4 ms·
I think I agree with your sentiment, but saying that nothing is as good as Bitcoin for sending money over the internet, and that’s why it has value, is just rea
by minitoar 6y ago
I think I agree with your sentiment, but saying that nothing is as good as Bitcoin for sending money over the internet, and that’s why it has value, is just really not a compelling argument.
- toss1 6y agoYup. And that key feature of "sending money over the internet" thing hasn't worked out very well so far I've been in it, found the entire system wanting, and got out. May get in again to enjoy gamble in speculative bubbles, but as a technology, it is still in the early and massively-sucking and increasingly-sucking days. Like railroads in the 18th century, they utterly changed society, but most investors lost their shirts along the way. Scalability, inconvenience, insecurity, and transaction fees all suck to varying degrees at varying times. It used to be that transactions would take 5-10 minutes to confirm, and that seemed mildly inconvenient but acceptable due to low costs. Now, it varies wildly, e.g., between 61 and 426 minutes in the last week [1]. I just bought software last night and the transaction cleared in fewer seconds than I could notice, and that network handles orders of magnitude more transactions than does BTC. Costs. BTC transaction costs used to be astonishingly trivial. Now, they fluctuate wildly depending on network congestion, and average well over $20/transaction. [2] This is nuts. A BTC transaction has to be over $750 to be better than break-even compared to a 3% credit card transaction fee. The days of buying a pizza with BTC are long gone, unless you want to pay more in fees than for the pizza. But it might make sense for buying a Tesla, if your fiat-exchange costs on the input side are not too high. Inconvenience and insecurity. You can either keep your BTC with someone else, which means you don't control it ("Not your keys, not your coin", see also multiple exchange hacks and/or exit scams), or you must roll your own. This involves first, extensive research to avoid selecting a wallet that has either been deliberately designed to steal your bitcoin, or just has unknown vulnerabilities. Then, you need to have absolutely solid key management to avoid both having your BTC stolen bay targeted malware attacks, or losing it because you lost your key, crashed your drive, etc. I find this daunting with a solid tech background, getting individual users to successfully and conveniently do it is not going to happen. Paul Graham made an excellent point about this yesterday [3] In short, without MASSIVE improvements, BTC and many other cryptos are nothing more than gambling on vaporware. And that is ignoring the deliberately engineered-in catastrophic energy consumption, which is literally more than the energy consumption of entire countries, when we need desperately to be cutting energy usage to minimize anthropometric global warning Sure IFF [4] the system could be scaled to handle billions of transactions per day at a cost of pennies per transaction (compare with ACH network costs), and with a usable and secure UI/UX, sure it WOULD be fantastic. But over a decade since those promises, all of the metrics: scalability, security, transaction time, costs, are tracking in the wrong direction. I really wish someone could point me to some cruptocurrency that has these solved. I'd genuinely like to see it and would use and advocate for it. But until then, I can only hope they enjoy their gambling. [1] https://www.blockchain.com/charts/avg-confirmation-time https://www.blockchain.com/charts/avg-confirmation-time [2] https://ycharts.com/indicators/bitcoin_average_transaction_fee https://ycharts.com/indicators/bitcoin_average_transaction_f... [3] https://twitter.com/paulg/status/1364986808900202513 https://twitter.com/paulg/status/1364986808900202513 [4] If and Only If
- mitchdoogle 6y agoI know I'm just a single point of data, but I routinely buy things on the Internet with Bitcoin that cost $200-300 USD. I always set the fee to just under a dollar and have never had any issues with the transaction being confirmed within 20-30 minutes, including in the last few weeks. Of course, paying with Bitcoin saves the seller about 3% of the final cost, and they need not worry about chargebacks, which saves them more money in the long run, so they offer a discount when crypto is used. It's a win-win for me and the seller.
- toss1 6y agoInteresting, I was several years ago seeing the transaction delays climb into the hours, but not the fees (ya, maybe I should have set my fees higher, just went for average). I wonder how much variance there is in the network and if you are just lucky, or somehow frequently get picked up in maybe a less busy part of the network, or if there's some attribute you unconsciously use that gives your transactions so much better than average performance? Either way, enjoy your advantage while it lasts!
- udkyo 6y agoI've done multiple txs in the last few weeks - all with "normal fees," 2 took 3 days to get a single confirmation. 1 took ~24 hours, although these were the outliers, several other txs happened in normal time frames. It's possible the app I use is just bad at calculating fees, but I'm thinking yeah that guy got lucky.
- 8note 6y agoIt's only a win for you if they weren't a scammer, or they didnt send you a broken item. Otherwise you've lost the chargeback feature
- tekknik 6y ago> Of course, paying with Bitcoin saves the seller about 3% of the final cost, and they need not worry about chargebacks, which saves them more money in the long run, so they offer a discount when crypto is used. It's a win-win for me and the seller. But where are the benefits for the buyer? I have to go out of my way to buy BTC, when I’m paid in USD, and the value changes so quick that if I don’t want to hold BTC the buying the exact amount will be difficult, get to deal with fun taxes at the end of the year, and all of this because “fiat is bad” and to give up my rights to a chargeback. And then one day SHA256 will be broken, as all hashing functions eventually are, and then what?