[joncampbelldev]

I'm not entirely clear what you mean. Are you referring to the disaster artist consultant mentioned in another comment? I don't think anything can stop someone like that from wrecking things short of not hiring them.

Or do you mean that small incremental improvements shouldn't be attempted because a proper review of whether they are worth it or not will be "wasted effort" if its decided they are not worth it?

Or something else? Not being sassy, genuinely unsure.

As for me personally, I would never hire an outside consultant to improve a codebase. The incentives are all wrong (long term maintenance to be done after they leave) and it's unlikely they'll be working on it for long enough to properly understand either the codebase or the domain.

[sys_64738]

It all goes back to what was the consultant actually hired to do? Was it to rewrite that code? Who is watching what the consultant actually does?

[joncampbelldev]

Please may you specify what consultant you're referring to here so I can answer more thoroughly.

In general, I highly doubt anyone would hire a consultant just to "rewrite code". Rewriting code is not a business objective. Improving application performance is, delivering features more efficiently and reliably is. These are some examples of objectives that might involve rewriting code (or not, not every codebase is a swamp in need of draining).

In the context of this article though, the code to be "improved" should be anything you touch. Did you use a function in building some new widget and you noticed it had no tests and no docs so it took you ages to integrate? Then add a test and the minimal doc string.

In terms of who is watching the consultant, I'm unclear why you're asking me, a person who has just said they would never hire a consultant to do anything with code. But if I had to, I would get someone of at least equivalent technical knowledge. However I'm pretty sure thats not how it usually works, which is probably why inept consulting companies rinse big companies for so much money for such lacklustre results.

But overall I really have no idea what you're talking about with regard to consultants and rewriting code. None of which seems to have much relevance to this article about making tiny improvements regularly. Unless your comment was in the wrong thread?