[wonderingdev]

And that's a great explanation of the theory behind that :) and it delights me, every time. But the question, alas, remains unanswered: "would everything work well anyway without such a framework?"

Basically I'm curious about this: we're all applying this framework, taking it for granted. It was there, we accepted it. But I wonder: what if it wasn't there?

I'm building a machine, I know the minimum requirements to make it work. I build it like that and then I think: should I also add to it this other cog as well? Would it improve _somehow_ the machine? Well, the other folks are doing it, heck, I'll do it as well!

Now you have a heavier, more complex, machine, with one more cog, and you don't even know if you really need it.

Sorry for the brutally simple example, but I'm still missing an evidence we really need these frameworks.

[josephmosby]

If you accept these postulates as true: * on a long enough time horizon, companies must either adapt to their environment by changing their product, or they must cease to exist. * a changing product means changing the abilities required to deliver that product at some level * changing the abilities required to deliver that product means the humans must change as well

then you have to conclude: On a long enough time horizon, the humans must change their own abilities to match the changing requirements of the product, or the company will go bankrupt.

Now, given that conclusion: you are correct, goal-setting is not required. And we have evidence that it is not required in companies that mostly outsource technical labor on an as-needed basis rather than set goals to continually realign employees with company objectives.

In some way, you have to choose an approach for calibrating changing employees and changing company objectives. Goal-setting is a low friction approach for calibration when the company prefers employees with internal domain knowledge over time. There are plenty of other approaches for aligning needs and available labor as well, but they are generally not preferred by both companies and individuals which is why you see the prevalence of this approach.

[wonderingdev]

On such a long term horizon, you should consider staff turnover, which is never zero. So, no, what you're inferring from those "postulates" is incorrect.

I worked for a big dotcom company in the past and

1) there was a cap on headcount (as it's usually the case), so a lot gets outsourced

2) they don't care about internal domain knowledge: everybody should be replaceable at any time, ramp up costs are not even taken into account. Companies that wants employees to set goals usually are big and usually can afford to lose people with "internal domain knowledge". Even if, of course, they'll try not to do that, but it's still not that big of a deal.

While I really like philosophy (really, I do!), I'm a very practical guy and I'm still missing the evidence that I'm looking for. And I'm pretty sure it won't come in the form of a syllogism :)