Is Jetbrains going to remain relevant in the next 10 years?

I want to start this by saying how much I like Jetbrains. Installing Resharper for the first time and discovering what it was capable of was one of the most exciting moments in my .NET career. I felt like I was getting super-powered overnight. I liked it so much that I begged my boss to buy it for everyone in the shop. He didn’t say no, but his delay made me by it myself. I couldn’t stand living without it. Later on he ended up buying 43 licenses, and after that, for whatever bizarre reason, my e-mail ended up being the “main contact” for Thomson Reuters and one day I received an e-mail asking me if I wanted to renew the 2000+ licenses registered “on my name”. Anyway. I kind of feel like they like me to :).

After my experience with Resharper, my mindset was that whatever they produce should be of equivalent excellence so, when I first started learning Node.js and React back in early 2015, I didn’t hesitate and bought myself a WebStorm license. It seemed to be fairly up to date with whatever nerdy-fashionable-gimmicky library or framework was trending at the time, and React was included, so I was happy. I compared it with Atom, which was one of the major choices at the time, but the seamless Git integration, the refactorings, the smart code completion and error highlightings kept me tied to WebStorm. Atom seemed too slow to add so little to the table (at least at the time).

Then, a few months later, mid 2015, Visual Studio Code came along. I remember my first thought was:

What the he*k is Microsoft thinking? I mean… GitHub creates Electron, then Atom on top of it (Not necessarily in that order). The developer community seems to be more engaged to it than it would ever be to any Microsoft product. Because… You know, Microsoft. And then Microsoft comes in and expects to use GitHub’s own weapon to create another Atom? No way this will ever work.

I didn’t even bother to check whether I would switch from WebStorm. I wouldn’t.

Then, later 2015, Jetbrains announced the most controversial, rant-provoking license change that I have ever witnessed in the dev community. They would switch to a subscription based model. Reddit went on fire, lots of posts with repeating complaints and heated comments. Jetbrains even seemed to be reconsidering but they came back with a final decision that didn’t change that much. I personally didn’t think it was all that bad at the time. I mean, I would end up paying for it anyways. Maybe it would even help the company to create even greater products.

Fast forward to mid 2016, Microsoft releases VS Code 1.0 and for the first time I started reading positive reviews about it all over Reddit, Hacker News and Twitter. People seemed to like what they were seeing. One of the top praise points was how fast new releases were coming out, and how the dev team seemed to be hearing the community and making their UserVoice the actual road-map.

At that point I decided to install it and compare it with WebStorm. I thought: “Not quite there yet, but promising.”. I remember complaining about the absence of tabs and how fast it took for the tabs to actually be included. I felt like I couldn’t quite replace Webstorm yet but I started to use VS Code casually. I liked the Git integration, the simplicity in the UI, how fast it was, and the fact that everything seemed to work kind of naturally. I mean, it wasn’t as smart as WebStorm but the syntax highlights and code inference didn’t break with modern JavaScript like React. TypeScript worked as a charm (no wonder) and CSS, Less and Sass seemed OK. Ah, and the fact that they released new versions like crazy (VS Code and Node.js should have won the 2016 release frenzy awards). Ah², and the fact that for the first time I got to debug Node.js just by hitting a button without having to pass through a satanic ritual or selling my soul the devil. Ah³, and the fact that I installed the Python Extension and it went so seamless, with PyLint, debugging and code completion support, that I see reasons why people would favor VSC over PyCharm, even though PyCharm has a free version and is an awesome tool.

As time went on, I started to pay more attention to the problems WebStorm always had and I previously overlooked. Probably because I now had an alternative and because the new subscription model was weighting, even if not that much. It was slow. Some bugs were never fixed (like the moving file refactoring bug that would screw up all the ‘imports’) and the refactorings seemed less important for me now for some reason (maybe I got more experienced?). The point is, I started to think whether I would renew the subscription or jump to VSC altogether. I didn’t renew it, even though I still find WebStorm an awesome tool.

Now let’s analyze the VCS vs. Atom situation.

I don’t know why the h*ck GitHub thought it was a good idea to invest in a text-editor but I’m glad they did (I still think they should have created Electron to be the basis of the best possible cross-platform Git GUI ever, not a text-editor, but this is material for another article). Anyway, I see a lot of value in Atom nevertheless.

Microsoft, in the other hand, has a lot more motive. Nadela’s objective is clearly the cloud and that is going quite well. They want to create the best possible cross-platform editor and make it popular so they’ll be able, in the future, to use it as a leverage to push devs to Azure, no matter what technology they’re using. VSC doesn’t seem like a leverage to today because they need more traction first, but it will when the time is right. By the way, I was wrong about my first reaction about VSC, both about the editor and about Microsoft. It seems that Nadela has changed the company so much.

Here are the top 2 comments on a Reddit thread comparing Atom and VSC:

I used Atom exclusively on mac for development (php, vue and some angular) for several months and loved it.

I switched to VSC a few weeks ago because I loved VS when worked on a .NET stack. Honestly, it’s about the same in terms of functionality, I like that it’s a little more streamlined and I think it handles syntax highlighting better than Atom (in terms of laravel blade).

Really, just download and try it out for a bit, download the extensions you need and go to town. There is very little difference between Brackets, Atom and VS Code.

Source.

Atom and VS Code are very similar in concept, but their execution is wildly different. (Although less so now that Code has added tabs to the editor window.) The biggest difference between the two is that Code’s performance is orders of magnitude better than Atom’s. For example, I used to keep a second editor (usually Sublime) ready for opening large files, because Atom just couldn’t… Fucking… Do it. I no longer have Sublime installed because I didn’t see a reason to put it on this machine after switching to VS Code for most of my editing.

Source.

Now back to Jetbrains.

I expect most of Jetbrains revenue to come from IntelliJ based IDEs and Resharper. Oh wait, they have a 2015 annual report (2016 not available yet). I was almost right.

If you take the time to look at the report you’ll see that 2015 was an awesome year for Jetbrains. But, there’s always a but, I’m not sure if the next 10 years will look as bright. Because…

1. Atom and VSC are getting strong, I personally think VC Code will win because Microsoft has a brazillian of money and they “need” it to support Azure in their fight against AWS. Both Atom and VS Code are starting as text-editors on steroids but I believe they will eventually evolve, specially VS Code, to the IntelliJ level in the long run. And it will all be facilitated by how easy it is to code extensions for these editors, especially because you can do it in JavaScript and everybody loves it. As it happened with Python already, I think these editors will become smart enough to minimize the need for IntelliJ IDEs in lots of languages.
2. .NET Core, in terms of coding, seems to be the priority for Microsoft now. I don’t see the full-fledged Visual Studio losing any steam in the short term, but I think this will happen in the long run, in favor of VSC. Besides, with every new release of Visual Studio, there’s a bit less need for Resharper. Microsoft might catch up. They might was well end up buying Jetbrains or the Resharper division as they did with Xamarin (not that I’m rooting for it, competition is healthy).

I want to wrap this up by stating that I wish the best for Jetbrains but I think they’ll need to reshape to stay relevant in the next 10 years, if they’re not bought. I don’t see any reason they why wouldn’t (reshape). They’re full of enthusiasm and ridiculously talented minds. They’ll find a way.

Thanks Microsoft, Jetbrains and GitHub for making my coding career so enjoyable.

Learn more about me here: andrerpena.me