[opensource-dev] Indicating privacy implications of preference settings (was: STORM-34 Test Binaries)

Carlo Wood carlo at alinoe.com
Tue Dec 28 17:05:43 PST 2010


Ok, good points. I take back what I said :).
The best way forward seems to put it where people might first
look and then make clear that everyone can see them with text
right there.

On Tue, Dec 28, 2010 at 10:58:01PM +0100, Boroondas Gupte wrote:
> On 12/28/2010 09:40 PM, Celierra Darling wrote:
> 
>     It seems a little unclear to try to communicate "this can have privacy
>     implications" by putting the setting on the privacy tab.  It might be
>     better to write the setting label so it's more explicit (i.e. something
>     like "Show my favorites to anyone under 'Start At' before logging in" or
>     "Show my favorites under 'Start At' on login (before authenticating)" or
>     etc.).
> 
> I have to agree that placing settings on the "privacy" tab is not a good way to
> indicate the fact that they have privacy implications. It goes against the
> principle that a setting should be placed where users expect it. I.e., if you
> know (or suspect) a setting exists, but don't know where to find it, where
> would you first look for it? Education about the consequences of changing the
> setting can be postponed until after you found it.
> 
> Of course, besides looking at tab labels, placing similar or related settings
> together can help in finding them. But the common property "affects privacy"
> seems to me too week a reason for placing settings on the same tab, when that
> property is not the goal of a setting but rather a side effect (even although a
> side effect inherent to the goal itself, not just to the setting's
> implementation).
> 
> Further, when a user finds a setting on the "privacy" tab and realizes this
> placement means the setting has privacy implications, how would she or he know
> whether enabling or disabling the setting grants higher privacy? Of course, if
> one can derive from the description of a setting what it actually does and then
> thinks it all through, one gets the answer as well as how one's privacy would
> be affected. While the users are probably well capable to do that, do they want
> to do that—when they could just as well be told openly about the setting's
> effect, e.g. like in Celierra's labeling proposals above?
> 
> Boroondas

> _______________________________________________
> Policies and (un)subscribe information available here:
> http://wiki.secondlife.com/wiki/OpenSource-Dev
> Please read the policies before posting to keep unmoderated posting privileges


-- 
Carlo Wood <carlo at alinoe.com>


More information about the opensource-dev mailing list