5 ms·

> If you keep it framed like that, it's more exciting to have a successful mission & likely to continue to get funding, than if you set high expectations and fa

> If you keep it framed like that, it's more exciting to have a successful mission & likely to continue to get funding, than if you set high expectations and fail. This seems to be the more plausible reason. You keep the bar for success low so you keep getting funding for subsequent missions. Isn't that pretty much how every organization with public funding sets goals?

TaylorAlexander 5y ago
I mean the low bar is that they fly three robots to Mars and two of them land safely and one of them makes multiple flights on the surface of another world so it’s a pretty high low bar!
- sfifs 5y ago
  I mean sure it's cool but I think the point is they've been so consistently spectacularly successful in blowing past mission objectives that it's a bit hard to imagine achieving this wasn't the "expected result" (maybe we can cut a bit of slack for Pathfinder/Sojourner which was the first complex mission after the Climate Orbiter debacle and over-delivered only 12x).
  - StavrosK 5y ago
    The issue here is that there's a mismatch between the probability distribution and result. In reality, either the equipment will fail some time on entry/initialization and won't be able to complete any missions, or it will work for one and thus probably work for a hundred of them. If you promise one run and fail to complete it, it's a disappointment, but eh, it was a high-risk mission anyway. If you promise a hundred runs and fail to complete even one, you'll catch a ton of flak. If you promise one and complete a hundred, you're a hero. The math checks out that way, especially when one single mission is good enough to get the funding anyway. Everything else is literally a bonus, why promise ten when they're paying you for one?
    - sfifs 5y ago
      Yes - there's clearly an incentive misalignment in setting & communicating realistic mission risks & expectations which is not too surprising given the politics behind how NASA is funded. What happens over a period of time in any organization if you consistently under-commit & over-deliver is that people stop believing in your "official" under-committed number ("It's just on-paper") and start creating their own informal "realistic" expectations (which are possibly different across stake-holders) and start taking (possibly mis-aligned) decisions based on these. As you may imagine, such situations tend to blow up from time to time.
ForHackernews 5y ago
If you're smart, that's how your private organization sets its goals, too. Under-promise and over-delver keeps stakeholders happy.
chha 5y ago
Isn't this the exact same everywhere else? Your boss wants to know how long you'll spend on a job; if you tell him 30 days and spend 30 days you're just good at estimation. If you do it in 20, you're a star for doing it under budget!

TaylorAlexander 5y ago

I mean the low bar is that they fly three robots to Mars and two of them land safely and one of them makes multiple flights on the surface of another world so it’s a pretty high low bar!

sfifs 5y ago
I mean sure it's cool but I think the point is they've been so consistently spectacularly successful in blowing past mission objectives that it's a bit hard to imagine achieving this wasn't the "expected result" (maybe we can cut a bit of slack for Pathfinder/Sojourner which was the first complex mission after the Climate Orbiter debacle and over-delivered only 12x).
- StavrosK 5y ago
  The issue here is that there's a mismatch between the probability distribution and result. In reality, either the equipment will fail some time on entry/initialization and won't be able to complete any missions, or it will work for one and thus probably work for a hundred of them. If you promise one run and fail to complete it, it's a disappointment, but eh, it was a high-risk mission anyway. If you promise a hundred runs and fail to complete even one, you'll catch a ton of flak. If you promise one and complete a hundred, you're a hero. The math checks out that way, especially when one single mission is good enough to get the funding anyway. Everything else is literally a bonus, why promise ten when they're paying you for one?
  - sfifs 5y ago
    Yes - there's clearly an incentive misalignment in setting & communicating realistic mission risks & expectations which is not too surprising given the politics behind how NASA is funded. What happens over a period of time in any organization if you consistently under-commit & over-deliver is that people stop believing in your "official" under-committed number ("It's just on-paper") and start creating their own informal "realistic" expectations (which are possibly different across stake-holders) and start taking (possibly mis-aligned) decisions based on these. As you may imagine, such situations tend to blow up from time to time.

ForHackernews 5y ago

If you're smart, that's how your private organization sets its goals, too. Under-promise and over-delver keeps stakeholders happy.

chha 5y ago

Isn't this the exact same everywhere else? Your boss wants to know how long you'll spend on a job; if you tell him 30 days and spend 30 days you're just good at estimation. If you do it in 20, you're a star for doing it under budget!