I'm designing a web application that is locally focused on JSON objects to do all the data stuff (and after that I would pass it to a server to cloud synchronize). However, somebody in an answer from another question suggested me to use Node.js to up/download data directly formatted as JSON.
I read some stuff at the developer's website, and they claim it's as a web server. However, I'm not sure if it's just a data server (which makes sense to me) or it could be a replacement to Apache (I would go for no).
I'm aware that your answers depend on the needs of my application, so, to sum up: it should be able to dispatch data from a database ([No]SQL, whichever) and synchronize it back when I choose (either in the background or by direct action). The client-side should store the data as a mid-step (before syncing and after downloading the stuff) in JSON. Eventually in the future, it should need to synchronize across a bunch of accounts at the same time (you know, team modifications, just like Google Docs and so).
And now, back to the question:
Which are the pros and cons of Node.js?
I'm pretty sure that the direct use of JSON objects from the server to the client and the pushing calls are pretty good reasons to choose Node.js. However, I'm not sure if security (either my code or the data itself) or concurrent connections are mature enough, or I should go for a typical Apache with PHP files querying a MySQL, and in the early future poll instead of push (or find a pushing server).
I know the question may sound easy to answer if you're already on one side (just node, node+Apache, just Apache), but these days I'm from nowhere. Thanks for your answers, mates.