5 ms·

With open source projects there aren't any managers, and yet good quality software gets produced. I'm interested to know people's thoughts on that? Perhaps in o

by tlocke 4y ago

With open source projects there aren't any managers, and yet good quality software gets produced. I'm interested to know people's thoughts on that? Perhaps in open source projects the 'maintainer' roles are analogous to being a manager? On the other hand I've been a manager of a software team in a company, and I'm also a maintainer of a couple of open source projects, and I'd say that the two roles are very different. Does it mean that we don't actually need managers? Really I don't don't know the answer and I'm raising the question to see what other people think.

rc00 4y ago
I would start by suggesting that the incentives and motivations aren't aligned but there's more to it than that. With open source projects, most commonly, everyone wants to be there and has some level of the same objectives, possibly because they are consuming the final product as well as contributing to it. Here, the open source maintainer is a sort of shepherd for the flock. As a software team in a company, not only is everyone not equally motivated, the incentives are also not the same. A company has a financial incentive that may not align with the technical one. A company also has the pressure of deadlines where an open source project can be delayed or even in a state of maintenance mode. This drives a different culture as well. The purpose of the manager is to deliver value for the company while the role of the open source maintainer is to deliver value to the project. I would simply posit that the two roles are not the same. There are maybe some overlaps from the standpoint of overseeing finished code and the delivery of releases, but the hierarchy, org structure, motivations, and incentives create very different situations.
- cma 4y ago
  Is there any example of a long-lived successful open source project from a non-technical manager that leads a team of people on board with his or her vision?
  - rc00 4y ago
    It depends on your definitions of long-lived, successful, and non-technical but Canonical/Ubuntu and Mark Shuttleworth seem like an apt example.
    - cma 4y ago
      He received an undergrad degree in information systems.
      - rc00 4y ago
        Bachelor of Business Science degree in Finance and Information Systems[1]. Depending on the curriculum and institution, Information Systems is not a technical degree in the same sense as Computer Science. That is why I quantified with "it depends on your definitions." I'm not sure if Bob Young (founder of Red Hat) meets the definition either since he is no longer involved these days. There's a bit of a theme developing in my mind clearly as these are all the enterprise and open source projects. I admittedly don't know much about the founder(s) of SUSE. (In my mind, I might be ascribing some notion of success to being a profitable enterprise. You may not feel the same.) Nonetheless, those are examples to the hypothetical and I have not researched any others though I am sure they exist. I think one important thing to keep in mind is the type of person/personality these projects tend to attract in the first place. That is likely why there are a predominant amount that are technical. This detracts from the original point though. [1] https://en.wikipedia.org/wiki/Mark_Shuttleworth#Early_life https://en.wikipedia.org/wiki/Mark_Shuttleworth#Early_life
dasil003 4y ago
Engineering managers exist to ensure the company is getting sufficient value out of the hundreds of thousands they’re paying for each software engineer. If goals were always crystal clear and every SWE always had a good sense of how their work translated into business outcomes and prioritized accordingly then no, managers would not be necessary. Those conditions are often true at very early stage startups, and never true at any company over 100 employees.

by tlocke 4y ago

rc00 4y ago

I would start by suggesting that the incentives and motivations aren't aligned but there's more to it than that. With open source projects, most commonly, everyone wants to be there and has some level of the same objectives, possibly because they are consuming the final product as well as contributing to it. Here, the open source maintainer is a sort of shepherd for the flock. As a software team in a company, not only is everyone not equally motivated, the incentives are also not the same. A company has a financial incentive that may not align with the technical one. A company also has the pressure of deadlines where an open source project can be delayed or even in a state of maintenance mode. This drives a different culture as well. The purpose of the manager is to deliver value for the company while the role of the open source maintainer is to deliver value to the project. I would simply posit that the two roles are not the same. There are maybe some overlaps from the standpoint of overseeing finished code and the delivery of releases, but the hierarchy, org structure, motivations, and incentives create very different situations.

cma 4y ago
Is there any example of a long-lived successful open source project from a non-technical manager that leads a team of people on board with his or her vision?
- rc00 4y ago
  It depends on your definitions of long-lived, successful, and non-technical but Canonical/Ubuntu and Mark Shuttleworth seem like an apt example.
  - cma 4y ago
    He received an undergrad degree in information systems.
    - rc00 4y ago
      Bachelor of Business Science degree in Finance and Information Systems[1]. Depending on the curriculum and institution, Information Systems is not a technical degree in the same sense as Computer Science. That is why I quantified with "it depends on your definitions." I'm not sure if Bob Young (founder of Red Hat) meets the definition either since he is no longer involved these days. There's a bit of a theme developing in my mind clearly as these are all the enterprise and open source projects. I admittedly don't know much about the founder(s) of SUSE. (In my mind, I might be ascribing some notion of success to being a profitable enterprise. You may not feel the same.) Nonetheless, those are examples to the hypothetical and I have not researched any others though I am sure they exist. I think one important thing to keep in mind is the type of person/personality these projects tend to attract in the first place. That is likely why there are a predominant amount that are technical. This detracts from the original point though. [1] https://en.wikipedia.org/wiki/Mark_Shuttleworth#Early_life https://en.wikipedia.org/wiki/Mark_Shuttleworth#Early_life

dasil003 4y ago

Engineering managers exist to ensure the company is getting sufficient value out of the hundreds of thousands they’re paying for each software engineer. If goals were always crystal clear and every SWE always had a good sense of how their work translated into business outcomes and prioritized accordingly then no, managers would not be necessary. Those conditions are often true at very early stage startups, and never true at any company over 100 employees.