these possibilities must then be assessed by bureaucrats, and this causes our best to spend half of their time writing proposals instead of working and researching. A complete waste of time. Let the people who know the most about their subject freedom to take risks, and then they are given the spoils of their rewards if they are proven correct.
Prizes are much more efficient than grants. Prizes should be given to academics according to the value they produced. I have no issue if the academics choose to spend some of the windfall profits of their prizes on trials.
This is where prizes come in. If you create a drug and keep it secret, the government can grant you a large prize as an offer to reveal your secret.
This is a much better compromise. The company will end up with more money(and faster!) as a result of this prize than under a patent system - since the patent system induces dead-weight losses, and the government will end up with more lives saved.
A selfish player will claim that they will coordinate with the group, and then vote red in private. A coordinating player will pick what the group chooses, whether that be red or blue. You are talking about a coordinating player here. Yes, in this case if all players agree to red, it's obvious you should all pick red. It's completely safe.
You don't even have to go that far from the original question. If instead of the entire world being a single game, if you have hundreds of millions of sub-games where 9 random people are placed within, what should you do?
Surely some of those groups are going to be filled with selfish red pickers. Should the kind coordinating players still go blue? All the red pickers are going to lie that blue is sensible. I suspect that more coordinators will die in this way than the always blue pickers if every coordinating player went red.
So now the full-world version only has the law of large numbers on their side, but they have no way of knowing just what percentage of the population is a selfish red picker. Going for team blue is the much riskier option that can yield catastrophe.
Your concern can easily be addressed by having a 100% tax on the value of land and then distributing this tax money through a citizen's dividend.
But this isn't all that related to the productivity dividend. That's more natural by setting banks to have a full reserve requirement. At that point new innovations would yield deflation, which would be harmful. The Fed would then have to step up and be the creator and distributor of new money. That money would then be the UBI, distributed equally to all.
Landlords should be pushing for an LVT so that in a downturn that can sync up all at once it is the local government that needs to take that hit. It makes no sense that they currently are the ones to take on society level risks. The LVT is a de-risk to their business. In a downturn they would have no capital losses, and instead only see land taxes fall.
For those that don't know, an LVT causes land prices to drop, where a tax on 100% of the rental value of the land would cause land prices to fall to zero. This would allow landlords who were able to own many more properties, and could use their funds building out extensive retail spaces, and have many tenants pay them rents.
The disconnect here is that they built it this way, but still call it a question and answer site and give a lot of power over to the person who created the question. They get to mark an answer as the solution for themselves, even if the people coming from Google have another answer as the solution.
If they were to recreate the site and frame it as a symptom and issue site, which is what the interview described, that would yield many different choices on how to navigate the site, and it would do a lot better. In particular, what happens when two different issues have the same symptom. Right now, that question is closed as a duplicate. Under a symptom and issue site, it's obvious that both should stay as distinct issues.
Riolu is a uniquely terrible example. While he used Cheat Engine to slow down gameplay, he could have just as easily used TAS to record and replay his inputs since TrackMania is deterministic. This is still possible today. This will always be possible even with Kernel level anti-cheats.
I'll note here that the work that Nadeo has done on the matchmaking aspect is in line with what I'm thinking and should be expanded throughout the online gaming space. A division 10 COTD player will never see a cheater. If cheaters do show up, as they commonly do in Weekly Shorts leaderboards, the community ignores it. Their region leaderboards do a much better job than typical games of bringing the community together and they promote continuity. When top players smurf COTD on a new name, the community sniffs it out within the hour. TM doesn't need anti-cheat.
The problem here appears to be the banning. If the cheaters are never banned then they will continue to only play with other cheaters, and everyone is happy. And in fact, to a normal player I doubt they care very much if the player is legit and smurfing or if they are not legit and cheating. That player ruins the game they are in.
The ranking system needs to be a better determinate of skill, especially early in a new accounts life, so that they can stop harming normal players games. This might mean changes to the rules of a game to allow this to be done better. The match-maker should take this into account, so that if a player does go up against a player that was far from the skill level that they end up at, it should protect that account from being placed with new players for a time so that they can forget about it.
For the example you choose for Chess, you might force players to do Chess Puzzles before they can queue for their first match. A normal player would then never see any cheaters.
This argument is that the match-making software is incompetent. If what you say is true, and the match-maker could determine skill with any ability, then the cheaters would quickly find that the only people they match-make with is other cheaters. The non-cheaters don't care that the cheaters exist if they never play against them.
Yeah, this is pretty clear. The community for any competitive game if you are a member of the top 100 players is always amazing. These players play the most, they end of seeing each other over and over, and you build up a rapport with the other players and can start to play against specific peoples play-styles.
However, for the vast vast majority of the player-base who is top 50% in skill, the fat normal distribution nearly guarantees that most of the people they play against will never be seen again. And therefore there is no harm for them not to be toxic to them, so most people only ever experience toxicity in online competitive games.
Server browser games solve this because players end up with "home" servers where they come back to over and over, and over time build communities who do the same. This was taken away from the players when we moved to matchmaking, and many in the player-base have a bias against matchmaking because of it.
But this is in no way required, and merely a result of gaming companies to do any work on this front. It would be extremely easy for these games to add an arbitrary community tag to the matchmaker that would attempt to put people in games with players that they have not previously reported. The matchmaker might take a little bit more time, but since these players are in the fat normal distribution, their average matchmaking times will still be incredibly low.
This is why the best form of UBI is a Citizen's Dividend funded through a Land Value Tax. Any increase in rents through the CD just make their way back into higher taxes that then raise the CD.
What this is going to allow is a way to flatten org-mode, which will massively expand the amount of people willing to use it. Put anything you wish into your own data collection, and you can instantly pull it up with a prompt. That service would then allow anonymized queries of other peoples data.
If we don't get AGI, the LLM that are starting now and don't have fresh data from people's queries won't be able to get going. The internet will quickly become stale. This will be sped up by the spam that the LLM will be used to create.
Walking through this scenario I don't see anyway for this not to end in a network effect monopoly where one or two services wins.
Is this going to end up into a single model, where its trained on text and images and audio and videos and 3d models, and it can do anything to anything depending on what you ask of it? Feels like the cross-training would help yield stronger results.
If we begin doing this, and as a result, none of the academics can get a job as they are all filled by the 'embedded' students, I don't mind seeing how that shifts everything.