Continuing on with the coding for v0.3 of the Eddy Framework, another enhancement has been added in time for the drop scheduled for this weekend. This feature solve one of the major issues with Web 2.0 programming - what happens when a remote Data Source goes down?

We all have seen the issue before, you integrate a new API, or consume an RSS feed on your site - only to find out your site is rendered useless for a couple hours because the remote server supplying the data has gone down. In an effort to fix this, if caching is used for any Data Source (as it should be whenever possible for performance reasons), the cache will not be dumped until a response has been received from the remote server. In simple terms, you will be safe guarded against a remote supplier of data which is suffering some downtime.

It is also important to know that this new feature can also be applied to other avenues, such as allowing you to upload new data files, or switch out databases behind the scenes. Since during the switch over process these Data Sources will not be available, the old versions in the cache will be used until they come back online.

Special thanks to John Lovell from CodeValve.com for the idea behind this new enhancement.