1 |
On Mon, 10 Apr 2017 20:41:42 -0400 |
2 |
"William L. Thomson Jr." <wlt-ml@××××××.com> wrote: |
3 |
|
4 |
> Git shows really low stats. You cannot argue with facts. |
5 |
|
6 |
Please stop using git as a measure of contribution. Its inaccurate at best. |
7 |
|
8 |
I've had to hand test and triage dozens of bugs in the time it takes |
9 |
to do one commit. |
10 |
|
11 |
A lot of useful developer time is entirely invisible from Git, git is only |
12 |
one viewport, and if you use it on its own, all you see is a distorted reality. |
13 |
|
14 |
For instance, Arch testers *by far* have the highest commit numbers. |
15 |
|
16 |
But I don't think it would be fair to read that as either an indication that they do /more/ work |
17 |
than other developers, nor would I discredit that number as being entirely meaningless. |
18 |
|
19 |
Just the *kind* of work they do necessitates they do more commits in the course |
20 |
of that work than is otherwise normal. |
21 |
|
22 |
But you could theoretically create a bot that pumps out commits, and you wouldn't treat |
23 |
it as the glorious saviour developer. |
24 |
|
25 |
In short: |
26 |
|
27 |
"git" |
28 |
|
29 |
Is not |
30 |
|
31 |
"facts" |