4 ms·
Yeah that's the "they have x, why don't i?" attitude that causes a lot of problems. It should be an "everyone should have x!" attitude instead. At the end of t
by WWLink 6y ago
Yeah that's the "they have x, why don't i?" attitude that causes a lot of problems. It should be an "everyone should have x!" attitude instead.
At the end of the day, everyone's human, so there should be a little bit of slack in the staffing - long term - so that when someone's unavailable, nobody is drowning.
- owenmarshall 6y ago> At the end of the day, everyone's human, so there should be a little bit of slack in the staffing - long term - so that when someone's unavailable, nobody is drowning. I once as a very junior programmer asked a lead why we kept planning 75% of our capacity every month when we were hitting - and exceeding - our deliverables like clockwork. Then one month we had to do a death march to build features that had some $$$ tied to them. Did I ever learn why a little slack was critical.
- WrtCdEvrydy 6y agoI had a manager who kept saying "don't use fear-based estimates" since he had been picked to right the train. He looked good in front of the directors until he had to deploy the thing that was obviously rushed since "we can't use fear-based estimates". He wasn't able to pat his own back before the alert on the outage went out (all system outage)
- nicbou 6y agoThat's the argument I made above. If they get paid 100% of their salary for 80% of the output, so should everyone else. I compared it to smokers taking smoke breaks while non-smokers aren't allowed to take random 5 minute breaks.
- rootusrootus 6y agoThe problem is that nobody can even define output accurately. I know people who output more in 25% of the time than some of their peers manage at 100%. But that's not how compensation in this industry is structured, so it is moot.
- nicbou 6y agoSure, but it's fair to say that if someone has 75% availability and must miss meetings because of external events, their output is affected. Besides, I am talking about a drop in output for a given employee, so the comparison to other employees is pointless. If my productivity dropped by 25%, I'd hear about it pretty quickly, and I doubt my explanation would matter much in my performance review.
- rdtwo 6y agoYour company must have some good metrics or they are measuring clicks per minute if you think they would notice 25% output growth very quickly. Likely you just wouldn’t get promoted and that might happen to your coworkers in due time
- nicbou 6y agoYou are getting hung up on arbitrary numbers in a fictitious example
- sokoloff 6y agoAt a place where smokers can take random 5-minute breaks, I can practically guarantee that non-smokers can as well.
- nicbou 6y agoSmoke breaks are not really an issue for white collar workers, but I don't think it invalidates the relationship between the two situations. I don't think this has to be explicitly stated.
- myko 6y ago> At a place where smokers can take random 5-minute breaks, I can practically guarantee that non-smokers can as well. Never seemed okay for non-smokers to take these breaks in the Army
- AzN1337c0d3r 6y agoI have a coworker who doesn't smoke but will call it his smoke break and take the 10 minutes and just go for a walk around the building.
- rdtwo 6y agoAnd that comparison breaks down as soon as you go from making widgets to creating nebulous IP. Who can really say if the smoker on non smoker writes better code in their 8 hrs