The article harps on how it would be nice to have an objective and simply measurable metric. I'd refer to the classic statistics quote "The combination of some data and an aching desire for an answer does not ensure that a reasonable answer can be extracted from a given body of data." which seems to be the case here.
There's probably a consensus that reading code is harder than writing code. You're not going to get a good evaluation of a developer's output without looking at that output, and it's not going to be simple, there are no shortcuts - if you need that information, you have to have a competent developer look at that output and they'd then be able to tell how good or bad the quality is, and how the quantity of work relates to the time spent. This can be done in code reviews.
Other than that, you'll simply have to accept that you'll use subjective metrics - which is also fine, this isn't the only domain where accurate measurements are hard/expensive enough to do them sparingly.
There's probably a consensus that reading code is harder than writing code. You're not going to get a good evaluation of a developer's output without looking at that output, and it's not going to be simple, there are no shortcuts - if you need that information, you have to have a competent developer look at that output and they'd then be able to tell how good or bad the quality is, and how the quantity of work relates to the time spent. This can be done in code reviews.
Other than that, you'll simply have to accept that you'll use subjective metrics - which is also fine, this isn't the only domain where accurate measurements are hard/expensive enough to do them sparingly.