Now that 1-minute blocks are locked in for May 15, I’d like to ask a practical question from the wallet side. I work in BD and support at Gem Wallet, not as a developer, so this is about what ordinary users see and ask.
Most users won’t follow the CHIP discussions, so whatever they learn about the change will likely come from their wallet, an exchange or a support chat. A few things I’m not sure about:
Before May: should wallets say anything at all, or is the change mostly invisible for people who just send and receive?
Waiting for confirmations: if a wallet shows an estimated wait, what’s a fair way to describe it after the switch? “About a minute” is the average, but people get frustrated when a wait takes three or four times longer than they were told. Showing a typical range might set better expectations.
Exchanges and confirmation counts: many users know “my deposit needs X confirmations.” If exchanges change their required number, users may ask why. Is there a simple way to explain that more confirmations doesn’t mean a longer wait?
Do users need to do anything? For example, update their wallet app before May 15, or watch out for anything if they’re using an older app afterwards?
If there were one short, neutral explainer that wallet and support teams could all link to, that would help avoid mixed messages. Is anything like that planned, maybe on Consensus.cash?
Yeah I plan to draft some document(s) to help this, each aimed at some category of downstream software (wallets, pool software, exchanges, etc.), no later than few moths before May.
For the neutral explainer, I’d suggest a short “what changes for you” section: most users won’t need to do anything, but they should install wallet updates when offered. For confirmation counts, phrase it as “your exchange requires N blocks; faster blocks mean those blocks usually arrive sooner, though the exchange may adjust N to keep the same security.” Avoid promising a fixed wait—say “often a few minutes, sometimes longer” so a three- or four-minute wait doesn’t feel broken.
Well, I’m no expert, but I hope this might be a useful contribution, based on my intuition:
I think each wallet app can choose the way to reflect it that seems most convenient to them, but I like the idea of the 1-2 minute range. Naturally, there will always be exceptions that fall outside that range, so I don’t think this is something that can be fully resolved from within the wallet itself. This is one of the reasons why faster blocks represents an advantage, because you can’t inform the average user of everything, so a “small” frustration of a couple of minutes is better than a big one of an extra 30 minutes.
Ultimately, this also complements the information users might encounter elsewhere—for instance, if a fellow bcasher explains that “average” still allows for occasional, minor exceptions (like a 2- or 5-minute block), or even rarer, longer delays. I think that explanation becomes more convincing when the difference is a few minutes and the average is 1 or 2 minutes.
I think a friend bcasher is the most reliable source for that kind of explanation.
We have to assume that they will. They could multiply the confirmation requirement by 10 or choose an intermediate number. Certainly, if there is any opportunity to request an adjustment to the required number of confirmations, this is it. We can request a dynamic, smaller requirement for relatively lower amounts. They can say no, but it’s up to us to argue; if we don’t try, we won’t succeed.
What could centralized exchanges tell their users?
They could explain that, while they may require more confirmations, the total wait time could actually be shorter. They don’t have to produce a paper, just mention that there is something called “variance”, which depends a little on luck, and that 10 blocks of 1 minute will generally mean less wait time than 1 block of 10 minutes.
That, of course, could be the response of a well-intentioned exchange or one that respects its business too much to lie to its customers. A less scrupulous exchange might omit that, but in any case, we “Bcashers” are here to point it out. Exchanges have the ability to create narratives, but they are not the only ones who can participate in shaping public opinion.
I think that depends a bit on the wallet you use. There are probably several models. Some where the changes fall on the server, others where the application you use needs to update its code.
I think it would be useful to analyze which wallets might require the latter, but (at the risk of being wrong here) fortunately they are probably not the massive multi-currency wallets that everyone uses, because I would think that many of them depend more on servers than on verifying the more detailed characteristics of each block themselves.
What if I use something with more options and BCH-only that needs an updated version by May?
Well, probably a bcasher friend will keep you updated, or maybe you already know about faster blocks.
@bitcoincashautist separate guides for wallets, exchanges and pool software sound ideal. Wallet teams will be watching for them.
@BitcoinCashPodcast the FAQ page and a short video are exactly the kind of thing support teams can link to.
@bchvenezuela good point that it depends on whether a wallet relies on its own servers or checks blocks inside the app. That’s a useful question for each wallet team to answer for its own users. I also like the way you put it: ten 1-minute blocks usually mean less waiting than one 10-minute block.
On the range question, the wait for the next block follows an exponential distribution, so the odds come straight from the protocol rules. Computed for a 1min target, about 63% of blocks arrive within a minute, 90% within 2.3min and 95% within 3. Roughly 1 block in 20 takes longer than 3min and 1 in 55 longer than 4. On the tempnet our listener counted a median gap of 40.0 s over 28,052 blocks, close to the 41.6 s the rules give. So “usually under 2 mins, now and then up to 4” would be honest. I can work out the same for 3 or 6 confirmations if that helps.
Here are the same odds for 3 and 6 confirmations, computed from the 1min target and assuming the transaction makes the next block. For 3 confirmations the median wait is about 2.7 mins, 9 in 10 are done within 5.3 and 99 in 100 within 8.4. For 6 the median is about 5.7 mins, 9 in 10 within 9.3 and 99 in 100 within 13. So “usually under 5 mins, now and then up to 8” for 3 and “usually under 10, now and then up to 13” for 6 would be honest. The more blocks you wait for, the closer the total sticks to the average, which is the variance point @bchvenezuela made.