I know that the 58,000 votes are over a couple of decades, but as I read it it was 58,000 different voters, not 58,000 votes in total, which is still significant.
I'd be interested to see more information on this, maybe the number of non-voters are less than the report, but if thousands of votes are invalid, then that is a problem in my mind, especially since there have been some pretty close votes in some races, such as Al Franken's initial victory.
My concern about this is that Texas is relatively strict in voter laws, and many other states are much more lax. We should do all we can to ensure that only valid votes are counted.
I agree that no one should vote who is not legally eligible. That said, this claim of 58,000 illegal votes will not be proven true, because it's not true. It's a smaller version of trump's claim after the 2016 election that "millions and millions," of illegal votes were cast.
Credible studies from Loyola Marymount University and Arizona State showed that the number of fraudulent votes cast In US elections are miniscule. Beyond minuscule, actually. The ASU study found 10 fraudulent votes in reviewing a decade of votes. The Loyola study found 31 cases of voter impersonation out of over a billion votes. That's .00000003 of the number of votes cast. Far, far less than bman's suggested .003 of the vote (which I'd argue would be unacceptably high.)
These types of false claims and unconscionable exaggeration serve only one purpose. An atmosphere of fear among minority voters and suppression of the legitimate vote. If what you're selling is worth buying, you'd want to make sure everyone who can vote, does.