The evidence is unclear to me that it increases turnout very much. That's partly because of the way the evidence gets presented.
Maybe my own opinion is not worth something, but for what it's worth, I quite like strolling down to my polling place and standing in line with other Canadians, and I think there is something to be said for that. That said, it's a pretty small good, so if we can increase turnout by securely increasing online voting, then maybe we should pursue it.
The only qualification is that you can't easily get back people's sense that there is integrity in the electoral system. We've a pretty foolproof system right now for counting votes, for making sure that it's secure. You can imagine an election in which things go a bit pear-shaped with online voting, and in that case the invitation to have distrust in the system and the invitation for political actors to appeal to that may not be worth the risk. On balance, it's not clear to me that it's a solution that doesn't invite more potential problems.