You claim that snap offers real benefits but list none and dismiss detractors as irrational and suggest that instead of complaining about Snap users who have no desire to use snap should invest their own efforts to improve something they have no desire to use instead of being critical.

You are correct that this appears to be EXACTLY like the systemd debate.

Snap HOPES to provide an easier environment for developers to target and thus provide a richer ecosystem for users to enjoy. This like trickle down economics probably isn't real. Like literally every other time Canonical decided to go their own way they will provide an inferior option that isn't taken up outside their own ecosystem before eventually giving up and joining the crowd. Unless it attracts highly hypothetical new developers to linux it offers nothing but downsides to users.

- It's tied to a close source server run by and solely controlled by Canonical with no ability to add software channels like virtually every other major software distribution model for Linux. This means not only could Canonical exercise undue control over how their users use software on their platform it means others including repressive governments could force it to on their behalf.

- Users may only install the most recent version of software and will be updated to the most recent as soon as it comes out. -- This means that if devs push a buggy version you are stuck with it until its fixed. If it isn't fixed for months you just can't use the software. Bugs that effect everyone will probably get fixed immediately. Bugs that effect niche features or a smaller number of users are liable to go unfixed for longer. Please see bugs that are open for years at a time.

-- In case of developer getting compromised ability to push updates to all users as soon as users machine is online means that a substantial portion of user base can be hit within minutes and almost all within hours. If a new version had to get pulled in and then distributed at irregular intervals a new version would take at least weeks to compromise most users. This would give users/packagers/distros/developers time to realize what is going on before all users are effected.

- For some reason they are slow to start

- Waste users bandwidth and storage even with one or the other is dear.

- Results in 17 different apps having 17 different version of a dep 16 of which have known security vulnerabilities because apps don't use system libraries that get updates.