Hey Eltonlin, this sounds really cool and I hope you can make an impact in improving access to education. Some time ago I wrote a blog post titled "Professors as Creators" which discusses the impact the Creator Economy could have on education - how in other fields (e.g. writing and videos) creators are increasingly leaving large organisations like newspapers and film studios to create content on their own using platforms such as Substack and YouTube, and how this could impact education next. I think you might like it as your platform designed around these ideas.
I'll be interested to see how the longevity of e-sports pans out. Most games will be woefully outdated after a decade: will the audience move to the next big game, or stay playing the old games? Or will video games eventually plateau to a point where significant technological advancements are no longer made?
This year saw a time where Chess became the top trending app on the App Store. The strength of traditional sports and games is their timelessness - can e-sports ever match that?
abstract strategy games like chess, go, and hex are fun ti play but difficult to watch for the layman. You tend to just be watching people think rather than the game itself unless you’re really well versed in it.
Games like counter strike and StarCraft are still entertaining to watch even if you don’t understand the nuances.
Hi, I'm the author of the blog post. Just wanted to say, I would be very interested in reading more about the experiments you ran on getting GPT-4 to describe its plans. GPT-3's explanations were confidently incorrect, as usual.
Hi, since you're commenting here and appear to be interested in exploring things further, I just wanted to point out that once GPT-4 is available through the API you could use a LangChain Agent[1] to maintain the board state externally and feed it back in automatically with every new move, so that the playing field would be more level in terms of memory. You could also inject instructions about explaining its plans either as system messages or as per-prompt instructions.
Hi all, today I'm launching a side-project I've been working on for a while: Web3 Readers' Club. I think the internet lacks spaces for discussion: most sites operate in the form of public posts where the goal is to get you post as much visibility and attention as possible. I think this creates a lot of toxicity as people are incentivised to write "hot takes" that capture the most attention. This is especially true for crypto and Web3, where people have a monetary incentive to shill projects they bought into.
Web3 Readers' Club tries to solve this problem by building around semi-private group chats. This makes it easy to discuss topics in small groups, without the need to perform for a large audience.
We're just starting out so for now this is just an experiment, but we'll see where it goes.
I saw lots of people criticizing PG's essay, but I really liked it. Free will and agency are interesting philosophical topics to wonder, even if they don't hold much practical value.
I think there's something to comparing education (especially when delivered remotely) to YouTube. There's a huge variance between the teaching skills of teachers, just as there is variance between how entertaining YouTubers are. The difference is, the best teachers get paid the same as average ones, while nobody knows the name of an average creator.
I wrote briefly about the topic in a blog post, titled "Professors as Creators". It explores the idea briefly, and how treating teachers as "creators" could add value not just to remote learning, but to in-person lessons as well. https://medium.com/age-of-awareness/professors-as-creators-h...
Hey crawfordcomeaux, I noticed you've shared some interesting ideas about communities in Hacker News! I wonder if you have an opinion on whether remote working will change where and how people choose to live?
Cities still have a lot going for them by providing the best access to services (especially after covid is over), but one thing I see happening in the future is the establishment of "remote working villages" in smaller towns. Price and quality of housing will be the main selling point, but they'd also have access to nature and a good sense of community (in fact the projects could be self-organised and funded by the residents if you find the right people).
This concept is important to the way I think about crypto. It feels like a war between two elites (TradFi vs Crypto insiders and VCs), if crypto wins it it will not "democratize finance" when 90% of all Bitcoins have been mined in the first decade of the project's lifespan.
(This is not an attack against crypto in general, there are ways to design a cryptocurrency so it's equal to all participants regardless of the time they buy in, it's just that they fail to gain adoption because of a lack of VC funding and support from the crypto community, as neither can make a quick buck out of it)
Well it is not really a war, so much as a rebalancing of power dynmics within Tradfi circles. You can look at the impact of computer on capital markets since the early eighties through the evolution of the social structure of investment banks labor force. It slowly became less and less blue blooded thanks to the need for quantitative skills.
But there was still a massive moat for institutions due to the need for trust, backed by a regulatory framework. And entrenched institutions means entrenched elite. What crypto provide is a temporary workaround for a faction of the financial elite to build rival institutions.
But the crucial thing is member of the new "crypto" elite are for the most part "junior" (in the status sense) members of the Tradfi elite. So for the rest of us peons, it will stay business as usual !
Thanks for the great article! I wished the section on DeFi addressed the biggest elephant in the room, overcollateralization. A DeFi borrower always needs to lock in more assets as collateral than what they are borrowing, thus defeating the whole purpose for taking a loan aside from financial speculation. And there's no real way to solve it: all mechanisms I've heard of either try to replicate some form of background checks on borrowers (not decentralised, and uses the real world), or claim to use real-world assets as NFT collateral (once again, uses the real world). Thus I can't see a way for DeFi loans to ever be used for mortgages and other loans regular people actually need.
Of course DeFi is more than just loans and borrowing, but that's an aspect that's constantly promoted by evangelists so I think it's an important point to mention.
This is a great point. Some ways I'm seeing this addressed are in the form of,
- Credit ratings from the real world being moved on chain, both by startups and by established companies (see Fitch report from Oct 21, pay-to-access only).
- Credit ratings across and on chains (https://www.credprotocol.com/) to make sure that a defi OG on eth can have access to collateral on other chains.
- Chainlink and others are working on products like CCIP (https://chain.link/cross-chain) to facilitate creating more systems like this through Oracles vs having to be reinvented.
I would imagine this state of affairs gets better over time.
Why would anyone want their credit history on a public blockchain?
Besides, credit histories often have incorrect information that needs to be corrected, and often creditors will cut you a deal or just be nice ("goodwill adjustments"). Blockchain history is immutable, so how do you handle this? Sure, you can write something to the blockchain that amends an existing entry, but the existing entry is still there to see, even if it was erroneous. I would bet that some lenders would use "has more than $SOME_NUMBER amended entries in history" as a negative signal, even though it's not necessarily fair to characterize them that way.
Hell, negative things on credit reports (in the US) disappear after 7 years. Sure, you can tell lenders to disregard negative things older than 7 years, but do you really expect them to do that?
Great point. I think having this data encrypted on a public system for anyone to verify will lead to much more openness in the ratings system. It has the potential to be a far cry than the black hole the US system is today. I don’t pretend to predict the future, but if you look at the past, any time “open standards” have replaced closed black box systems - it has generally benefitted the end consumer.
With zero knowledge proofs you can prove that you have a specific credit rating or above to anyone who needs to know, without having to reveal what your credit rating is.
Bad marks on credit reports fall off after certain amount of time. you can also dispute it in court or settle it for less than owed. None of which translate to a blockchain and which beg the question of privacy when it comes to collections outside of generic consumer transactions.
I think DeFi basically justifies the high market cap of ethereum. You have a decentralized exchange working 24h, with basically no downtime.
Compare it to normal exchanges:
- need to register
- need kyc
- hard to tranfer LARGE sums cheaply
- your bank can block your assets as well as your gouverment
- fees are sometimes very high as well. Example tried to convert one currency to an other 1% fees in europe. I think you would pay less in a DeFi setup.
- Now imagine you could buy stocks and so on on defi. That would be a killer application.
I could go on but the point is. DeFi solves a problem.
I have trouble imagining who might want to do this. It seems like a really small niche? Normal people don't need to trade or move large sums of money 24-7. An ATM and/or credit card is good enough.
I mean 24-7 comes for free. Security comes in some way for free as well.
Large amount: its relative, so even for 10k USD you would pay 100 USD as fee. You get similar fees for purchasing stocks and many have more then 100k in equity. So it isn't really a small niche. Rather a huge market that may get more simple and cheap through DeFi.
Ok maybe in the us. Anyway are you sure it is really free in regards that they don't give you a worse price or won't execute if things move in the wrong direction like Robinhood did...? My point is, its just inconvenient for me to rely on a third party without able to enter the "true" market on myself.
The question is, worse price than what? In the US you’re supposed to get at least the best price quoted at any stock exchange, the “nationally best bid and offer” (NBBO). Surprisingly, you can often do better, but this is still a good price. I don’t think you can easily do that with cryptocurrency where you normally only trade at one exchange?
Maybe relying on a third party is useful if they can get you better prices without needing accounts with all the exchanges? And brokerages normally give you a better price than any exchange. This is called price improvement. [1]
Robinhood got in trouble for falsely claiming that they get better prices than their competitors and not disclosing that they get payment for order flow, but their customers still got the NBBO I think? Although see [2].
People make a big deal about payment for order flow because they don’t like the idea of someone else making money off them, but that doesn’t mean you got a bad price.
Actually you can do it for cryptocurrencies as well. There are a lot of aggregators that execute or split your order across different exchanges.
As you said yourself, brokers may offer somehow better prices. My point is just, it is not possible to interact directly with the exchange, even if it maybe for some people more convenient to use a broker. There is no free lunch, more third parties means in my opinion worse price or higher fees.
You'd think that, but apparently retail customers get better prices because market makers are more likely to lose money trading against a professional who knows something is coming, and the smaller spread from lower risk for them is enough to pay for the business model. They still make money on the spread but it's lower.
It's a bit weird that it works that way, but there is a logic to it. Price discrimination can work in your favor.
(And having gotten many literally free lunches from an employer, there is no harm to it if you know why they do it.)
Got it. Sounds interesting.
Just one remark about it, europe seems to be banning practices related selling the order flow. As you said it sounds a bit weird. Lets say non transparent.
So maybe you don't always get a good deal.
PS:
Crypto isn't always better in this regard since, there are some frontrunners and so on. The difference is that it is completely transparent.
This is still mostly restricted to those living in the USA. It's also only a fraction of the investment market too, to invest in higher risk things such as startups you need to already have a net worth of $2M or make $250k/year+ which is only the top 1% of people. I'd like to see a lot more democratization of finance than that.
> your bank can block your assets as well as your government
It wasn't until recently that governments (arguably) became good at using technology to physically regulate banking activity, and it's only because the financial system became centralized enough for them to gain visibility into what was going on. Before then, and - to be honest, even now - government regulation was operationalized through the threat of repercussions and consequences for non-compliance.
Aside from reducing visibility into the transactions themselves, I don't understand what, exactly, DeFi or any of the other acronyms for ledger based systems do to stop this. Does DeFi, crypto, or whatever it's being called at the moment make it easier to avoid government oversight? Sure, but if you are doing something illegal in the process and get caught, there will still be legal repercussions.
> hard to transfer LARGE sums cheaply... fees are sometimes very high as well. Example tried to convert one currency to an other 1% fees in europe. I think you would pay less in a DeFi setup.
Is this still true? I'm pretty (very) sure that Interactive Brokers and other platforms allow for extremely cost effective currency transfers.
1% is true, for a lot of banks in my country. IB may be an exception, without going into details I had trouble with their kyc.
DeFi is a buzz word. Lets take Uniswap. You get direct access to an exchange. It works 100% of the time. You pay the same fees as everybody else.
Traditional exchanges: try to get the same fees as a HFT shops. Yeah forget it. Even through IB you don't get direct participation on the exchange. So in my perspective the traditional system is kind of rigged in the way how different actors are able to participate. At least at the moment DeFi is fair in this regard. Also you get full access to the assets you are holding without any third party..
A more important limitation for defi loans is that nothing other than blockchain assets can be used as collateral. You can't use the loan to buy anything real without having enough money to buy it already.
A mortgage lets you buy and live in a house even if you don't have enough money to buy the house.
That is a bit too simplified. You can add yield bearing tokens as collateral. That basically means that as long as the yeild is more than the cost of the loan the loan wil pay back it self.
Hmm, that's sort of interesting but it seems like it presumes you own the income stream to begin with, and that it's already been turned into a cryptocurrency token. What are some examples?
Not totally true. You could, for example, use as collateral tokens that represent staked Ethereum (rocket pool ETH for example) so you are effectively using the ETH to generate staking yield, and also using it as collateral simultaneously.
The ability to wrap ownership in a smart contract can extend to pretty much any use case, as long as the lending protocol supports using the wrapped token as collateral.
CREAM's Ironbank does undercollateralied/uncollateralized loans. They approve lending to protocols but they can easily decide to take risk on individuals if they want.
Likely there will be competition, and bad actors causing no recourse. But non-performing loans are easy to see so much faster than the traditional system, could easily limit losses with lending caps.
I didnt really understand the part about loans only usuable for financial speculation, you can cash those loan proceeds out for USD to do whatever you want. But even if you couldnt, financial speculation is called “good debt” anyway, compared to consumptive spending, so I dint understand the criticism standard here.
Can you provide an example of someone claiming defi can provide solutions for loans?
The best I can come up with logically is something like cardano's prism that effectively puts a persons encrypted identity on the blockchain which allows that person to reveal things like their grades etc. to future partners as a means of quickly establishing trust. Maybe something like that could enable defi loans? Even that is a stretch.
Searching for "DeFi loan" returns a deluge of results with https://defirate.com/loan/ at the top for me. They seem content claiming that secured loans are useful loans.
Pretty interesting. A lot of people here are focusing on permanence, but for me the main difference on addressing by content vs by host name is the loss of authorship it opens up for the web. Since a research paper of essay is referred to by its hash, the owner effectively gives away all control of the work when it's first published. There will be no editing, no taking down unwanted work, and no real way to build an interactive website that allows dynamic linking to other materials by the author.
It's interesting how the same people promoting the "creator economy" also tend to promote the cryptocurrency space and IPFS without an ounce of self-awareness. IPFS sounds awful for creators of all kinds in the same way as BitTorrent was awful for artists. I can definitely see a use case for IPFS as a file storage for trustless systems such as smart contracts, which are designed as immutable, trustless systems.
You don't lose authorship, you lose ownership, which arguably you don't have under the regular internet either, ie. Once uploaded, anyone can mirror your content. Arguably under ipfs you literally cannot lose authorship if you put the author in the content because then authorship is part of the content hash used to address/load it.
IPFS claims to solve this with their name service IPNS which can update to point to a new hash with a revised file. Where the original hash can be cached and used but users can refer to the NS version and get the latest version. But last I saw, the name on IPNS had to be frequently pushed by the original server or it would go away.
IPNS doesn't work. Whenever you press them on this point, they'll admit it doesn't work.
But IPFS has the crypto problem of conflating the stuff that works now with the stuff that's hypothetical in its marketing, and not admitting that the latter is janky nonsense that doesn't bloody work.
Again, I'm not sure when it wasn't working, or when it began working (it's always worked since I've been around), but IPNS has made huge strides, I use it every day. Even https://ipfs.io is using IPNS, it's very popular.
Yeah this is what I saw 4 years ago. Shame it still isn't working. Insane how crypto projects spend all this effort on flashy landing pages, marketing, hype. But if you actually try to use the product, you find out it just doesn't actually work.
https://medium.com/age-of-awareness/professors-as-creators-h...