Jump to content

  • Log In with Google      Sign In   
  • Create Account






Barfing out some thoughts

Posted by ApochPiQ, 19 June 2014 · 194 views

It's an interesting time, that's for sure.

Work has been consuming a monumental portion of my time and energy lately; and while I can't talk about it yet, I can say that it has definitively been a lot of fun and I really can't wait to unleash what I've been working on. There's a ton of potential in it and it'll be endlessly entertaining to see how people react to the whole thing.

Outside of work, I've been pursuing a few various hobbies a little more than I have in the past. This loosely includes doing a bit more gaming than I have been recently, which is a nice change. My PS4 still sits unused until some actually good games come out for it, but until such time as I have a meaningful purpose for that console, I'm getting a ton of mileage out of replaying (or finishing) games on the 360 that have been gathering dust.

On top of this, I'm working on buying my second house, which isn't really a giant ostentatious display of ludicrous wealth... more a function of not having had a chance to sell the first one yet. Either way, buying and selling homes is a pretty exhausting affair all around, and a significant number of my already-limited spare brain cycles have been stolen by that entire mess.

All that is really just background. I'm not really interested in bitching about how busy I am or whatever; my point is that there's not enough hours in the day to do all the things I wish I could do.

Speaking of things I wish I could do... Epoch has been suffering because of all this other stuff. I still have a massive list of things I want to do with both the language itself and the surrounding tooling... there simply isn't enough clock to go around.

Yes, the language self-hosts now; yes, the compiler supports almost everything it needs to in this revision of the language; yes, the IDE is slowly shaping up into something actually worth using. But there's so far to go, and my priorities have simply been elsewhere.

Some of my shift of focus is actually due in large part to Apple's recently-announced Swift language. I have very mixed feelings about Swift. A careful observer might notice that there are some conspicuous similarities between Epoch and Swift; this is, I assure you, completely coincidental.

What is not coincidental is that someone noticed those similarities about a year ago - long before even the idea of Swift was public knowledge. The story really isn't all that long or complex or interesting, but the bottom line is that I nearly wound up working for Apple on their compiler and IDE tools. It was a very, very close thing, and in the end I made the exceedingly difficult decision to decline the opportunity, for mostly personal reasons.

The point of all this is that seeing Swift announced - and delivering many of the very concepts that I've talked about building into Epoch's core language and tools over the years - is kind of difficult for me. To be perfectly clear, nobody appropriated anything from me or Epoch without my knowledge and/or consent. The fact that Epoch and Swift have such startlingly parallel goals is why that opportunity happened, not because of it.

On the one hand, Swift gives me a lot of validation. They've observed the same weaknesses and pain points in contemporary languages as I have, and settled on many of the same approaches to alleviating those problems. There's a lot of things in the language and tools that I wouldn't have done, but that's kind of like saying one painter used a different hue and brush stroke to paint his sky than some other dude did... completely meaningless and kind of obvious. Different creators will wind up producing different creations.

The flip side is that Swift means Epoch will forever be seen as the copycat, not the originator of some of the core ideas that I've been pushing for so long. In the end nobody cares who had the idea - they care who popularized the idea. And Apple has without question pushed a lot of my personal grievances directly into the spotlight, and offered up great solutions to them. I don't regret the choice I made, but it is kind of sad that I didn't get to be a part of doing that.

In a way I feel like Epoch no longer has a chance to be judged on its own merits, although I know that's not really true. It's more of a feeling of having let myself down a bit. When I first envisioned Epoch over eight years ago, I genuinely wanted it to be a major game-changer in the programming languages space. Between Go, Rust, and Swift, many of the key points of the first language proposal have been addressed.

To be sure, there's still room for the unique blend of ideas that Epoch has become, and I still feel that Epoch can be a better solution to certain classes of problem. If nothing else, it will certainly continue to be the language I'd like to program in whenever possible.

"Getting awful crowded in my sky."




I'm a little hesitant about how effective I think Go/Rust/Swift will actually be in shaking up the language landscape. On the one hand, they are clearly all addressing key problem points, but Go seems to have siloed itself into the distributed computing space (where it competes against the much better-established Erlang and Scala), Rust is still stuck in the category of interesting-but-not-ready-for-prime-time academic research languages, and if Objective-C is anything to go by, Swift will likely never gain traction outside of Apple's walled garden.

 

All of which goes to say: I still think that a truly pragmatic language for the future has an opportunity to gain real traction. Provided of course that it can differentiate itself from Clojure, which I see as the other logical contender for that title.

I would not give up hope. You have loaded too many hours to simply stop.

August 2014 »

S M T W T F S
     12
3456789
10111213141516
171819202122 23
24252627282930
31      
PARTNERS