Mobile development's future is a big question mark

In the past 5 years, mobile devices have taken the world by storm, creating new opportunities for both consumers and vendors: truth is, the landscape of mobile development still looks very immature and we should be prepared to see a radical change in the next 2 to 5 years.

A suboptimal development industry

As we all agree, duplication is something bad in software engineering: it leads to unmaintainable, error-prone code that requires a signicant effort to be fixed.

If you are a mobile factory or a product-company with its own apps-development team, you might be familiar with the annoying problem of fixing, or implementing, the same thing twice.

Nowadays, unfortunately, duplicate efforts are the norm while developing mobile apps: you have to reproduce the same behaviour across different platforms (ios and android, but add blackberry or windows phone if you’re not that lucky) which led people to create tools like cordova.

Problem is, these tools ain’t the solution: they don’t work as nicely as native apps, they probably hide too much from the developer and vendors aren’t keen on supporting them that much. If you, like me, gave stuff like ionic a try, you probably first found yourself surprised by how nice your demo app would look and behave, to then just find out that the more you were implementing, the quirkier it looked.

On another note, the decision on how to structure the team isn’t trivial as well: shall I get devs who can hack on both platforms? Let’s create 2 separate teams? Shall I get 2 lead developers or 1 guy that has a good sense on how both platforms work?

We need things to be better: just like the web, mobile is now a scary land of vendors-dictated “standards”. We don’t need another JavaScript.

The need for standards

I don’t think mobile is much different to what JavaScript used to look like until 3/5 years ago: a lot of vendor-specific standards, weird tools that try to uniform these platforms and a plethora of things that just don’t feel “right”.

We need uniform APIs, we need to be able to rely on the same toolchain (for example, official staged rollouts on the Apple Store) and we need the same kind of transition the JS ecosystem has seen: vendors need to come together and design all of these things altogether, agree on a minimal interface to share and get ready to kill the mess.

Of course, there are a bit more complications here, as we are talking about very different platforms, with different development tools and workflow, which is why I think we will see huge changes, from this point of view, in the next 3 years.

How will it look like?

I have been telling people for a while that I do not believe this multi-platform, fragmented ecosystem can last for long, as I think it will eventually lead vendors to agree on a common platform to work on.

I don’t necessarily think it needs to be Objective-C, but it definitely won’t be Android on Java. At the same time, I don’t know how interested Apple would be in making Swift / Objective-C a thing for the masses (the masses are interested though).

At the end of the day, thinking of one, unified mobile platform seems kind of crazy today but, in 5 years, devs would look back and wonder “how could we do it like that, to be wasting efforts on so many different platforms?”.

My personal feeling?

At the end of the day, all we need is to give a popular language a simple API to be able to completely access your device.

That language’s name is probably JavaScript1. And that API is probably something like JavaScriptCore.

Since React Native seems to be already halfway through the journey of making all of this real, my feeling is that it won’t take long until vendors will realize that the easiest thing to do is to give JavaScript a (real) shot.


  1. I know that it sounds a bit counter-intuitive since I earlier said that “we don’t need another JavaScript”, but there I was referring to what JS used to look like 3/5 years back (= a mess). We’ve come a long way, though there’s still a lot we can do :)

In the mood for some more reading?

...or check the archives.