Group one are the Trumps and Putins. The guys who want to control everything. They always want to centralize everything with themselves standing at the top. Sadly a lot of people with power/money are like that.
Group two are the dont-care folks. The ones that sign up to every next facebook because they don't care if their data is misused. The ones that make jokes about Zuckerberg today, but still use FB. They won't put up with the slightest bit of uncoolness or usability discomfort no matter what harm it does to their freedoms.
Sadly both these groups work very well together. Basically I gave up hope that anything reasonable can be achieved as long as both these exist. There's only the option to join either group and live with what these groups bring with them.
(On Android, Hello gets access to your contacts, your rough location, the ability to read and modify your storage, use your camera, and view your Wi-Fi connections)
I understand why it wants contacts, to make it easy to add people, but again, it can be done in a "let the OS offer the contact picker" way, if we're making a privacy-respecting app.
If Google is expected to be the steward of protecting users, they should not allow apps on the Play Store which don't use the most privacy-respecting method possible to implement the needed functionality.
In fact when I use reddit from my phone (through Firefox) I force it to use the non-mobile version as I find it much faster (as long as you open every link in a new tab).
I understand the need for such verification when a webmaster wants to prevent easy re-registrations of trolls and/or banned members, and it's still better than Google's invite only system for when they release something big like gmail or g+, but it's a dark pattern all the way.
[1] I stated "used" because there are plenty of apps (and underlying networks) related to decentralized social platforms that are already developed and in existence...more people simply have to make use of them.
While true, they probably won't be the first answer. You need a some of these networks with real traction (or at least one huge one) so they can derive a protocol from the greatest common factors between them. To preemtively make a protocol sans popular implementation has little value and often ends in low adoption. As we've learned, the success of a protocol is more about its popularity than its presence or quality. So I say let these networks gestate and once the market (of people, not money) starts picking winners, then begin your abstraction.
And for those (of us) working on solutions to this problem space now, keep going. You don't need a committee or standards doc or whatever. You just need an awesome impl.
It's fairly trivial to build a webapp around APIs that they've probably already developed for android/ios apps, so it's hard to forgive them for not having it ready by now.
None of that matters for me though, because when I attempted to sign up from my laptop I got the message 'Sorry, we're mobile only'. I see no compelling reason to be mobile-only at full launch. Mobile-first for your MVP with early adopters, yes that makes a great deal of sense. But when a company does a full launch I think they should support as many of the major platforms as they can (Windows, Mac, Linux, Android, iOS).
Also, I am not convinced that we need yet another social media company (YASMC, "Yazz-em-see"?) that is centralized and will by necessity make money by monetizing attention.
Cal Newport thinks social protocols are the answer, as he discusses in http://calnewport.com/blog/2018/03/20/on-social-media-and-it.... I agree. I worked on this as the OASIS XDI technical committee co-chair for a while, and am working on something new now.