[Chicago] Golang vs Python slides

[Chris Foresman]

It would probably look a lot like the existing C version. Though I suppose it would be easier to graft asynchronous execution if it was possible to program around the GIL. I’m not sure it would have significant performance improvement over the existing C Python; Go is fast, but it’s not necessarily faster than C. I think the primary advantage in Go is that it makes it relatively trivial to write concurrent systems, and such a system can be much faster than one that is not. It’s also typically much faster building binaries since the language is designed for optimal compilation. There might be optimizations that could be gained by building the runtime to use goroutines, but that is just wild speculation on my part; I don’t really know enough about the internals to say that’s true for sure.

Maybe if it were a special version of Python that implemented a keyword to execute a function in a goroutine, and support for channels, might get you some kind of hybrid that maybe performs about the same at regular code but a big boost in asynchronous code.

The more I think about it, the more I wish I had infinite free time to just re-write it in Go and see what happens :P



Chris Foresman
chris at chrisforesman.com



> On May 15, 2015, at 10:13 PM, kirby urner <kirby.urner at gmail.com> wrote:
> 
> how about Python *implemented* in Go, what would that look like?

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/chicago/attachments/20150516/a34f9138/attachment.html>