I don’t see how it’s possible to get a 5x win in compile performance just by changing the build system. If you do a CPU profile of the build of a large project it’s almost all in rustc.
-
-
This Tweet is unavailable.
-
Replying to @sayrer @BRIAN_____ and
Man, if a JVM were a dependency to build Rust the language would be dead by now.
2 replies 0 retweets 3 likes -
This Tweet is unavailable.
-
Replying to @sayrer @BRIAN_____ and
It trumps all other concerns. Because if Rust weren’t easy to get started with, we wouldn’t be having this conversation in the first place. Language ecosystems from small players have to optimize for ease of use, otherwise they die. Google can afford to do otherwise.
1 reply 0 retweets 2 likes -
This Tweet is unavailable.
-
I don’t recall ease-of-use as part of the pitch to me either, I have to say. It’s great, but only once it’s off the ground enough that people _want_ to use it. EOU promotes adoption, and you don’t want a lot of 0.1 adoption anyway IMO.
1 reply 0 retweets 1 like -
This whole conversation is driving me crazy because we’re starting from the position “Bazel good, Cargo bad” without even stopping to understand what the specific problems that Rust is currently facing are. So far I’ve heard talk about caching LLVM, but that is a Travis issue.
2 replies 0 retweets 1 like
I don’t have enough domain experience to have an opinion on whether, or to who, we should outsource our CI or not. I am certainly willing to believe that we could do better than Travis.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.