Hacker News Reader

Comment by transfire | Hacker News Reader

I thought Bitcoin had run into a fundamental problem that transactions were now taking hours co complete. And that an internal battle among developers on the fix remained unresolved. Did I miss something?

View on HN

The Bitcoin network works perfectly fine and always has. The issue you are referring to was during a DOS attack on the network, which still wasn't an issue if you paid slightly higher fees to prioritize your transaction. The developer split on the blocksize increase was pretty heated for a while, but has largely blown over due to Bitcoin still working and the sky refusing to fall as some predicted. However, scaling is still a major issue that some very smart people are working on.

If Bitcoin users were naturally making more and more payments or there were more and more users, and then it ran up against the block size limit, how would you know whether there was a DOS attack or not? Wouldn't it look like a permanent DOS attack?

Did it blow over because the block size limit was increased?

I call it a DOS attack because blocks aren't regularly full anymore and the transaction data showed that the flood of transactions were "peeling chains", meaning they mostly had the same source. There is blockchain data to back this up.

The blocksize hasn't been increased yet. It blew over due to wallets improving their fee support. If you are in a hurry and need quick confirmations you increase your fee by 10 cents, but if the transaction doesn't need priority you may wait for quite a while.

Currently the calculations show that for your transaction to be processed first it should have a $.06 fee. Transactions with fees down to about $.01 are basically guaranteed to be processed but may take longer. Many 0 fee transactions are processed but they have no guarantee.

The modern wallet softwares all calculate and offer different fees based on the expediency of the transaction.

"The fastest and cheapest transaction fee is currently 60 satoshis/byte, shown in green at the top. For the median transaction size of 226 bytes, this results in a fee of 13,560 satoshis (0.06$)."

I think it was more that the amount that you need to pay to get your transactions to complete in a reasonable amount of time has gotten higher.

That's not a problem for Steam, since the fees are still much lower than credit card processing fees for larger transactions (as in at least several dollars). It is a big problem for people who have dreams of micro-transactions via bitcoins.

Nah. That was just media insanity. Bitcoin functions just fine in the face of crazy politics. The whole point of it is to be unkillable.

Bitpay (steam's payment processor) accepts 0-confirmation transactions

Doesn't that open Bitpay up to double-spending attacks?

Bitpay tells Steam it was paid right away. Steam gives you access. Bitpay tells Steam when it get confirmed, and then pays out.

If it gets doublespent, they tell Steam that, and Steam simply revokes the sale, because it is digital and linked to the account.

They watch the network for red flags, good propagation fights new transactions

Because credit card payments and wire transfers take less than hours to complete?

Yes, credit card payments, CNP, usually takes seconds to complete. I may take days (or month depends on the country) to settle.

Exactly like bitcoin payments then, except bitcoin settlements only take hours.

Reply on news.ycombinator.com