[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Social-discuss] Comments: GNU social relationship manager

From: Bjarni Rúnar Einarsson
Subject: Re: [Social-discuss] Comments: GNU social relationship manager
Date: Thu, 24 Jun 2010 09:38:11 +0000

On Thu, Jun 24, 2010 at 2:09 AM, Jason Self <address@hidden> wrote:
Rich Hilliard <address@hidden> wrote ..

> So, what is the right default

Share nothing, IMO.

Sure.  That should probably be the default setting, but the system should also by default make it easy and intuitive to change. The whole point of social networking is sharing.

The way I would prefer is to, by default, create a bunch of common groups (tags, lists, ...) which users can easily add people to. Groups like 'Family', 'Friends', 'Coworkers' and of course 'The World'. Then when granting (or revoking) access to things, it should be easy to reference these groups.

It would be incredibly user un-friendly if the user had to define all their sharing groups from a blank slate, pre-creating some makes the feature instantly visible and understandable.  It should obviously be possible to edit, add and delete groups, but if you start with nothing pre-defined then people won't have an easy place to start.

Share nothing by default, means your groups start empty (except for 'The World') and nothing is shared with any groups, but when you post something that would be one of the first questions: which group(s) should have access to this new thing? And when you add a relationship, you will be asked which groups that person should be a member of.

I also really think people should be encouraged by the UI to set access for entire groups, not individuals. Pushing people towards the group abstraction will make it much easier for them to manage permissions later on, which is actually a critical part of protecting their privacy. If you have to configure every person individually, then making changes later becomes a huge amount of work - which means protecting your privacy becomes a huge amount of work, which is exactly what we want to avoid...

Sorry if this was redundant. :-)

Bjarni R. Einarsson

reply via email to

[Prev in Thread] Current Thread [Next in Thread]