hurricaneSlider

The name Open Invention Network is a somewhat misleading name as the goal of the initiative is much narrower than the name implies (just protecting the Linux project.) Is there a patent non aggression project which aims to freely license any application of a given patent?

I'm currently working on a prototype for a consumer facing application which hopefully will have global applicativity.

I've often heard the adage that it is never too soon to worry about localisation and timezones, but am wondering whether when trying to develop a MVP whether this holds.

I know there are a number of founders here, so would like ask: "When did you start worrying about localisation? Would you have approached localisation differently if you were to start over again?"

For a recent project we developed our application atop Akka.Net (a C# implementation of an actor system) We had a great experience with it despite the architectural learning curve.

Serverless or function apps seem to be quite popular at the moment. But in my mind they appear to be simply a specialised case of the actor model. I've not used serverless in production, but at a glance, actor systems seem to win in terms of flexibility, theoretical soundness and maturity of tooling. In contrast I've heard horror stories about the kinds of hacks developers who've embraced function apps have made in order to squeeze out decent performance.

A compelling argument in favour of function apps appears to be: cost, hosted services providing an easy means to scale out and relatedly, platform buy in from AWS, Azure, et al. But equally I could easily imagine a PaaS version of an actor system which could share the same attributes.

Is it the language that actor systems are couched in? What makes "serverless" more compelling?