Cool! Have you measured the impact of any bundle sizes to see what the load time improvement might be? Note that JS engines are changing all the time, and the impact can vary from codebase to codebase (even have a negative impact in some cases!), so it's good to actually be measuring this stuff.

In optimize-js I've been meaning to overhaul the benchmark system to provide better measurements, but if you'd like to use something yourself, I'd recommend marky for adding performance.marks and measures and then just mark the time before inserting a script element into a page and then mark the time after. Curious to know what you find!

Is there anywhere we can get an ELI5 (explain like I'm 5) explanation for this? I don't quite understand why this causes such a performance improvement?

@citypaul

http://stackoverflow.com/questions/440739/what-do-parentheses-surrounding-a-object-function-class-declaration-mean

@citypaul This description specifically addresses the performance improvement https://gist.github.com/getify/938c4a46013244cbd3cd750c0bc767ce

@nolanlawson, we ourselves don't have large enough example repos to warrent measurable results, we have had folks like @MikaAK who has used the separate plugin I wrote emulating this behavior. We would love to collect a list off large applications so we can do further perf analysis. Hard to get hands on in an easy way.

@gavinwahl Thanks for that gist. That's exactly what I was looking for 👍

We would love to collect a list off large applications so we can do further perf analysis.

@TheLarkInn A large real-world repo with webpack is
https://github.com/instructure/canvas-lms

Thank you @statianzo! Have you tested out a build against webpack in master? I'm interesting to see the gains. Will probably test it myself anyways.

@TheLarkInn I've been doing some internal tests at Domo with @aaronfrost using the webpack-1 branch. I'll work on getting you some numbers next week.

Much appreciated!