Open amv opened 10 years ago
Reading the existing documentation I had some issues with it. I found it rather opaque– that is, hard to get it into my head. It start swith an architecture overview which I really don't care about when I'm first reading the docs. Worse, that architecture introduces new jargon.
Think about how unix man pages are usually laid out– I think there's a lot to be said for their organization.
Having it start with a focus on how to use it would be particularly helpful.
For instance, I'd like to see https://github.com/meetings/gearsloth/blob/master/examples/json-client.js embeded in the documentation, along, perhaps, with a note about what it does, eg "Execute the 'log' job with a payload of 'kitteh' in one second"
I'm a bit confused by https://github.com/meetings/gearsloth/blob/master/examples/json-worker.js though… isn't submitJobDelayed provided by gearsloth? Shouldn't the worker be for the 'log' function?
The audience break out suggested would go a long way to improving the documentation I think. I would highly recommend putting those into separate files though. Having them together means people don't have an easy and obvious way of knowing when they're done reading.
I propose that the documentation will be restructured to cater to 4 different types of readers:
For "end user" the important things would include:
For the "sysadmin" the important things would include:
For the "advanced end user" the important thing would include:
For the "contributor" we should have at least: