This is exactly what I was thinking. It seems crazy to roll this out instead of parse until there's a very clear reason why - and it's actually going to be more fun to use parse anyway since you get away from this kind of stuff.

On top of that, building this out is unlikely to match much of parse's capabilities unless a huge amount of work is put in (by you). And once it's 'done', it will no longer have that magical property of getting better without more work on your part, whereas Parse will continue to grow and kick (even more) ass.

There's almost certainly some part of this approach that would be fun, but I can't really think of it off the top of my head.

with apologies to patio11 for (mis)appropriating his meme.

What feature can Parse build for you so that you don't need to hook up with Heroku just to manipulate data? Perhaps a way for you to author data manipulation macros in the data browser and the ability to run these macros based on events?

That would be a good thing, yeah. Maybe let me write these macros with javascript (which they could interpret on the server side) or any popular scripting language that fits the bill.

Cool. I work there and I have been thinking about server side scripting for a while now. Would love to talk to you more about it; shoot me at my first name at Parse.com

-Andrew

In particular, I'd like to find out from you what kind of manipulation you have been needing to do. This would help us define the scope of this feature.