4 ms·

> Musk’s team has already tried to evaluate the productivity of Twitter employees by asking engineers to print out the code that they have written in the last 3

> Musk’s team has already tried to evaluate the productivity of Twitter employees by asking engineers to print out the code that they have written in the last 30 to 60 days. Musk also brought in Tesla engineers to look over Twitter code. At one company I worked we had a bug in the software for the device we were manufacturing. There were millions of units at peoples' homes and the bug caused couple hundred of these units to loose the content of their flash, every day. This in turn a huge ongoing loss both financial and reputational for the company. Because of the nature of the bug, we spent something like 3 months trying to find it. We even built a special lab with dozens of units hooked to specially designed rig to have a chance to observe one unit in the process of getting wiped. Finally, one guy with an oscilloscope cracked it. It was electromagnetic interference between two different circuits within device causing the flash to receive a command telling it to wipe itself clean. The fix was one line, two characters really. I am sure Tesla engineers will do a good job evaluating worth of each one of those lines people printed out.

marvin 4y ago
If I had planned and written those two characters, I would have appended a comment like yours and any sensible engineer would have praised the solution. It’s fairly obvious their metric isn’t «number of lines of code».
- twawaaay 4y ago
  My point is, looking at the code you are unlikely to be able to judge the value of it. Code is not valuable in itself. Code is liability because it costs a lot now and will have to be maintained in the future. Frequently, best solutions and decisions are the ones that result in avoiding to write any code. As to code, I put a lot of effort into solving complex problems with simple code. Anybody can write complex code. It takes experience and effort to make stuff look simple.
  - kamaal 4y ago
    I think their point is in the era of OKRs, they are more interesting in knowing if you have been coding up the next billion dollar app, or working on maintenance stuff to keep lights on. While the latter is equally important, its the former that brings in the next billions in profit.
fleddr 4y ago
You know you can talk back, right? When you'd bring that tiny print of code and give your explanation, do you honestly believe a Tesla engineer is not equipped to understand the value of those few lines?

marvin 4y ago

If I had planned and written those two characters, I would have appended a comment like yours and any sensible engineer would have praised the solution. It’s fairly obvious their metric isn’t «number of lines of code».

twawaaay 4y ago
My point is, looking at the code you are unlikely to be able to judge the value of it. Code is not valuable in itself. Code is liability because it costs a lot now and will have to be maintained in the future. Frequently, best solutions and decisions are the ones that result in avoiding to write any code. As to code, I put a lot of effort into solving complex problems with simple code. Anybody can write complex code. It takes experience and effort to make stuff look simple.
- kamaal 4y ago
  I think their point is in the era of OKRs, they are more interesting in knowing if you have been coding up the next billion dollar app, or working on maintenance stuff to keep lights on. While the latter is equally important, its the former that brings in the next billions in profit.

fleddr 4y ago

You know you can talk back, right? When you'd bring that tiny print of code and give your explanation, do you honestly believe a Tesla engineer is not equipped to understand the value of those few lines?