[afarrell]

> If a rewrite is necessary, then the sooner you start the sooner you'll finish.

A 8-person team might easily rely on more C/C++ code than could be rewritten by them in 3 years. This jumps to 10-40 years if you include external Open Source codebases like Nginx.

So, the team needs to prioritize and to do things in steps. While they are rewriting some C/C++ code, they'll need to still maintain other code.

...and they need to solve enough business problems to keep the lights on.

[lmm]

> So, the team needs to prioritize and to do things in steps. While they are rewriting some C/C++ code, they'll need to still maintain other code.

Sure. But I don't think kind of portability-for-reliability work described in the article is effective work that should be prioritised. Yes, it will catch some bugs, but there's not actually much value in that.