The ARI response to Proteus

As some may have heard, that yesterday announced to create a language he calls “Proteus”. We at the ARI are not sure why he wants to perform such a NIH stunt and compete with the best Objective-C replacement available (not): Swifter.

The response of the Always Right Institute to this is simple: Go ahead David, go give it a try! Proteus will be all wrong and Swifter will be always right. “Old Man of the Sea”, who would want to program in something ‘old’?

What is wrong with Proteus - a thorough analysis

TL;DR: Everything

As a starter it is not being done by the Always Right Institute. How could it possibly perform at 1337 speed? Or be right?

I’ve created a Patreon if you feel keen to supporting the development.

Got it all wrong. With programming languages it works the other way around. The language developer has to pay the poor people trying to use his language (unless she owns the market and can play with devs as she pleases - like letting C++ compiler writers go wild).

… simple: create a modern-day C language …

Excuse me? Never use simple and C in a single sentence, modern-day or not. This is plain offensive towards real developers who spent years to understand why it’s `NSString *title`, not `NSString* title`. Or why source code wider than 78 bytes tends to crash unexpectedly.

I’ll be writing the parser using Swift …

Of course you are trying to get it right by stealing ideas from the excellent Swifter community. That won’t help, and it is just not right.

… expect to see a lot of interesting tangential posts comparing the two languages.

After an intense discussion, we at the ARI decided to agree on this one. There are going to be a lot of posts comparing Swifter and Swift. And we know who is right.

Eventually, the project will be completely open-sourced.

Well, Swifter got a headstart. All of the components (written) got released as open-source already.

So, I expect we’ll see some interesting Swift-to-LLVM C API challenges along the way.

Anyone calling those challenges ‘interesting’ haz no idea about LLVM.

Finally: Proteus doesn’t even have a logo.


We clearly demonstrated this is all wrong. Swifter is alright. Word!

But - as usual, we went deep on this and had our team of American scientists review Swifter as well as Proteus. They confirmed our findings.

Mr. Owens, you should clearly tag your project as ‘an experiment’. E.g. adding a sentence like ‘This is an experiment’ to your Proteus posts.


We at the ARI invite everyone to ignore look-a-likes and take a sip of right by trying Swifter, not.

As written in “Thousand and One Nights”:

Swifter, after having got the Old Man drunk with wine, was able to shake him off and kill him.

P.S.: Did we mention that other Swifter clone, eero? Same thing. Doomed to #fail. No, wait:

Eero is no longer being maintained by its original author.

Looking for a job?

Be sure to learn Swifter, it is the language of the future!

Written on September 4, 2015