On May 30, 2012, at 10:21 AM, Kris Borchers wrote:
On May 30, 2012, at 8:36 AM, Bruno Oliveira wrote:
The rest of the security implementation on DS will be ready in a couple of months, so we must wait. For now my next move I was wondering about to start R&D into client libs, specifically start to explore backbonejs persistence strategy and crypto-js.
I would say keep the R&D more general. I think it would be more useful to research general persistence strategies and once we have started creating our different demos, see how easy/hard it is to work the best strategy into those demos.
If we narrow it to general persistence & security strategies I'm fine with that. What I want to avoid is keeping it so general it has no focus or end goal.
Also, in terms of persistence, I had an idea. If you look at AmplifyJS, they have extended the jQuery.ajax functionality to allow for caching and pulling from the cache when it's valid. What might be a good starting point would be to find a good cross-browser technique for detecting network connection (not easy from what I have seen) and further extend AmplifyJS to only invalidate the cache if it has expired and a network connection is present. If there is no connection, continue to pull from the cache until a connection exists. Something like that.
What you are talking about is along the lines of what I envisioned. Start with persistence access, and security, then start adding features like cache checking, local storage, etc...
aerogear-dev mailing list
aerogear-dev redhat com