Hey, if it ends up saving time and stress after those two days it was worthwhile.
Seems to have become one of the fundamental rules of the Internet now, I approve 👍
it wont, it’s just more enjoyable to automate a task than to do it manually
There’s something really satisfying about running a script that you know would save time. Even if the overall time is probably a negative.
I wrote a script that would log me into our AWS EKS stuff. I typically would have to copy these 7 lines and look up which cluster version I’d need. One of my lines just pulls all the clusters and I use fzf to select the cluster I wanted. Takes away all the pain and makes me feel smug. Love it.
It’s a two minute task, but it happens randomly between the hours of “romantic dinner with my wife” and “ten minutes after the baby finally went down for a nap”.
Automation also cuts down on mistakes.
Or greatly amplifies them if you coded it wrong.
deleted by creator
Test 1: Locking process. Unexpected error encountered. Exiting immediately. Test 2: Waiting for process unlock to proceed. Test 3: Waiting for process unlock to proceed. Test ...
The exception is if it’s open source and can save thousands of people two seconds.
Doesn’t even have to be the case. A 2min task done every (work)day, takes up a bit over 7 hours/year. After 2½ years it will be a benefit to have automated it!
The benefit of automating is really measured in hair loss and extra time to grab another coffee.
Only if the requirements stay the same for 2.5 years. Otherwise there’s probably another week of time trying to update the initial work, then just throwing it away and making a new solution that’s theoretically easier to update.
If changing requirements mean you need to update the script, then updating the script is part of your job. QED. I don’t see the problem with a little job security.
Yeah, just add it to the ‘amount of work you are putting’.
When setting up git hooks for my project, I looked at other’s OSS hooks first. That shaved off significant hours off of my Research.
Yeah, and you build skills and reusable code base that’ll be useful for automating/ simplifying future tasks 😎
Some years of this, you get to the point where you can solve damn near everything quickly and people think you’re some magical shit-wizard
I don’t know how to code, but I’m already a shit wizard.
you get to the point where you can solve damn near everything quickly and people think you’re some magical shit-wizard
This is basically my work life, and its almost a problem because I’m the first guy people call when they need something done.
The perils of being competent. /s
Fucking competence. I wish I was bumbling fool with severe Dunning-Kruger more often than I care to admit.
And if it’s a task that will need to be done thousands of times a month or even year, you should thank them for it.
Obviusly it has to be done once evey year…or a decade.
With this as a guide, it would be enough if the task needed to be done twice per day to break even after five years.
That’s also only assuming that one person is doing that 2 minute task. If you automate a 2 minute task for 5 people, then it’s only 1 year.
I just realised I was being waay more efficient than I needed to be.
deleted by creator
Any shell is the duct tape of computers.
The best engineers can do anything with duct tape.
I can spend 2 minutes scanning a page for a certain word every time I need to search for something.
But I’m very happy somebody spent the time to code Ctrl+F.
Ctrl+f code has to be some of the most efficient automation ever written. Time spent was probably about day and the time saved in work hours is probably in the trillions at this point.
Yes, but since it runs automatically every day and emails my team the results, I don’t have to remember to do it on my own. I don’t even have to be working that day. Taking “my ADHD memory” out of the system is always a win.
Accessibility in general seems to be a huge benefit to automation a lot of people here including op are overlooking, which is a great shame but unfortunately not surprising…
A two minute task done with 0 errors and once it’s done that 1440 times it’s a net positive
Garbage collection/memory safety in a nutshell.
If you choose to do a task manually over and over instead of automating it, you might as well go dig ditches and hammer nails.
Programmers after slacking off for two days to improve their factories in Shapez2 because they already automated all their tasks.
Non automated tasks remain in the inbox for a week, so spending 2 days automating them means they’re finished earlier.
One more advantage is that you now have the full process well documented (via code) and if you realize some change is needed you can repeat the task quickly.
deleted by creator
Sometimes I’ve gone the other way. My manager complained once that I didn’t automate a task to get business metrics, and I responded to say that it currently takes me 1 minute to pull the metrics and paste them into a spreadsheet and print a PDF. Automating this would take at least several days, for a slide that changes constantly, and where the data often requires a deep dive into why the data is how it is. What’s the point in automating something that I already need to manually look at?
They raised it with our PM, and their response was “fair, I wish I hadn’t bothered to automate things last year”.
If the cost would give a higher benefit, sure, automate it so that it spits out a spreadsheet every week and do the manual stuff separately - but automating something “because you can” is junior level shit. My time is valuable, let me work on stuff that actually matters.
