Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Study finds 83% of software developers feel burnout

usehaystack.io

Post reply on HN
Print view View original

211–220 of 308 posts

Re: Study finds 83% of software developers feel burnout

#211

Post by eeZah7Ux »

sethammons wrote: post #83
varelaz wrote: post #16

Developers are getting too spoiled. Did someone do the same research for teachers for example.

I've been a teacher (and a manager, all levels of software engineer up to principal, a construction worker, an inner-city and high gang-presence high school math teacher, a graphic designer, and worked under a yard maintenance guy from Mexico, and several other small jobs. The software stuff has been among the easiest jobs (even with mission critical systems failing in the middle of the night affecting millions of do…

You are considering only anecdata from one person.

Only in one country.

And comparing software development only with stressful jobs.

And if you still think that burnout is not a bigger issue in the software world, go around and look at how many 40+ or 50+ years old work in software.

Re: Study finds 83% of software developers feel burnout

#212

Post by kube-system »

hallway_monitor wrote: post #195
Spartan-S63 wrote: post #156

Earlier quoted context omitted.

I agree. Ideally, we get to an end-state where it doesn't matter what hours (or how many) you work as long as your team's outcomes are being met (and the outcomes are reasonable). There are a lot of caveats and complexities in that statement, but it's a state we can keep pushing towards as an industry.

I really like this idea but I'm not sure how to implement it. If Bob works twice as fast as Jim, does Bob get to go home at noon or does Jim have to work late?

There are various types of employee pay arrangements other than hourly and salary that do this (even though, some salaried jobs do work the way you describe), like piece work.

https://en.wikipedia.org/wiki/Piece_work

And it's already really common for non-employee contract work to function that way. The guy who mows my lawn gets paid the same amount no matter what speed he mows it.

That being said, there are plenty of salaried jobs where people are assigned measurable responsibilities to attend to and don't have assigned work hours.

Re: Study finds 83% of software developers feel burnout

#213

Post by ottermax8 »

throwawayswede wrote: post #56

This is such a great example of bullshit studies. Yeah I'm gonna go on a limb and say this is a marketing campaign. It's an insult to science to call this a "study". From the pdf: > The fieldwork of this study was carried out at remarkable speed, from the 23rd to 24th of June 2021 Data was collected in 1 day. Is this a joke? No it's not! > The population sampled included Software Engineers aged 18+ living in the Unit…

Yeah I can do a survey too :)

Re: Study finds 83% of software developers feel burnout

#214

Post by kristopolous »

titzer wrote: post #2

If you were a professional painter and every couple of months, someone snuck into your art studio, threw out all your brushes, paint, canvas, and then dumped a huge load of requests for you to "fix" your old paintings, I'm pretty sure you'd get burned out pretty quick.

It's more like you get scrawlings from toddlers on your desk every day and the boss says "these were supposed to be the blueprints to the next skyscraper. I paid a lot for this so you can't throw it away" and then you need to work with crayon scrawlings and somehow get the job done

Re: Study finds 83% of software developers feel burnout

#215

Post by techsin101 »

abz10 wrote: post #160

It could be genetic. RCCX gene mutations may make people more susceptible to CF/ME and being a software developer. For most people CF/ME will present as burnout.

it's very weird when people take correlation and present it as causation. I feel there are at least few steps in statement that people with that gene mutation are more susceptible to burn out. Like people with that gene mutation "like to be independent", "like variety", and therefore lack of those in regular 9-5 software development leads to burn out. How it is presented as a causation has only one practical application. Gene xyz is bad, don't hire people with gene xyz.

Re: Study finds 83% of software developers feel burnout

#216

Post by mateo411 »

danenania wrote: post #203
mateo411 wrote: post #201

Earlier quoted context omitted.

If we use a 95% confidence interval and we have a sample size of 258 then the margin of error is around 4.5%. If the sample size is 130, then the margin of error will be around 6.5%. That's a little high, but probably still good enough.

The sample size you need also depends on variance and the presence of confounding variables. It’s impossible to say whether n is “big enough” in the abstract; it depends completely on what you are measuring and how you measure it.

We are sampling a Bernoulli distribution for measuring burn out. A software developer could be burnt out/ not burnt out.

The variance of a Bernoulli distribution is P(1-P), which is (.83)(.17), since 83% of the respondents answered that they were burnt out.

Re: Study finds 83% of software developers feel burnout

#217

Post by danenania »

mateo411 wrote: post #216
danenania wrote: post #203

Earlier quoted context omitted.

The sample size you need also depends on variance and the presence of confounding variables. It’s impossible to say whether n is “big enough” in the abstract; it depends completely on what you are measuring and how you measure it.

We are sampling a Bernoulli distribution for measuring burn out. A software developer could be burnt out/ not burnt out. The variance of a Bernoulli distribution is P(1-P), which is (.83)(.17), since 83% of the respondents answered that they were burnt out.

But why are they "burned out"? Is it because they're a software developer or because they live in an advanced post-industrial civilization in the midst of a global pandemic? Could they feel burned out today but not tomorrow? Has the sun come out in the UK at any time in the last two weeks? Do all the participants even agree on the definition of "burnout"? Were there five other similar studies done that we don't know about that were discarded because they didn't produce the desired click-baity conclusion?

I could go on and on.

Re: Study finds 83% of software developers feel burnout

#218

Post by milesvp »

SketchySeaBeast wrote: post #197
milesvp wrote: post #194

Earlier quoted context omitted.

I personally really like maintenance programming, but I wonder if I've been lucky. All the systems I've ever inherited, I've been able to make impactful changes to them such that they stopped being that one system that causes everyone production headaches. Even the systems that I couldn't make much changes with directly, I was able to isolate, and route other systems around them, such that new data stopped depending…

You may have been incredibly lucky, or have a secret sauce I don't understand. I've worked a pair of maintenace positions. The previous one was incredibly creaky and had no other systems, so changes were scary and there was no recourse to move to something else. The only I'm working now I'm maintaining client code so I don't have any agency there either. Maybe it's agency? Ultimately I don't feel like my contribution…

Yeah, maybe agency. I don't know. It helps that I'm fairly social for an introverted engineer. One of the biggest systems change I ever made was related to a legacy url that the main page had to use. It seems like the type of thing that should be simple to make, but due to the sheer amounts of redirects and multiple systems involved it was really painful to change the primary domain. Only like 2 people in the company had any idea the amount of effort necessary to avoid breaking countless things. But when I finally flipped the switch I spent the entire afternoon running around the company and showing random people I knew in other departments the home page. I got a lot of "huh?"s. literally no one was impressed. But, I can have an infectious enthusiasm, so as I pointed to the URL, and said "look! that was actually really hard to do", people would at least come around and say "ok? good job?". They still weren't particularly impressed, but they at least understood it wasn't trivial. Couple this with a good PM who I knew was also evangelizing some of this stuff, and you can get a lot of agency over time in an org.

So maybe the trick is evangelizing? I really don't know, I just know I've been successful, often in spite of the organization I was in. I'm also constantly aware of the amount of luck involved in everything I do, so it's always hard for me to believe that I was the important factor.

Re: Study finds 83% of software developers feel burnout

#219

Post by walshemj »

p_j_w wrote: post #133
throwawayswede wrote: post #56

This is such a great example of bullshit studies. Yeah I'm gonna go on a limb and say this is a marketing campaign. It's an insult to science to call this a "study". From the pdf: > The fieldwork of this study was carried out at remarkable speed, from the 23rd to 24th of June 2021 Data was collected in 1 day. Is this a joke? No it's not! > The population sampled included Software Engineers aged 18+ living in the Unit…

Your criticism of the sample size is unfounded. You absolutely can have meaningful results with only 258 people provided you have good sampling. If you want to criticize their sampling, then by all means do it, but that's not what you've done here.

Normal election poling uses 1k sample size as a minimum and 3k is preferred.
Post reply on HN