Ten Things You Can Do with GraalVM (2018)

[xonix]

For me the most relevant selling point of GraalVM is beeing able to compile to native self-contained executable. But on the other hand, Golang delivers on this aspect much better (or at least much faster).

[fulafel]

This gist narrowly predates Clojure GraalVM tooling that enables eg Babashka and Clojure usage in cloud FaaS with low cold start latency. And quick startup CLI tools like clj-kondo (a Clojure linter).

It seems there's some WebAssembly related things cooking GraalVM, currently just support for running wasm binaries but will be be interesting to see if eventually we get support for targeting WebAssembly (if WebAssembly one day gets good GC support).

[kasajian]

The decision to make Clojure JVM based and so slow to startup has slowed it's adoption to a much more significant level than anyone in the Clojure community is willing to accept.

[amelius]

One thing they didn't mention you can do with GraalVM:

Getting sued by Oracle.

/s

[chris_overseas]

/s noted, and sure Oracle are litigious, but the GPL community version of GraalVM, plus "early adopter" enterprise version licence[0] seems farily clear, and reasonable - at least as far as Oracle are concerned?

[0] https://www.oracle.com/downloads/licenses/graal-free-license... https://blogs.oracle.com/java/post/graalvm-free-license

[appleflaxen]

You nailed it

This was the first thing that came to mind when I saw this.

[psd1]

No sarcasm tag required. Exposure to oracle is a business risk. I expect any architect to think pretty hard before allowing anything owned by oracle into the stack.

[gavinray]

RIP Chris, he was a true gem in the community. Often answered questions I had on the Graal slack.

https://news.ycombinator.com/item?id=33893120

[manojlds]

Buzz around GraalVM died very quickly though. I had high hopes for it and innovation based on it.

[ccleve]

I still have high hopes. But man, the pace of development on this project is slow.