off-topic 2022-01-22 | Slack Archive

andy.fingerhut02:01:31

Was wondering if anyone knows of any sources of data or other evidence for this statement: Approximately X% of open source code was written by people paid to develop it. I know that percentage has changed over time, but I would suspect that as of 2022, it is well over 50%.

andy.fingerhut02:01:58

I think I mean something like "percentage of lines of code released as open source", not "percentage of projects, regardless of size or complexity". I am nearly certain that if you count every project as equal to every other, the percentage is under 10%, because of tons of tiny throwaway projects released as open source.

seancorfield02:01:36

It doesn't really answer your Q but it talks about the massive shift in OSS over the years https://techcrunch.com/2019/01/12/how-open-source-software-took-over-the-world/ and it notes that for OSS projects that have been commercialized by some company, likely over 90% of the lines in those projects comes from employees.

seancorfield02:01:35

I think it would be hard to measure "total number of lines" globally, but I think almost any "major" OSS project these days has corporate backing and there are people in those companies specifically paid to write/maintain those projects.

gklijs03:01:17

Yes, either directly, where the company is based on the project or one of the biggest users, or indirectly via a foundation, like with GraphQL and Rust. Seems almost like once an OSS project becomes both big and popular enough, one of those needs to happen, or the project will fail.

hiredman04:01:33

yeah, but the number of opensource projects that get that big is very small

hiredman04:01:11

react has a team of 9 out of 60k fulltime employees at facebook

hiredman04:01:51

I think, unless you come up with some criteria to weed out the long tail of opensource projects, the answer is going to be a vanishingly small percentage

☝️ 1

andy.fingerhut22:01:55

So even potentially harder to measure, but "used by at least 10 people in the world" code would weed out a significant part of the long tail, I think.

andy.fingerhut02:01:06

So some 1000 line Clojure project I did as a hobby is insignificant compared to Linux, for example, in the measurement I'm hoping to find data for.

sova-soars-the-sora20:01:24

recently i moved into a fresh apartment and i found the silence to be very welcome but i wanted some background sound. i find the sound of thunder soothing. i read once somewhere [reliable sourcing, right?] that astronauts on the space station are played the sound of thunder rolling once a day for some interval to help keep them sane well i must have imagined that information because i have not been able to corroborate it again anywhere but i now play the sound of thunder at significant volume for some intervals during the day and today i hear my neighbors blasting the sounds of ocean waves 😂

p-himik20:01:30

Good timing - I have quite some thunder right now, along with hail. :)

⚡ 1

jaide21:01:47

Every once in a while I get a bit tired of music but found this to be a nice UI for dynamically creating an ambient soundscape which includes thunder as well https://defonic.com/

sova-soars-the-sora22:01:16

@U8WFYMFRU nice contrib! 😃

mjw23:01:47

I'm curious to know, how many of you work in organizations that use dedicated QA teams to verify expected behavior before work can be considered done? Do your QA teams write tests, and either way, how do your organizations prevent the presence of QA from resulting in developers cutting corners and reducing quality under the assumption that QA will catch all the resulting problems? I'll admit that I have a very hard time understanding why developers shouldn't be expected to fully own the quality of their work (or why developers can't be tasked with smoke testing each others’ work), but so many organizations continue to utilize dedicated QA teams, so I can only assume there are reasons for continuing to do so.

2022-01-22

Channels