user consent, experience with using the new HTMLLocal storage option ...

[Cantor, Scott]

On 2/1/17, 2:54 PM, "users on behalf of Steven Carmody" <users-bounces at shibboleth.net on behalf of steven_carmody at brown.edu> wrote:

> we're interested in hearing abut any experiences sites have had testing, 
> exploring, working with this new option for storing a user's decisions.

Not using it for consent, but I've been using local storage since the initial rollout of V3 at OSU last Spring to maintain the option to deploy logout. If we don't ever do it, I'll yank it because it would just be pointless overhead, but while it's slightly annoying when you know what it's doing and why due to the extra hops, it's not causing any problems or degradation.
    
> We're interested in all levels of experience. Very curious, tho, as to 
> whether any site has this approach in production use...

I would not use it for "show the attributes" consent, for all the reasons it's obviously not a good fit. I think it's very usable for a kind of "authorize release of data to R&S services on this device" poor man's consent that is easy to understand and repeat occasionally.

As a technical matter, local storage has been highly successful and is one of the more insane things I've ever implemented that actually worked.

I would implement it in the SP in a minute if it weren't for the problems with having that run in the middle of an application sequence, but I suspect I will implement it there for use in more controlled application interactions that can compensate for those issues.
	
-- Scott