Go (The Programming Language)

go-logo-black

There’s a companion post to this piece in which I pretty much lay out my experience of programming languages. If you’d prefer to avoid reading through all that crap, I can summarize by repurposing one of my favorite lines from Barry Humphries (“I’m no orator, far from it, I’m an Australian politician.”) — I’m no systems programmer, far from it I’m a web developer. Go is explicitly a systems programming language — a tool for implementing web servers, not web sites. That said, like most programmers who live mainly in the world of high level / low performance languages, I would love to program in a lower level / higher performance language, if I didn’t feel it would get in the way. At first blush, Go looks like it might be the “Holy Grail” of computer languages — easy on the eye and wrist, readable, safe, and fast. Fast to compile, and fast at runtime.

What’s my taste in languages? Something like JavaScript that compiles to native binary and has easy access to system APIs. Unity’s JavaScript is pretty damn close — except for the system APIs part (it’s also missing many of my favorite bits of JavaScript, and not just the stuff that’s hard to write compilers for). Actually if I really had my druthers it would be JavaScript with strict declarations (var foo), optional strong typing (var foo: int), case-insensitive symbols, pascal’s := assignment, = a binary comparison operator, syntactic sugar for classic single inheritance, and ditch the freaking semicolon.

Of course, I am not — as I have already said — a systems programmer, so things like the value of concurrency haven’t really been driven home to me, and this is — apparently — one of Go’s key features.

Highlights

  • Clean syntax — it looks easy to learn, not too bad to read, and certainly easy to type. They seem to have managed to climb about as far out of the C swamp as is possible without alienating too many people (but I’m a lot less allergic to non-C syntaxes than most, so maybe not).
  • Explicit and implicit enforcement of Good Programming Style (via Gofmt in the former case, and clever syntactic tricks such as Uppercase identifiers are public, lowercase are private in the latter).
  • Compiler is smart enough to make some obvious ways of doing things that turn out to be serious bugs in C into correct idioms.
  • Safety is a design goal. (E.g. where in C you might pass a pointer to a buffer, in Go you pass an array “slice” of specified size. Loops use safely generated iterators, as in shiny but slow languages like Ruby.)
  • Very expressive (code is clear and concise, not much typing, but not too obscure — not as self-documenting as Obj-C though).
  • Orthogonality is a design goal.
  • Seems like a fairly small number of very powerful and general new “first class” concepts (e.g. the way interfaces work generalizes out both inheritance and (Java-style) interfaces, channels as a first class citizen are beautiful).
  • It really looks like fun — not hard work. Uncomplicated declarations mean less staring at a bunch of ampersands and asterisks to figure out WTF you’re looking at. Fast (and, I believe, simple) builds. It makes me want to start working on low level crap, like implementing decent regex… Makes me wonder how well Acumen would work if it were implemented as a custom web server, how much effort this would be, and how hard it would be to deploy…
  • What look like very fast compiles (and in his demo video (above) Pike recompiled more code in less time on a Macbook Air) — but then I’ve seen insanely fast C++ compiler demos that didn’t pan out to much in the Real World.
  • Really nice take on OO, especially interfaces (very elegant and powerful).
  • Almost all the things you want in a dynamic language — closures, reflection — in a compiled language.
  • Really, really nice support for concurrency and inter-process communication. Simple examples — that I could understand immediately — showed how you could break up tasks into huge numbers of goroutines, perform them out of order, and then obtain the results in the order you need to, without needing to perform mental gymnastics. (Lisp, etc., coders will of course point out that functional programming gives you these benefits as well in a simpler, cleaner way.) The 100k goroutine demo was pretty damn neat.

Lowlights

  • To quote a friend, “The syntax is crawling out of the C swamp”. So expect the usual nonsense — impenetrable code that people are proud of.
  • Concurrency is way better supported in the 6g, 8g, 5g version than the GCC (the first muxes goroutines into threads, the latter uses one thread per goroutine).
  • They promise radical GC, but right now it’s a promise (and no GC in the gcc version yet).
  • Pointers are still around (although, to be fair, largely avoidable, and reasonably clearly declared).
  • No debugger (yet), obviously very immature in many ways (weak regex, and see the point about GC, above).
  • The syntax seems a bit uncomfortably “open” (nothing but spaces where you are accustomed to seeing a colon, or something), but it’s a matter of familiarity I suppose.

Unknowns (at least to me)

  • How well does it compare to evolutionary improvements, such as Obj-C 2.0 (which is garbage-collected) combined with Grand Central Dispatch (for concurrency support and hardware abstraction, but at library rather than language level) and LLVM?
  • Will GC actually work as advertised?
  • Is someone going to step up and write a good cross-platform GUI library? (It doesn’t need to be native — it just needs to not suck — something that has conspicuously escaped both Java and Adobe.)
  • Will we have a reasonable deployment situation within — say — six months? E.g. will I be able to ship a binary that runs on Mac / Linux / Windows from a single source code base?
  • Will it get traction? There’s a lot of other interesting stuff around, albeit not so much targeting this specific audience (and it seems to me that if Go delivers it will appeal to a far broader audience than they’re targeting since it’s essentially very JavaScript / Python / PHP like with — what I think is — obviously bad stuff removed and better performance.

Conclusion

I’m not quite ready to jump on the Go bandwagon — I have some actual projects that have to ship to worry about — but it does look very, very attractive. More than any other “low level” language, it actually looks like fun to work with. The other newish language I’m attracted to — for quite different reasons — is Clojure. Because it lives on the Java VM it’s a lot more practical now than Go is likely to be for some time, and because it’s essentially “yet another Lisp” it affords a chance for me to really build some competence with a fundamentally novel (to me) programming paradigm. In a sense, Go is more attractive because it’s closer to my comfort zone — which speaks well for its chance of success. Clojure because of its conceptual novelty is almost guaranteed to remain obscure — the latest Lisp variant that Lisp adherents point to when complaining about <insert name of language people actually use here>. Time will tell.

Javascript Dick Waving FTW

Ars Technica’s rather gushy coverage of Squirrelfish Extreme (and the somewhat critical comments on that coverage) prompted me to do some of my own testing. Certainly, Squirrelfish Extreme keeps Webkit (and, ultimately, Safari) among the contenders for best endowed web browser, but my results were a little less definitive.

Google’s V8 benchmark suite is here. The Sunspider benchmark suite is here. Domaeo (the Mozilla test suite) is here. You might like to check them out in a few browsers. I just tried the current Safari release, the current Firefox release, and the Webkit Nightly Build. As I understand it, the Webkit speed claims are based on testing the JavaScript engines in isolation from the command line. Of course, very few folks run JavaScript this way…

The following are my very slipshod results obtained by firing up browsers and trying each benchmark once or twice. (I wasn’t careful about shutting stuff down, minimizing background activity, closing extra tabs, etc.)

Google’s V8 Benchmark Suite

This set of benchmarks takes less time to run and appears to be less thorough. (It certainly produces less immediately intelligible results… you just have to accept that big numbers are a good thing.)

Firefox 3.0.1 (OS X 10.5.5)

  • Score: 213
  • Richards: 226
  • DeltaBlue: 238
  • Crypto: 205
  • RayTrace: 162
  • EarleyBoyer: 245

Safari 3.1.2 (OS X 10.5.5)

  • Score: 218
  • Richards: 124
  • DeltaBlue: 173
  • Crypto: 187
  • RayTrace: 309
  • EarleyBoyer: 401

WebKit r36685 (OS X 10.5.5)

  • Score: 1312
  • Richards: 1317
  • DeltaBlue: 990
  • Crypto: 2553
  • RayTrace: 575
  • EarleyBoyer: 2035

This is just wonderful. (I tried to download a Firefox Nightly to check out Tracemonkey for myself but figuring out how took more than ten seconds, so I gave up. Here are some stats from Windows (note that these are running under VMWare Fusion, so it’s not fair to compare to the Mac figures directly).

Firefox 3.0.1 (Windows XP Professional/VMWare Fusion)

  • Score: 208
  • Richards: 207
  • DeltaBlue: 264
  • Crypto: 157
  • RayTrace: 180
  • EarleyBoyer: 251

Google Chrome (Windows XP Professional/VMWare Fusion)

  • Score: 2117
  • Richards: 2593
  • DeltaBlue: 2482
  • Crypto: 1692
  • RayTrace: 1456
  • EarleyBoyer: 2685

Internet Explorer 8.0 Beta 2

Note: IE8 interrupts each run with a “script is running slowly” dialog. I tried to click this off as quickly as I could but it definitely hurt the benchmark slightly. I ran it three times, dismissing the dialog as quickly as possible and took the best score.

  • Score: 58
  • Richards: 47
  • DeltaBlue: 52
  • Crypto: 50
  • RayTrace: 69
  • EarleyBoyer: 75

Sunspider

Sunspider’s results are really detailed, and I’m not going to post the whole things here, but I got the following bottom line results from a single run of each. In the Webkit case, this was with half a dozen tabs open (including this blog entry), while Chrome had two tabs open, one of which was displaying the results of its earlier benchmark (not very taxing, I expect).

Webkit r36685: 1614.6ms +/- 7.5% (running native under OS X)

Google Chrome: 1458.2ms +/- 0.4% (running under XP/VMWare; yes, faster is better)

Dromaeo Results

Again, the results are very detailed and also seem very real world (they use code from a lot of widely used JavaScript libraries such as ExtJS, Dojo, and Yahoo Widgets as part of the suite). You can see them for yourself here (Webkit) and here (Chrome). During the Chrome run, Chrome kind of seized up, whichwas probably Windows XP’s fault. So the overall result may be garbage, but if you look at the details you’ll find that V8 is clearly faster at some things and Squirrelfish at others. What this says to me is that both have quite a bit of headroom even yet.

Webkit r36685: 6652.60ms (running native under OSX)

Google Chrome: 10540.80ms (running under XP/VMWare)

So on both Google’s and Apple’s preferred benchmarks, I’m seeing better performance out of Google Chrome, but not by much. Dromaeo, which is a very DOM-intensive suite and also appears the most comprehensive, favors Webkit. (And, yes, IE8 is pitiful. No, I am not going to try Sunspider or Dromaeo on IE8.)

The really annoying thing is that Apple isn’t very aggressive about pushing advances in webkit out to the masses. Or, to put it another way, they use new versions of Safari to sell new versions of Mac OS X*. Chances are Squirrelfish Extreme‘s performance will be a key selling point of Snow Leopard when it ships, instead of a free upgrade to Safari, which is what it ought to be. In the mean time, we’ll probably need to use some other Webkit-based browser or Webkit Nightly, or simply use Firefox to enjoy better performance.

Note: as a comment — rather rudely — points out, Apple doesn’t charge for Safari upgrades. But it does hold new features back so it can announce them as though they were new OS features. (Edit: and, actually, it doesn’t support Safari on versions of Mac OS X before 10.4.x, so technically I was completely correct and the aptly named “bs” was full of bs.) A case in point is the debugging functionality added to Webkit over a year before it became available in Safari as part of Leopard. (And yes, you can get it for Tiger, but the marketing would have you believe it was a Leopard feature, and you simply can’t get it for Panther or earlier.)

And the really puzzling thing is that Microsoft seems to be happy to allow JavaScript performance to suck dog’s balls under IE8.

So, Javascript is fast now?

This site is the most interesting attempt I’ve found to seriously compare execution speed across languages. It should obviously be taken with more than a grain of salt since, apart from all the usual issues with benchmarks (artificiality, etc.), the quality of the code written in each language may vary greatly, as may the actual algorithms, and different algorithms may favor different languages, and different languages are used for different things and, obviously, benchmark comparisons try to force everything to do the same thing. Still, it is interesting.

Based on some quick spelunking, JavaScript (Spidermonkey) is comparable in speed to PHP or Ruby, perhaps a third as fast as Python, a fifth as fast as Lua, about a hundred times slower than Java, C# Mono, or Lisp, and a hundred and fifty or so times slower than C++.

What that says to me is that JavaScript should easily be able to get two or three times faster than Spidermonkey (because it has a lot in common with Python), but then it will get tough. It also speaks rather well of Lisp (which manages to be truly dynamic and still competitive with the big guns). Since it looks like Squirrelfish Extreme and V8 are both around twice as fast as Spidermonkey, and that if you took the best bits out of each from the detailed benchmarks and combined them you’d maybe get 50% faster again, it looks like we’re nearly out of low-hanging fruit in the JavaScript engine race.