6 ms·

My pet peeve is the incorrect definition of the "10x dev" The "10xDev" is one that's 10 times more effective than the least effective developer; NOT 10x the av

My pet peeve is the incorrect definition of the "10x dev" The "10xDev" is one that's 10 times more effective than the least effective developer; NOT 10x the average dev nor 10x the median dev. The 10x is the absolute best versus the absolute worst. It's not the good vs the bad. Honestly there's no surprise here. What field doesn't have such a major difference between the most talented vs the least talented? The discussion is overblown. From Peopleware: "Count on the best people outperforming the worst by about 10:1" "Count on the best performer being about 2.5 times better than the median performer" "Count on the half that are better-than-median performers out-doing the other half by more than 2:1" Most "10x dev" articles should really be using "2.5x" if they're comparing the best with the median. Again, a 2.5x difference between the good and the bad is again, not surprising. It's also worth noting that the study Peopleware uses is from 1985 and sampled about 600 devs. The amount of pixels spilled over this basically ancient software study is honestly hilarious.

Tade0 6y ago
The "10xDev" is one that's 10 times more effective than the least effective developer; NOT 10x the average dev or 10x the median dev. What about people who have negative effectiveness?
- codemac 6y ago
  Or who have positive effectiveness when looking solely at their production, but negative effectiveness of the team or company's production?
  - paulryanrogers 6y ago
    At one point I think I was that dev. Software engineering is as much communication and human relationships as computer science.
- dolni 6y ago
  A friend of mine has used the phrase "activity != productivity" before. I think it fits these kinds of people well.
  - high_byte 6y ago
    great, I might use that one. I used the term output, but perhaps activity is a better description of it.
- high_byte 6y ago
  i^2 dev!!! hahaha
CharlesW 6y ago
> The "10xDev" is one that's 10 times more effective than the least effective developer; NOT 10x the average dev nor 10x the median dev. If it were typical to interpret "10X developer" that way, a new extremely unproductive developer would instantly make everyone else on the team 10X developers. (Personally, I've always considered it "10X the median developers productivity".)
high_byte 6y ago
I have to disagree on that one. because in practice, the difference between min and max would be several order of magnitudes larger than 10. (IMO) I did consider mentioning average or median, but of one self over different times rather than comparing different people. As others mentioned, 10x is symbolic not to be taken literally, and I have not done a study to feel comfortable with the number 2.5x (as I said, I personally don't agree with this number) btw, I'm looking up that study. do you mean "Peopleware - Productive Projects and Teams"?
- valbaca 6y ago
  Yes. Chapter 8. Pulls from several studies. See Notes, page 229
  - username90 6y ago
    But that study says that a majority of people submitted work with defects. So if you count the time until a defect free product is delivered it would massively increase the difference in performance, as the slower people typically had more defects.
SahAssar 6y ago
> The "10xDev" is one that's 10 times more effective than the least effective developer; NOT 10x the average dev nor 10x the median dev. I'm pretty sure that very few have this definition of the term.
username90 6y ago
Those studies just gave already filtered people a small project and only measured the students who actually completed the project. So, problems with those studies which would imply that the difference is bigger in the real world: 1. Filtering means that the population was more homogeneous than real world populations. 2. Talent matters more the larger project you work on as a very big part of being fast is designing good abstractions to simplify the problem. 3. Not counting people who didn't complete the project means that we remove the long tail of worst developers. If you do include the worst developers productivity which was 0 then it isn't 10x, it is infinite x.
theamk 6y ago
I don’t think this is is true. Imagine accidentally hiring someone who cannot program. Does that mean the entire team is suddenly 10x people?

Tade0 6y ago

The "10xDev" is one that's 10 times more effective than the least effective developer; NOT 10x the average dev or 10x the median dev. What about people who have negative effectiveness?

codemac 6y ago
Or who have positive effectiveness when looking solely at their production, but negative effectiveness of the team or company's production?
- paulryanrogers 6y ago
  At one point I think I was that dev. Software engineering is as much communication and human relationships as computer science.
dolni 6y ago
A friend of mine has used the phrase "activity != productivity" before. I think it fits these kinds of people well.
- high_byte 6y ago
  great, I might use that one. I used the term output, but perhaps activity is a better description of it.
high_byte 6y ago
i^2 dev!!! hahaha

CharlesW 6y ago

> The "10xDev" is one that's 10 times more effective than the least effective developer; NOT 10x the average dev nor 10x the median dev. If it were typical to interpret "10X developer" that way, a new extremely unproductive developer would instantly make everyone else on the team 10X developers. (Personally, I've always considered it "10X the median developers productivity".)

high_byte 6y ago

I have to disagree on that one. because in practice, the difference between min and max would be several order of magnitudes larger than 10. (IMO) I did consider mentioning average or median, but of one self over different times rather than comparing different people. As others mentioned, 10x is symbolic not to be taken literally, and I have not done a study to feel comfortable with the number 2.5x (as I said, I personally don't agree with this number) btw, I'm looking up that study. do you mean "Peopleware - Productive Projects and Teams"?

valbaca 6y ago
Yes. Chapter 8. Pulls from several studies. See Notes, page 229
- username90 6y ago
  But that study says that a majority of people submitted work with defects. So if you count the time until a defect free product is delivered it would massively increase the difference in performance, as the slower people typically had more defects.

SahAssar 6y ago

> The "10xDev" is one that's 10 times more effective than the least effective developer; NOT 10x the average dev nor 10x the median dev. I'm pretty sure that very few have this definition of the term.

username90 6y ago

Those studies just gave already filtered people a small project and only measured the students who actually completed the project. So, problems with those studies which would imply that the difference is bigger in the real world: 1. Filtering means that the population was more homogeneous than real world populations. 2. Talent matters more the larger project you work on as a very big part of being fast is designing good abstractions to simplify the problem. 3. Not counting people who didn't complete the project means that we remove the long tail of worst developers. If you do include the worst developers productivity which was 0 then it isn't 10x, it is infinite x.

theamk 6y ago

I don’t think this is is true. Imagine accidentally hiring someone who cannot program. Does that mean the entire team is suddenly 10x people?