Stirring the pot and not necessarily what I think we should do…
safe:
This is nice in that anyone seeing a safe: URL is given a strong hint that it is not a web URL.
For example “safe:metaquestions” versus “metaquestions.com” or “http://metaquestions.com” etc.
Or “safe:blog.metaquestions” versus “blog.metaquestions.com” or “http://blog.metaquestions.com” and so on.
Lots of us (me me me anyway 
) like the cleanness and bold “this is safe” message of this, though as @Viv points out in the topic which @Pierce links above, it doesn’t necessarily improve security in the way we might assume (because people can still build bridges from legacy browsers and what not).
In the past I’ve been advocating things based on my hope that by supporting legacy browsers we would piggy back on all sorts of things that would help adoption of SAFEnetwork. I am no longer seeing this possibility, and so reconsidering. In fact, I wonder if we might be better off in this respect by ignoring them and just trying to provide the cleanest user experience with our own browser. If others want to build crappy links from old browsers that’s fine too, but when people search out SAFEnetwork, if we have the best or rather easiest solution, that’s what most people will use: they’ll follow our guide, download our client. Later, rival clients will follow (just as happened with Netscape, Internet Explorer and Chrome etc), but they’ll have to improve on the standard we set at this point in time in order to succeed.
So, and not wishing to make @cretz choke on his breakfast, this leads me to consider merging the browser and launcher. For two reasons:
- security is not only about making bad things really really hard. It is also improved by making the right thing the easiest thing for users to do. And if to get on SAFEnetwork you just download one client, and that is the one and only place you log in, and is both able to browse the SAFEnetwork and authorise third party SAFE Apps, why would you use a legacy browser which also needs a launcher and/or plugin?
-
this approach gives a vastly improved user experience IMO - and a much easier to understand and follow route to adoption than: download a “launcher” (a what? what does it do? what does it “launch”?..confusing IMO) and then set up one of several options (a plugin for my browser, oh wait I’m using chrome so I need a proxy, what’s a proxy, what’s a plugin…granny will never get on SAFEnetwork I’m afraid
) just to begin browsing the network. Eughhhh! Horrible.
A big question for us is how these choices affect user experience (both to get started and to use SAFEnetwork) because ease of adoption is so important. We just should not ignore this is my concern.
But TBH I’m wondering if this single step download the SAFEnetwork Browser (with built in launcher) might be the smoothest and easiest route - and therefore maximise adoption anyway.