[jtheory]

Cool, thanks for the added detail!

I suspect the main additions you'll need for this to be useful in a production environment will be around logging and accessing stats -- showing the end user that a particular page performed poorly doesn't help you unless you also know about it, with enough info to reproduce the problem. Also, with higher volumes, you don't want to be flooded with almost-identical data, so you can dig into strategies like only sampling a subset of actual requests, different handling for poorly-performing queries, etc..

I'll keep an eye on this, and if we do decide to integrate it I'll be in touch; thanks!