I am attempting to verify I perceive the query: Why Do We Use Change Descriptors when Constructing Wallets?
Context: I work on the Bitcoin Growth Equipment, the place one in every of our core varieties is the Pockets kind. It has sometimes required 2 descriptors (an exterior descriptor and a change (inner) descriptor). This concept is normal follow and used extensively within the trade, however when urgent myself to put in writing about it in our documentation, I used to be left questioning if I actually knew why this was the case. Moreover, customers that request the Pockets be usable with just one descriptor had robust opinions about how the follow of utilizing 2 descriptors was possibly outdated and never required underneath sure circumstances. One consumer instructed us that it was a mistake to double down on the inner/exterior descriptor schism, and that it was a relic of the previous of folks that use electrum wallets.
As a library maintainer, that is the type of factor I need to have a strong grasp on. I am additionally main our documentation efforts, and plan on together with a web page on this.
Right here is my present understanding and questions.
Benefits of utilizing 2 descriptors
- It doubtlessly permits you settle for 0-conf cash when they’re coming to alter addresses, since you should have been the one to provoke that transaction and won’t double-spend your self.
- Privateness when utilizing a public electrum server: in instances the place you would possibly ship a complete xpub to an electrum server, you are higher off not giving up your complete keychain however solely half? (at this level it appears trivial to reconstruct transaction historical past however nonetheless, I suppose not giving it out is healthier than doing it).
- Having only one descriptor forces you to maintain monitor of addresses you have given out when constructing transactions, as a result of in any other case your pockets would possibly use an exterior tackle already given out as change, and if this tackle given out to somebody finally ends up being utilized by them, you may double transactions at that tackle.
- In instances the place a pockets is recovered, change addresses can nonetheless be labeled as change.
Questions I Nonetheless Have
- Are there different pitfalls customers of single-descriptor wallets ought to pay attention to within the case the place wallets use extra personal sync instruments like personal Electrum servers and CBF, and utilizing trendy pockets practices that by no means give out repeat addresses?
- Does the “change” keychain come into coin choice in any respect in Core? For instance, are there algos for coin choice the place understanding if a UTXO is a part of the change keychain reasonably than the exterior keychain issues? If that’s the case, how does it come into play?
- Another historic data that is good to know and consider for builders of pockets can be appreciated.
Thanks!
I am attempting to verify I perceive the query: Why Do We Use Change Descriptors when Constructing Wallets?
Context: I work on the Bitcoin Growth Equipment, the place one in every of our core varieties is the Pockets kind. It has sometimes required 2 descriptors (an exterior descriptor and a change (inner) descriptor). This concept is normal follow and used extensively within the trade, however when urgent myself to put in writing about it in our documentation, I used to be left questioning if I actually knew why this was the case. Moreover, customers that request the Pockets be usable with just one descriptor had robust opinions about how the follow of utilizing 2 descriptors was possibly outdated and never required underneath sure circumstances. One consumer instructed us that it was a mistake to double down on the inner/exterior descriptor schism, and that it was a relic of the previous of folks that use electrum wallets.
As a library maintainer, that is the type of factor I need to have a strong grasp on. I am additionally main our documentation efforts, and plan on together with a web page on this.
Right here is my present understanding and questions.
Benefits of utilizing 2 descriptors
- It doubtlessly permits you settle for 0-conf cash when they’re coming to alter addresses, since you should have been the one to provoke that transaction and won’t double-spend your self.
- Privateness when utilizing a public electrum server: in instances the place you would possibly ship a complete xpub to an electrum server, you are higher off not giving up your complete keychain however solely half? (at this level it appears trivial to reconstruct transaction historical past however nonetheless, I suppose not giving it out is healthier than doing it).
- Having only one descriptor forces you to maintain monitor of addresses you have given out when constructing transactions, as a result of in any other case your pockets would possibly use an exterior tackle already given out as change, and if this tackle given out to somebody finally ends up being utilized by them, you may double transactions at that tackle.
- In instances the place a pockets is recovered, change addresses can nonetheless be labeled as change.
Questions I Nonetheless Have
- Are there different pitfalls customers of single-descriptor wallets ought to pay attention to within the case the place wallets use extra personal sync instruments like personal Electrum servers and CBF, and utilizing trendy pockets practices that by no means give out repeat addresses?
- Does the “change” keychain come into coin choice in any respect in Core? For instance, are there algos for coin choice the place understanding if a UTXO is a part of the change keychain reasonably than the exterior keychain issues? If that’s the case, how does it come into play?
- Another historic data that is good to know and consider for builders of pockets can be appreciated.
Thanks!
