Liam Gray

The issue on your site appears to be that ``` javascript $(document).ready(function () { Snarl.addNotification({ title: 'Issue', text: 'Testing, testing...' }); }); ``` Is running before Snarl can initialise itself....

I'll look into it soon. Could it be that mobile browsers don't have the `WebGLDebugUtils.makeDebugContext(context)` function? See [js/raytracer.js#L24](https://github.com/hoxxep/webgl-ray-tracing-demo/blob/master/js/raytracer.js#L24) for where I try to get the context.

Thanks for the swift response! Yeah I would be happy with either, as long as the guide makes it clear that we should be consistent, and possibly present both options...

@abonander apologies for the delay. I have pushed an alternative proposal to align with your comments, which is to allow setting the `in_memory` flag in the connect options, and also...

On second glance my pool timeout issue with `.max_connections(1)` still exists, but I still believe the PR could be worth merging for reasons outlined below. The parsing code applies this...

At first glance this looks like there's an issue with your `npm install` or Node install rather than electrolyte, as you're getting a TypeError inside `util.inherits` -- a native Node...

`IoC.Container().use` uses a hierarchy – the last `.use(...)` is the first one to be checked for an object. So while using a different container for each test is one option,...

That's a neat workaround to add individual objects into electrolyte. Note the `else ... return` statement is not needed though, as the function returns the same value, `undefined`, without it....

Please give more details about the issue. How are you trying to use electrolyte etc. with specific code samples.

A potential workaround could be to set `min_connections` to a larger number, which shouldn't leak connections in the same way as `idle_timeout`. In my limited testing of this workaround, 3...