Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
The sunk cost fallacy: Devs describe how it almost destroyed them (gamasutra.com)
101 points by tnolet on Dec 24, 2017 | hide | past | favorite | 20 comments


And for every article like this, there's another article explaining how Apple got big because they insisted on perfection.


Relevance? You can seek perfection while ignoring sunk costs.

Avoiding the fallacy just means you compare the costs of reaching "perfection" from your current position to the value of reaching perfection. The cost you ignore is the cost you've spent to reach your current position, because that cost was incurred regardless of your decisions now.


On how to deal with sunk costs, Seth Godin once said [1]

"Everything you own, all the clothing in your closet, your academic achievements and beyond is simply a gift. It is a gift that your past self is giving to your present self, and it’s up to you to decide whether you want that gift today.

It is as simple as that–you owe your past self nothing, other than the consideration of whether these gifts are helpful in the here and now."

I use this way of framing business decisions a lot nowadays. It's a very effective way of removing the emotional baggage that sunk costs come with, and focus on the actual goals you should be achieving.

1 http://www.breakthetwitch.com/gift-past-self/


I must admit, I've been put off Seth Godin recently.

But that quote you posted is something worth remembering.


Which, if you actually read the words of the people there was never the case.

They always strive for elegance and perfection but routinely compromised


You can tell they did some pragmatic compromising. Because, well, they're still in business.


Hah! Quite.


But how could they stop when they knew it could probably be half a milimeter thinner! :O


There are plenty of Apple failures.

https://www.macworld.co.uk/feature/apple/11-worst-apple-prod...

The truth is, in hindsight, you can make history and success/failure stories look like the result of anything you want if you think abstractly enough.


Developers didn't insist for perfection though. One of the greatest product designers of all time did.


You mean like this?

https://youtu.be/w96oTapYTKk


sounds like banner saga didn't need a ps vita port so much as it needed a vita re-implementation


Kinda sounds like it needed a performant reimplementation, period. This is what the game looks like. There's just a flat isometric plane of squares with animated sprites in it.

https://youtu.be/OBsRPhCm5sU?t=899

The Vita and the PSP both have a lot of tactical RPGs that are a lot higher in complexity, with actual 3D environments as well as randomly generated ones (much harder to handle than pre-made hard-coded ones).

Compare with best-in-class on that platform: https://youtu.be/nUhspS9VddQ?t=3539


This reminds me of Binding of Isaac. The game kept getting re-written poorly so it never ran well on some devices, such as the DS.


Yep. That one i wanted to play, but never could, since it literally wouldn't run at 60 fps anywhere ever.


What a painful process and with the promise of only a small return. It'd be better working for a bank or something.


Such is the wild life of a game programmer...


I love throwing out code. It makes me feel clean. :)


Throwing out code is easy, throwing out hours you've spent investing into something is hard (e.g. learning, promoting, developing, money, etc).

These aren't technical issues with technical solutions, these are issues relating to people's egos and their willingness to throw out acquired skills and effort.

For one example, you could go from being the biggest expert in that area to being behind others just by turning against a technology you've helped spearhead. I've seen it happen (and people resist it who knew they'd gone down a dead end).


I fully get what you are saying, but pruning code is a valid thing to do on it's own, and it doesn't necessarily mean changing tech or whatever.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: