When you encounter in your work some boring, repetitive (possibly not-programming) task how do you deal with it? Do you try to automate it immediately? Do you postpone automation till you encounter the task twice? Do you try to delegate it? Do you try to find something interesting in it? What do you do to make the work more joyful? Learn deeper to make it more interesting? Learn deeper immediately or when you need it regularly?
4
-
2Read stackoverflow and programmers.stackexchange a lot. =P j/k– TonyCommented Nov 16, 2010 at 13:02
-
7I delegate it to a junior team member, he-he :D– JasCommented Nov 16, 2010 at 13:13
-
Anyway you need to take break every 25 minutes for five minutes. In five minutes, you can socialize, drink coffee, call a friend, do some table exercises, goto loo, etc. These things will keep you occupied otherwise.– PradeepGBCommented Nov 16, 2010 at 17:27
-
I am not the downvoter but I think it is because you said "their fault". Some people react like Pavlovian dogs when confronted with "bad" words.– ThomasXCommented Apr 27, 2012 at 10:00
Add a comment
|
8 Answers
If automating it will take longer than getting the task done, then I'll postpone automating it until I have to do it a second time. But if I can automate the task faster than I can do it, then it's automated without hesitation.
No sense wasting time.
-
17There was this guy who automated sending out his weekly progress reports to his boss... Somebody started suspecting something is fishy half a year after the guy left for another company and the project reached 130% completion.– SF.Commented Jul 13, 2011 at 9:02
-
I think "the second doing it manually starts to piss you off" is a good criterion. Commented Aug 31, 2013 at 0:20
If there is a point and value in automating it, I will absolutely do so - time considering.
Example:
- Daily System Statistic Collection that could take up to an hour manually - Automate
- A Data Collection Routine once in a blue moon taking +- 15mins - I just do it
-
4+1 for automating where it's worth it, though I'd say even if the time taken to automate is equal to the saving it's a good thing given that it's then largely reusable and there's potentially been some learning in there. Commented Nov 16, 2010 at 10:35
-
@Jon, I belive that to write a good script you need to do the process manually once. E.i. quite unlikely that full automation will take <= time then at least half-manual processing (you use some programs for computer-related tasks anyway, right? :) )– AlexeyCommented Nov 16, 2010 at 17:52
-
considering 1st case - will you try to automate it at the very first day, or you do it first time manually and automate next day, or do it first time manually and automate right after that?– AlexeyCommented Nov 16, 2010 at 17:54
-
2@Alexey - Only automate once you understand it enough to automate it. If that takes a day, a week or a month of manually doing it, so be it. The last thing you want is an automated tool that breaks what was actually required. Commented Nov 16, 2010 at 19:43
It depends what the job is.
Some boring repetitive tasks (like going to weekly progress meetings) can't be automated. You just have to put up with them.
Some (like not having an automated build system) you might put up with forever if you only do the build once a year.
And some - you try to automate because it lets you do other more productive things, and also removes the possibility of human error.
answered Nov 16, 2010 at 9:07
If I have a boring task requiring minimal brain power, I save it either for those mornings when I am lucky not to have killed myself on the way to work or for long long Friday afternoons.
Unless I can automate it of course!
I would rather spend up to 1.5 times what it would take to do it manually, automating it instead. I attempt to justify this to myself beacuse I'm learning something whilst I automate and keeping my brain active.
-
Are you sure you're doing your math right? If you have to do it twice a week, and it takes you 10 times as long to automate it than write it, then your payback period is still 5 weeks. That's a pretty sweet payback period for any business. A lot will consider it a success if the payback period is 12 months. Commented Nov 17, 2010 at 11:35
Typically when I am tasked with something repetitive or boring I will almost always just do it atleast once. If I can see that the task may be asked of me or someone else again (even if I'm told its a one off) I try and take notes outlining the steps that were done. That way should I chose to add some automation I know exactly what I can automate.
But mainly it all depends on the how long the task takes to complete versus the time to automate. If it's a 10-15 minute process done once a month probably won't automate. May be a good diversion for moment. 10-15 minute process daily or even weekly, probably will automate. Man hours then start to add up.
-
Nice answer! I observed from my experience, that if i take notes how to do something it improves even manual performing of the task and makes it less painful.– AlexeyCommented Jul 23, 2011 at 7:00
Like others have said: if it's common enough to be worth automating, I automate it. If not, I just slog through it.
And if it starts to feel like the whole job is just one big pile of boring, repetitive tasks, then I burn out and move on. :)
answered Nov 16, 2010 at 19:51
I have three strategies for dealing with boring, repetitive tasks:
- Automate them.
- Assign them to an intern or junior programmer.
- Do them while listening to rock music.
And like @Bobby Tables said, if I'm required to do too many boring, repetitive tasks, then I move on.
