AI is not niche. Blockchain ledgers are because centralized ledgers are cheaper, faster, and controllable by law; which is what most people want if they spend a few seconds thinking about it.
This is the typical HN 2015 crypto knowledge. It was accurate a decade ago but isn’t any more.
- Centralised ledgers are multiple orders of magnitude more expensive (2.5% to 6%, a typical blue and white square checkout is 3.5%) versus something like $0.000025 on the most active blockchain)
- At their best (2-3 second confirmation) as fast as current gen blockchain networks and an order of magnitude slower than next generation (150 milliseconds block time so expected subsecond confirmations).
- Tokens have techniques like permanent delegate for OFAC compliance.
This isn’t meant to be a personal attack, it’s just that this view of crypto is akin to saying that ‘AI is customer service chatbots that don’t work’ - correct ten years ago but not anymore.
Axiom, YC W25 is the fastest growing company in YC history hitting 100 million in revenue in five months.
You're comparing different things to try to prove something that is obviously wrong to anyone who spends a few seconds thinking about it.
> Centralised ledgers are multiple orders of magnitude more expensive (2.5% to 6%, a typical blue and white square checkout is 3.5%) versus something like $0.000025 on the most active blockchain)
I was comparing cost of recording the transaction. Think for 2 seconds. Obviously, a centralized ledger is going to be cheaper. You are comparing the cost for completing a transaction on one side with the cost of completing a transaction plus fraud fraud mitigation, chargebacks, etc. on the other.
> At their best (2-3 second confirmation) as fast as current gen blockchain networks and an order of magnitude slower than next generation (150 milliseconds block time so expected subsecond confirmations).
Same mistake. Think for two seconds. Obviously, the speed of recording a transaction on a centralized ledger is going to be faster.
Whatever you build on a blockchain ledger you can build faster and cheaper on a centralized ledger.
People fooled by crypto grifters don't have enough economics education to understand "ceteris paribus" let alone everything that comes after in an introductory course.
I’d like to start by saying “think for two seconds” is not a respectful way to communicate.
Do I need fraud mitigation and insurance to buy a coffee or groceries?
Regardless of the capabilities of centralised networks when you last bought something using a Visa card was it hundreds of milliseconds or was it two or three seconds to confirm?
> People fooled by crypto grifters don't have enough economics education
That’s a very broad statement about a lot of people highly regarded in traditional finance.
> I’d like to start by saying “think for two seconds” is not a respectful way to communicate.
I'm assuming the readers here have enough technological sophistication to understand the problem in two seconds. This is not the case among general YouTube audiences being fooled by crypto grifters.
> Do I need fraud mitigation and insurance to buy a coffee or groceries?
Do you want to be on the hook for somebody else buying coffee and groceries on your dime? Most people don't. Thus, fraud mitigation.
> That’s a very broad statement about a lot of people highly regarded in traditional finance.
That's assuming that those "highly regarded" people are fooled by crypto grifters instead of profiting off the fools.
Do you think not engaging with the actual content of the poster and purposefully misinterpreting them disingenuously is a respectful way to communicate?
Don't be silly. Just because the Crypto/Blockchain promises of grandeur turned out to be mostly nothing, and the "fuck regulation, we are all the <hotel, taxi driver, you name it>" "gig economy" companies are losing customers as they show their true colours, and the AI bubble starts to burst, that doesn't mean that over-hyped niche "technologies" can't rake in millions of some poor VC investors money.
It just requires lateral thought.
Pay the completely untrained gig-workers peanuts to review AI written code to create a universal blockchain "bank without borders". I mean what could possibly go wrong?
To push back at your obvious facetiousness, I do think that both cryptocurrencies and machine learning are very powerful and useful technologies. Underneath the scammers, grifters and investors that jump on the hype train for their get-rich-quick schemes, and the general public that falls for it and fuels the hype train, both technologies have solid reasons for existing, and can be generally very useful to humanity.
So I reject the notion of throwing the baby out with the bath water as much as the hype bubble around them.
What we do need, as with any novel technology, is oversight and regulation. Which is difficult this time around when the grifters are also the ones in power.
> So I reject the notion of throwing the baby out with the bath water as much as the hype bubble around them.
I never mentioned throwing them out; I specifically mentioned that they are niche - they have useful applications in very specific domains.
That use isn't dependent on hype, because they don't need a bunch of investment dollars to somehow find a use-case and create artificial demand.
As an example, someone proposed a while ago that banks could use a blockchain to facilitate near instant foreign exchanges; for the sake of argument let's say that's a feasible thing for them to use it for, and is an improvement over their existing system for them.
The customer doesn't need to know and likely doesn't care that it uses a blockchain to facilitate the transfer. They care that the transfer was much quicker.
Eh, your entire comment was dismissive about the "technologies", and not unlike countless similar reactions about the topic. So any serious points you wanted to make were easy to miss within the sarcasm.
I do agree with your point that valid use cases don't need to rely on hype, but at the same time, the technology that powers it doesn't need to be invisible to users. Some amount of marketing around it shouldn't be an issue. The thing is that we've been flooded with it for the past decade+, and the critical thinking public has become desensitized to the bullshit. But this doesn't mean that underneath that there is zero legitimate value, as some people vocally claim.
Transparency logs like [Certificate Transparency](https://datatracker.ietf.org/doc/html/rfc6962) use permissioned chains (and other things) to distribute trust in the internet public key infrastructure.
My understanding is that CT uses Merkle trees, which can also be used in a Blockchain, but that seems to be where the association dies. Just because two things use the same underlying data type doesn't mean one is a use of the other.
My point is: when was the last time when K-means clustering or the Bayesian theorem got a cult, spawned a 10000 fake currencies and were touted as a panacea for all societal problems? The hype behind blockchain, which is just a distributed mathematically sound log of stuff, is just a bunch of misinformation shouted loudly by charlatans and uninformed enthusiasts.
It's the same thing with LLMs, they are basically an extremely sophisticated autocomplete, but I'll be damned if I don't see 1911 articles per day that claim they are sentient, and Sam the perv needs just a couple more billion dollars to reach AGI, just behind the corner. The average pundit's critical thinking is just so disappointing.
I'm not sure if you meant to reply to me? I generally agree with your points, but my specific reply was dismissing the idea that Certificate Transparency is a good use of Blockchain..
The problem is the crypto isn't used as a currency. It's used as a security, and a very speculative one at that. We don't need more securities, sorry.
Also, the benefits of crypto as a currency aren't all that. We already have instant money transfers that everyone uses day-to-day without any hitch. Crypto can do it too, sure. Keyword too.
There can be other benefits like transparency or anonymity but:
1. We can already achieve transparency through regulation and largely have.
2. Anonymity for currencies is pretty much bad. We tried that in the past, turns out lots of organized crime.
> The problem is the crypto isn't used as a currency.
False.
> We don't need more securities, sorry.
That's your opinion.
> We already have instant money transfers that everyone uses day-to-day without any hitch.
We do? Please let me know of an instant way to send money to any country on Earth, without exorbitant fees or trusting a single corporation.
> Anonymity for currencies is pretty much bad. We tried that in the past, turns out lots of organized crime.
Ah, yes. Let's abolish cash as well.
It's easy to strawman the case that all cryptocurrencies have no practical value. Groupthink is promoted on forums like this. What takes a bit of maturity and honesty is acknowledging that beyond the negative things cryptocurrencies have been used for, the technology is fulfilling the promise of evolving our financial systems. While you continue to scream and shout that it's a scam, millions of people will continue to benefit from it every day.
Okay, but most people aren't using bitcoin as currency.
> That's your opinion.
Okay, and most people's opinion. Yeah we don't need more securities, especially scammy ones. Sorry you're in the minority.
> We do? Please let me know of an instant way to send money to any country on Earth, without exorbitant fees or trusting a single corporation.
I don't pay for my fucking coffee in Venezuela. 99.99% of transfers don't go across countries.
And, if you need that, that's what credit cards are for. They're fast and I'm not going to get kidnapped and have my fingers stolen because I have an Amex blue card.
> It's easy to strawman
Really? You don't think appealing to usecases like "inter-country transfers" isn't strawmanning?
The reality is crypto-currency has exceedingly small usecases that aren't already covered by other technologies. And then, of those that are covered, the alternatives are proven, faster, cheaper, and safer.
> Ah, yes. Let's abolish cash as well.
Nobody ever said this, but we do put limits on cash to prevent laundering.
And it's saved a lot of lives. Sorry, I don't think letting organized crime flourish is worth for some principle of privacy you have. I don't care about you or your principles, I care about real outcomes, in the real world. And, fortunately, legislatures in pretty much every country agree with me.
It's hard to see the positives considering current cryptocurrencies are 99% rug-n-pull ponzi schemes with the longevity of a mayfly, and the ones sticking around like bitcoin, eth, etc, are just a long-run-con with a value volatility rivaling only the volatility of the currently in-fashion javascript framework.
I mean, most modern financial instruments are fairly similar, albeit slightly more regulated, so what do I know. In a short span, we have witnessed speculative buying of GPUs, crypto mining (aka using shit ton of electricity and hardware to produce NOTHING of value), dropshipping, and my favorite: the tremendous value that NFTs and all other web3 trash brought to society, I can't help but wonder why people are tired of the crypto fads, especially since crypto delivered NOTHING of its promises:
- you can buy limited amount of stuff with it and you often need to convert it to "real" money
- its unstable and unregulated, so it's just a breeding ground for scammers and snake oil salesmen
- It fails hard on its privacy/anonymity promise, as it's LITERALLY the most traceable thing in the world, especially considering wallet scramblers are outlawed in many places
So much negativity for no reason, all web3 so far has been super successful and extremely useful for society! /s
> Are people not finding these tools useful, even with the significant investment and hype in the space?
How exactly would someone find hype useful?
Hell, even the investment part is questionable in an industry that's known for "fake it till you make it" and "thanks for the journey" messages when it's inevitably bought by someone else and changes dramatically or is shut down.
It is slowly improving (albeit with some egregious bugs, like losing EFI data on export) but TBH even their x86 product pales in comparison to Parallels or VMWare Fusion, in terms of machine performance.
For those unfamiliar, it's a Lua library that gets executed in Redis using one of the various language bindings (which are essentially wrappers around calling the Lua methods).
With our multi-node redis setup it seems to be quite reliable.
How so?
GP complained about JS lack of types. I pointed out that most JS actually benefits from types, given it's typically authored in TS. No moving goalposts, no "true Scotsman" arg.
My point was agreeing with the GP's underlying point that types matter. Yes it'd be crazy to do serious programming work in JS without types -- that's why ~nobody does that. It's why TypeScript exists and ~everybody doing serious work on JS projects uses it. Their argument was almost a strawman, it refers to a hypothetical situation that doesn't pertain. Serious JS programmers use TS.
That ruby doesn't have types is also bizarre to me, but Ruby also sees monkey patching code as a positive thing too so I've given up trying to understand its appeal.
I appreciate that tools like Git Bug are not tied to a single repo host but they are tied inherently to a single VCS (git). BE was not reliant on a specific VCS, and it's a *really* simple format on-disk too.
With some motivation you could port git-bug to another VCS without too much problem. You would need to implement those interfaces [1]. The one you care about especially is RepoData, which mainly imply you can store a DAG, have references and push/pull. I believe other VCS (say mercurial) have similar concepts.
Or you could just as well plug a generic database there.
SEEKING WORK - Currently Remote from Thailand (UTC+7), with potential for short on-site trips if necessary. Relocating to Australia soon(ish).
I have 20 years experience across a mixture of Ops/Infrastructure/Dev-Tools and web application architecture/development, with a focus on solving unique/uncommon problems.
I’m looking for opportunities to help companies with
- server infrastructure setup/maintenance;
- backend/server-side architecture & development (primarily PHP MVC);
- system integration;
I currently work remotely from Thailand, on a reasonably flexible schedule (I can schedule work hours to get overlap with most other timezones when required) but on-site visits to pretty much anywhere are a possibility if required too.
Contact me via email (in my profile) if you want to have a chat about what you’re working on!
You mean the overhyped extremely niche technologies?