# Is there an error in 'The second metric is the lack of violations'?

The key indicator is an exploitable population. The more exploitable workers are the more likely they are being exploited. This means we need a way to measure each population’s exploitability. The second metric is the lack of violations. This is not an indicator of a lack of exploitation; a lack of known exploitation may be an indication that a domain may be running under-the-radar.

He underlined

The second metric is the lack of violations.

I don't see an English error here though I am a native speaker and he is a native German speaker so he actually learned English at one point where as I just 'know it.'

The comment was made with a sticky note in a PDF and I think he was on a transatlantic flight when he reviewed this paper; the sticky note looks like it is pointing exactly to "The second metric is the lack of violations." He marked "We need to know the domains, the expected rate of violations in each domain, and then the actual degree of violations in each domain." with the same comment.

• Seems fine in German. Maybe that's the problem?
– user14394
Sep 13 '16 at 0:51
• @user14394 I like you Sep 13 '16 at 5:37
• I ran this text through grammark.org and it came back more or less clean;1 passive voice case, 'being' is wordy, too many normalizations, and no transitions. Sep 13 '16 at 16:10
• You're a native speaker, and you spell English with a lower case e and German with g? Sep 13 '16 at 16:14
• Did the advisor actually write "fix english" on the page, or did he just underline the sentence? I would remark on the overuse of "exploit"+ending, which appears four times in the first three sentences and twice in the last sentence. Perhaps he underlined that sentence to point out that you've referred to the "second metric" without clearly identifying the first. Also, I wouldn't call the lack of violations a metric, but instead would refer to the number of violations as a metric. A metric is usually a measurable quantity, whereas a lack is a boolean. Sep 13 '16 at 16:39

You posit that there are two metrics

1) exploitability
2) lack of violations

For the first, you propose that some sort of measure for exploitability must be created.

For the second, you state it as if the measure is already known. By pointing out the need for

1) domain identification
2) normal run rate of violations
3) measured run rate of violations

it would be a way to identify if there is a lack of violations, basically how the metric is calculated, which you had not mentioned. The same argument could be made for the calculation of "exploitability" and I;m not sure why your advisor did not make them.

• Thank you Peter. The grammar is technically fine then. It is the logic that creates confusion. Is this correct? Oct 7 '16 at 12:53
• Yes, that is how I would understand your advisor's comments, but you could always just ask him what he meant. I suppose you could reword it to be "A second indicator is the absence of violations" since it would draw parallels with "exploitability" being an indicator. Oct 7 '16 at 13:15
• Thank you - I should use the same terminology metric!=indicator, so there is a mismatch there. That could be the source of confusion. I want to get my ducks in a row before asking about grammar (though he is patient with grammar issues). I will ask him. Oct 7 '16 at 23:16