3 ms·

"the planning tasks (e.g. writing emails, having meetings) have grown, though perhaps not entirely linearly." Or is it because of the cult of Process and Burea

by fsloth 10y ago

"the planning tasks (e.g. writing emails, having meetings) have grown, though perhaps not entirely linearly." Or is it because of the cult of Process and Bureaucracy? Skunk Works is/was one of the badassest engineering units in world history. They did projects that made sense, they had high individual responsibility, very flat organization , and encouraged communication between disciplines above all. The book "Skunk works" by it's former director Ben Rich contains as an epilogue a lamentation to the new way of doing things by lengthy meandering process and bureaucracy with an inflated middle management - instead of just doing things right. The book is the best book on the subject on how to do great technical things - but not by way of theory, but by of example. And I can tell you, it does not contain an inch of Scrum, six sigma, RUP or any of that. They had a simple high level process, and then they drilled down on specific processes on per project basis, as needed.

Bartweiss 10y ago
> The book is the best book on the subject on how to do great technical things - but not by way of theory, but by of example. The most pressing question is how well these techniques can be exported. Joel on Software gave us a great take on this quite a few years ago: https://www.joelonsoftware.com/2001/01/18/big-macs-vs-the-naked-chef/ https://www.joelonsoftware.com/2001/01/18/big-macs-vs-the-na... Basically, the best 'process' for a small team of smart, skilled people is for someone to hand them a blank check and a vague goal, and turn them loose for a few months. This is basically what Skunk Works did, and the results were exceptional. With a bigger team, or a less brilliant team, or a team less skilled at execution, blank checks and no guidance is far more likely to produce overspending and wasted time. Six Sigma doesn't have a hope of improving efficiency over Skunk Works, but it was at least intended to provide a minimum efficiency better than "screwing up a lot". The best cases are certainly hobbled by Process, but most of the world is too big and dysfunctional to match the best case. (None of which is to say that Skunk Works style projects are bad - I still think we need more of them. In particular, pseudo-startups without the time and fundraising pressure can be uniquely effective.)
- fsloth 10y ago
  Skunk works did not have blank cheques nor did they have vague goals. One reason to their success were explicit vehicle operational targets for each project. But skunk works was not only a small crack team of engineers, although this is what they are famous for. They built and operated several vehicles, with 'normal' personnel.
ArkyBeagle 10y ago
I've been on the local version Skunk Works team most places I've worked ( not THE Skunk Works, but teams modelled on it ). Eventually you find that a basket of crabs needs no lid. You can hear the click when tolerance for that sort of behavior ends, it's time to fresh up the resume and go. It's a really loud click if you are listening.
kogepathic 10y ago
> Skunk Works is/was one of the badassest engineering units in world history. Yes, Skunk Works is amazing. But I don't think it's fair to view their accomplishments in isolation. Skunk Works didn't have to worry about where to get money from, Lockheed provided Kelly Johnson with money and trusted him to spend in appropriately. In a normal company you have a lot of people dedicated to sales, marketing, and finance. The first two didn't apply at all to Skunk Works, and while I imagine they did have to plan and budget, I don't know to what extent they had to do controlling/taxes, etc. > The book "Skunk works" by it's former director Ben Rich contains as an epilogue a lamentation to the new way of doing things by lengthy meandering process and bureaucracy with an inflated middle management - instead of just doing things right. Yes, and is it not the least bit ironic that Skunk Works was within a company which got into its position through lengthy meandering bureaucratic processes which awarded them contracts for defense programs? > And I can tell you, it does not contain an inch of Scrum, six sigma, RUP or any of that. Please don't interpret my comment as saying "and I think the status quo in business is just fine." It's not. The current business processes we have created reward bureaucracy and paper pushing, at the expense of innovation. This being said, I also don't think the correct solution is to have engineers running everything. It would certainly help people in management to have relevant technical experience, but I studied around and work with engineers. The average engineer would fucking hate things like controlling finances, but it's necessary for the business to comply with tax regulations, etc.
- ArkyBeagle 10y ago
  China is run by engineers. So was Al Queda. We'll see how that experiment plays out.
  - RugnirViking 10y ago
    fascinating. Do you have a source? I would like to read further on such a statement
- fsloth 10y ago
  Of finances - Kellys was proud to occasionally return money to the military when they finished complex technical projects under budget. Sure, they had the financial backing of their giant mothership Lockheed but they did not spend frivolously and delivered. Probably because of their capability to deliver they could do things their way. I was not targeting your comment when I erupted about process frameworks - sorry about that. It was an outburst against the percieved "best practices". Like Cato's "By the way, Carthago needs to be destroyed". I agree - orgs need finances just like cars have instrumentation, but they should not be running all things unless that happens to be the orgs core business.

by fsloth 10y ago

Bartweiss 10y ago

> The book is the best book on the subject on how to do great technical things - but not by way of theory, but by of example. The most pressing question is how well these techniques can be exported. Joel on Software gave us a great take on this quite a few years ago: https://www.joelonsoftware.com/2001/01/18/big-macs-vs-the-naked-chef/ https://www.joelonsoftware.com/2001/01/18/big-macs-vs-the-na... Basically, the best 'process' for a small team of smart, skilled people is for someone to hand them a blank check and a vague goal, and turn them loose for a few months. This is basically what Skunk Works did, and the results were exceptional. With a bigger team, or a less brilliant team, or a team less skilled at execution, blank checks and no guidance is far more likely to produce overspending and wasted time. Six Sigma doesn't have a hope of improving efficiency over Skunk Works, but it was at least intended to provide a minimum efficiency better than "screwing up a lot". The best cases are certainly hobbled by Process, but most of the world is too big and dysfunctional to match the best case. (None of which is to say that Skunk Works style projects are bad - I still think we need more of them. In particular, pseudo-startups without the time and fundraising pressure can be uniquely effective.)

fsloth 10y ago
Skunk works did not have blank cheques nor did they have vague goals. One reason to their success were explicit vehicle operational targets for each project. But skunk works was not only a small crack team of engineers, although this is what they are famous for. They built and operated several vehicles, with 'normal' personnel.

ArkyBeagle 10y ago

I've been on the local version Skunk Works team most places I've worked ( not THE Skunk Works, but teams modelled on it ). Eventually you find that a basket of crabs needs no lid. You can hear the click when tolerance for that sort of behavior ends, it's time to fresh up the resume and go. It's a really loud click if you are listening.

kogepathic 10y ago

> Skunk Works is/was one of the badassest engineering units in world history. Yes, Skunk Works is amazing. But I don't think it's fair to view their accomplishments in isolation. Skunk Works didn't have to worry about where to get money from, Lockheed provided Kelly Johnson with money and trusted him to spend in appropriately. In a normal company you have a lot of people dedicated to sales, marketing, and finance. The first two didn't apply at all to Skunk Works, and while I imagine they did have to plan and budget, I don't know to what extent they had to do controlling/taxes, etc. > The book "Skunk works" by it's former director Ben Rich contains as an epilogue a lamentation to the new way of doing things by lengthy meandering process and bureaucracy with an inflated middle management - instead of just doing things right. Yes, and is it not the least bit ironic that Skunk Works was within a company which got into its position through lengthy meandering bureaucratic processes which awarded them contracts for defense programs? > And I can tell you, it does not contain an inch of Scrum, six sigma, RUP or any of that. Please don't interpret my comment as saying "and I think the status quo in business is just fine." It's not. The current business processes we have created reward bureaucracy and paper pushing, at the expense of innovation. This being said, I also don't think the correct solution is to have engineers running everything. It would certainly help people in management to have relevant technical experience, but I studied around and work with engineers. The average engineer would fucking hate things like controlling finances, but it's necessary for the business to comply with tax regulations, etc.

ArkyBeagle 10y ago
China is run by engineers. So was Al Queda. We'll see how that experiment plays out.
- RugnirViking 10y ago
  fascinating. Do you have a source? I would like to read further on such a statement
fsloth 10y ago
Of finances - Kellys was proud to occasionally return money to the military when they finished complex technical projects under budget. Sure, they had the financial backing of their giant mothership Lockheed but they did not spend frivolously and delivered. Probably because of their capability to deliver they could do things their way. I was not targeting your comment when I erupted about process frameworks - sorry about that. It was an outburst against the percieved "best practices". Like Cato's "By the way, Carthago needs to be destroyed". I agree - orgs need finances just like cars have instrumentation, but they should not be running all things unless that happens to be the orgs core business.