Update MailtoHandler

[mal-tee]

I propose a few changes to modernize our mailhandler. The idea is that the mailto header should be the obvious choice. Since websites nowadays can also handle mailto: links, the benefit of the specific handlers is dwindeling imo.

• rephrase the mailto string from software to something more generic, e.g. "Default mailer", "App", "Standard E-Mail Dienst"?
• move mailto to the bottom of the list, i.e., closest to the intial button
• make mailto primary

This could look something like this:

What do you think?

[baltpeter]

My two cents:

I agree with making the mailto link a primary button. That seems like an obviously good idea.

As for the rest, I'm personally really not sure. Mostly because I just really don't have a feeling for how people use and understand email.

• Yes, webmailers can register protocol handlers, but does that actually happen in practice and do users generally enable that?
  • Could we maybe recognize if the mailto link didn't work and offer the user more explanation and alternative options in that case?
• I've also considered dropping the MailtoHandler on mobile and just showing the mailto link assuming that users will typically have some sort of email app installed, right? Do people use webmailers on mobile? And even if they do, do the links we provide work there (at least for Gmail, they don't: Mobile version of Gmail doesn't support the link scheme in MailtoDropdown #1191 (comment)). No clue whether that's a good idea, though.
• I have absolutely no intuition for what label would be best understood by people.
• Do you think that having the mailto button last is most obvious? I know that we have a bit of prior art of putting the most-used choice at the bottom, but not to highlight it but because people have to consider the other options before choosing the last one.

This really seems like the textbook case for needing user testing, but I'm not sure how we could do that. :|