3 ms·

I really wish my experience were that cut and dry. This strategy seems to me to depend super strongly on the quality of the developers/engineers, and their comm

I really wish my experience were that cut and dry. This strategy seems to me to depend super strongly on the quality of the developers/engineers, and their commitment to getting work done (c.f. your examples: Skunk Works, Bell Labs, PARC, etc -- I really wish I worked in such an environment!). There are people who are good at seeming busy, but who wildly overestimate, take every opportunity to get bogged down on some trivial obstacle, chronically misunderstand objectives/values, can't consistently put three words in a row correctly, and just seem in general to be the weak link; those people need to be managed.

whiddershins 7y ago
Do they?
- sokoloff 7y ago
  Managed within or managed out, but I think yes.
waiquoo 7y ago
It's also really interesting the effect of management to drive timelines and innovation. If you look at the Apollo program, there was a ton of innovation driven by a incredibly fast design/launch cycle. Well managed timelines and goals can have a focusing effect (point the innovators in the most effective direction and give them the tools). The issue is when meeting the timelines and goals BECOMES the measure, rather than measuring effect. At that point it's just teaching the test and micromanaging.
- imglorp 7y ago
  That's an extreme and useful counterexample. Did anyone distill Apollo project management lessons learned into something we can apply now?
  - waiquoo 7y ago
    Of course. Here are couple interesting papers by NASA leadership from the Apollo era (PDF warning): https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19720005243.pdf https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/197200... https://www.nasa.gov/pdf/293253main_62682main_jerry_madden_forum7.pdf https://www.nasa.gov/pdf/293253main_62682main_jerry_madden_f...
ghaff 7y ago
Yes. I think you're seeing something of a survivor bias here. For every one of an example of a true and successful skunkworks project--and this one is obviously much different in scope than skunkworks aviation projects--there are tons of counterexamples. A lot of times there's a group of engineers who don't agree with whatever decision and go off and do what they want to do instead. I'm guessing when that happens, most of the time it never amounts to anything and never sees the light of day.

whiddershins 7y ago

Do they?

sokoloff 7y ago
Managed within or managed out, but I think yes.

waiquoo 7y ago

It's also really interesting the effect of management to drive timelines and innovation. If you look at the Apollo program, there was a ton of innovation driven by a incredibly fast design/launch cycle. Well managed timelines and goals can have a focusing effect (point the innovators in the most effective direction and give them the tools). The issue is when meeting the timelines and goals BECOMES the measure, rather than measuring effect. At that point it's just teaching the test and micromanaging.

imglorp 7y ago
That's an extreme and useful counterexample. Did anyone distill Apollo project management lessons learned into something we can apply now?
- waiquoo 7y ago
  Of course. Here are couple interesting papers by NASA leadership from the Apollo era (PDF warning): https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19720005243.pdf https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/197200... https://www.nasa.gov/pdf/293253main_62682main_jerry_madden_forum7.pdf https://www.nasa.gov/pdf/293253main_62682main_jerry_madden_f...

ghaff 7y ago

Yes. I think you're seeing something of a survivor bias here. For every one of an example of a true and successful skunkworks project--and this one is obviously much different in scope than skunkworks aviation projects--there are tons of counterexamples. A lot of times there's a group of engineers who don't agree with whatever decision and go off and do what they want to do instead. I'm guessing when that happens, most of the time it never amounts to anything and never sees the light of day.