Defacto
Member
Hi,
Some time ago there was discussion about registering username through email. Those usernames will point on Dash addresses (they can be used instead hard to remember normal addresses). There was discusion about this topic here: https://www.dash.org/forum/threads/why-does-evolution-require-email-registration.7819/
Why instead that we don't use something simpler? Something what is useful and tested through many years. Something what has well build infrastructure? I mean DNS.
For that purpose we can use TXT records in DNS. Let’s imagine DNS entry for dash.org like this:
so… the Dash wallet after entering “[email protected]” in “Pay to” field can find (resolve) address to which it has to send Dash.
In that way we can easy define “catch-all” addresses (like in mail):
In such way we can define that anything given before @dash.org domain will be directed to specified wallet address. DNS system allows to define multiple TXT type records. So we can imagine something that looks like this:
Such records should define that Dash transfer at first will go to better (more narrow) defined name (in this case eduffield)… and if there is no exact name found this transfer go to catch-all address type.
Of course we can also put everything in a single TXT record and additionally add versioning:
or even in last case define separate address for whole domain without giving recipient.
We can also take another approach and use different notation:
Let's imagine that we can create links like:
<a href=“dashto:[email protected]”> or
<a href=“dash://dash.org”> or
<a href=“dash://[email protected]”>
Some guys can claim that it will be easy determine what wallets resolves what queries (that DNS servers owners can effeciently deanonymize senders by theirs IP). That there will be lose of privacy which is so important in Dash. But what stands in a way that DNS queries will be executed by MN network? We have such advantage over other crypto currencies and we can’t use it.
Those same guys can also negate security of such solution by pointing for example on DNS spoofing. I think that DNS queries should force of DNSSEC usage. So in case of using proposed solution MN network should check if DNS query is signed securely.
WWW, mail and hundreds of other aplications uses DNS almost from begining of internet. It is well known and very well tested solution. If so much services and applications uses DNS so why we can’t use it for make remembering of addresses easy and in that way free people from cut&past of complicated crypto wallets addresses? Ha! Implementation of this will also eliminate some errors when given name is wrong and there is no catch-all record. In such case name isn’t resolved and transfer will be rejected.
This is only very early idea which emerged in only one evening. I'm sure that there are many things to thought about, but what we are for?
Pros:
- large and ready to use infrastructure
- known and well checked / designed system which exists so many years
- data will not be kept in Dash network (so resources of MN will not be consumed)
- easy updating (even without owning Dash client)
- easy usage of this solution in external systems. it easy to create simple scripts on WWW which will update receiving address without using Dash network
- no costs (no user registration paid in Dash). Of course there are costs of domain, but some people already have one domain. In other case anybody can still use many free domains.
- no changes on DNS system side
- another layer of hiding transactions? for example DNS records generated only for a specified time with random name. Something like PrivateSend but opposite way. Sender after specified time will not know where Dash go (of course this is just first idea)
- many more? ;-)
Cons:
- rely on other servers (not Dash related)
- any others?
Some time ago there was discussion about registering username through email. Those usernames will point on Dash addresses (they can be used instead hard to remember normal addresses). There was discusion about this topic here: https://www.dash.org/forum/threads/why-does-evolution-require-email-registration.7819/
Why instead that we don't use something simpler? Something what is useful and tested through many years. Something what has well build infrastructure? I mean DNS.
For that purpose we can use TXT records in DNS. Let’s imagine DNS entry for dash.org like this:
Code:
dash.org. IN TXT "dash:eduffield@Xem4RTrHuTG86ebQCdeEdtWwArLq3EFfMT"
so… the Dash wallet after entering “[email protected]” in “Pay to” field can find (resolve) address to which it has to send Dash.
In that way we can easy define “catch-all” addresses (like in mail):
Code:
dash.org. IN TXT "dash:@Xem4RTrHuTG86ebQCdeEdtWwArLq3EFfMT"
In such way we can define that anything given before @dash.org domain will be directed to specified wallet address. DNS system allows to define multiple TXT type records. So we can imagine something that looks like this:
Code:
dash.org. IN TXT "dash:eduffield@Xem4RTrHuTG86ebQCdeEdtWwArLq3EFfMT"
dash.org. IN TXT "dash:@Xem4RTrHuTG86ebfdsfsDDSDFSArLq3EZZZX"
Such records should define that Dash transfer at first will go to better (more narrow) defined name (in this case eduffield)… and if there is no exact name found this transfer go to catch-all address type.
Of course we can also put everything in a single TXT record and additionally add versioning:
Code:
dash.org. IN TXT "dash/1.0 edufield@Xem4RTrHuTG86ebQCdeEdtWwArLq3EFfMT @Xem4RTrHuTG86ebfdsfsDDSDFSArLq3EZZZX Xem4RTrHuTG86ebfdsfsDDSDFSArLq3EZZZX"
or even in last case define separate address for whole domain without giving recipient.
We can also take another approach and use different notation:
Code:
edufield._dash.dash.org. IN TXT "Xem4RTrHuTG86ebQCdeEdtWwArLq3EFfMT"
_dash.dash.org. IN TXT "Xem4RTrHuTG86ebfdsfsDDSDFSArLq3EZZZX"
Let's imagine that we can create links like:
<a href=“dashto:[email protected]”> or
<a href=“dash://dash.org”> or
<a href=“dash://[email protected]”>
Some guys can claim that it will be easy determine what wallets resolves what queries (that DNS servers owners can effeciently deanonymize senders by theirs IP). That there will be lose of privacy which is so important in Dash. But what stands in a way that DNS queries will be executed by MN network? We have such advantage over other crypto currencies and we can’t use it.
Those same guys can also negate security of such solution by pointing for example on DNS spoofing. I think that DNS queries should force of DNSSEC usage. So in case of using proposed solution MN network should check if DNS query is signed securely.
WWW, mail and hundreds of other aplications uses DNS almost from begining of internet. It is well known and very well tested solution. If so much services and applications uses DNS so why we can’t use it for make remembering of addresses easy and in that way free people from cut&past of complicated crypto wallets addresses? Ha! Implementation of this will also eliminate some errors when given name is wrong and there is no catch-all record. In such case name isn’t resolved and transfer will be rejected.
This is only very early idea which emerged in only one evening. I'm sure that there are many things to thought about, but what we are for?
Pros:
- large and ready to use infrastructure
- known and well checked / designed system which exists so many years
- data will not be kept in Dash network (so resources of MN will not be consumed)
- easy updating (even without owning Dash client)
- easy usage of this solution in external systems. it easy to create simple scripts on WWW which will update receiving address without using Dash network
- no costs (no user registration paid in Dash). Of course there are costs of domain, but some people already have one domain. In other case anybody can still use many free domains.
- no changes on DNS system side
- another layer of hiding transactions? for example DNS records generated only for a specified time with random name. Something like PrivateSend but opposite way. Sender after specified time will not know where Dash go (of course this is just first idea)
- many more? ;-)
Cons:
- rely on other servers (not Dash related)
- any others?