It's funny that they want to do these checks in a country where even the checks for voting are very iffy.
In any case, voting is substantially more intrusive. You must register with your full name and address, which is made public record. Each time you vote, that is also made public record (not who you voted for, but the fact that you voted). In states with closed primaries, your party membership is public record. In states with open primaries, it's public record which party's primary you vote in. It's way more invasive than a text box in your computer's account setup screen that asks for your age.
Anyone who tells you differently is lying or ignorant.
Some people think disenfranchisement is bad. Others see it as useful.
Specifically, PartyB thinks those people with inadequate ID skew toward PartyA voters. This has been the accepted wisdom for decades. So they are incentivized to make it harder for them to vote.
Interestingly though, PartyB might be wrong about the current population. PartyA, and those against disenfranchisement and imaginary crises in general (I count myself in this third group), do not want to blow up centuries of precedent especially if the consequences are likely to be undemocratic and unfair.
Luckily, this problem is wholly solved via selective enforcement.
How is it that you don't need an ID to exercise the rights of voting 'citizens', but you need one to exercise the right of 'people'? Consider that virtually all 'citizens' are also 'people', and even if you argue they are not, the portion of voting citizens that aren't 'people' is inconsequential compared to the supposed "10%" that can't muster an ID.
It's almost as if both sides of the argument are just using logically inconsistent arguments that just aligns with whatever gets the voting demographics they like. In fact, Vermont is the only state I know of that gives both full rights of citizens and full rights of people to those without ID in a manner consistent with the anti-ID argument usually presented.
Consequences of errors with guns are higher than with voting, because elections are audited and mistakes and fraud are found and reversed.
You cannot helpfully audit misuse of guns, after the fact.
But lets accept your premise as true.
You're proposing something like rank-stacking the risk of various rights of citizens and people and if they're high enough on the stack it's OK to to ID and if they're lower maybe it's not OK. That seems to move the goalpost quite a bit from your prior argument.
This happens before the winners are certified, and before they're given the ability to drop bombs.
I don't understand your confusion.
In the US, ACH transactions are reversible and trusted throughout the nation. Bitcoin transactions are not, and are not. This seems parallel to me.
Giving you the benefit of the doubt regarding the intent, why would anyone support a measure that demonstrably does not achieve what it intends, but instead denies you the right to vote?
This is how we know how extremely few problems there are, and how we catch the accidents (which are backed out, hence the delay between voting and election certification), and the fraud (which is extremely rare but of course also backed out).