4 ms·

You make a really strong point. I think the problem you described above is why, in truth, closed allocation tech companies are destined to fail. Closed/open all

You make a really strong point. I think the problem you described above is why, in truth, closed allocation tech companies are destined to fail. Closed/open allocation is a continuum in reality, but the more closed a technology company is, the quicker it rots. The "Real Googler Line" concept (and, as you know, I was only at Google for half a year and have no insight into what the place is like now) is the recognition that, above a certain level of recognized merit, people should be trusted to allocate their time and energies. As you said, people just below the trust thermocline where it goes from open to closed allocation (on whom the company relies to perform the most thankless, ugliest tasks) get resentful. You don't have a solution until you lower that trust thermocline down to the people who aren't capable of doing useful work, who you don't want to retain anyway. The only conclusion I can reach is that a closed-allocation tech company doesn't work. I realize that open allocation isn't right for all industries, but in software, it seems to have minimal downside, incredible upside, and no reasonable alternative. So, I agree with you. One shouldn't fix a tech company by elevating a couple top performers. (They'll get lonely.) One should go to full open allocation, and then those political issues go away.

nostrademons 13y ago
I don't think open allocation works either once you get past Dunbar's Number. Instead of making politics go away, it makes it worse. All the political considerations that are out in the open when you're managing a formal org structure become hidden in the backroom deals of who wants to work with who else and how they want to go about doing it. Here's another observation for you: there is a system where there is no org chart and nobody telling anybody else what to work on. That system is the startup ecosystem. And the end result of that system is what you term "VC-istan": some players become more powerful than others, and they freely choose who they want to associate with based on reputation, or how well their personalities mesh, or pattern-matching on previous successes, or who they went to college with. The only "solution" I could come up with is that there is no solution. If you give people agency and freedom of association, you will eventually run into situations where their freedom to act impinges on someone else's freedom to act, and it is impossible to let both of them work on what they wish without forcing one or the other to make tradeoffs. Or, alternatively, they could both choose to work alone, in which case it is impossible to build anything larger than what one person can do by himself. Once you see that tradeoff, it's possible to make choices that stake out various parts of that space, eg. another thread currently on HN [1] describes how your options are management, niche development, or consulting as you get older. [1] https://news.ycombinator.com/item?id=7373301 https://news.ycombinator.com/item?id=7373301
- thecage411 13y ago
  Can you give an examples political considerations that are out in the open in a formal org compared to open allocation? I think you could make a case for the opposite.
  - nostrademons 13y ago
    Example 1: You have a group of people who care deeply about performance, reliability, maintainability, and long-term health of the code base. You have another group of people who care deeply about user experience, development velocity, and pushing features to market quickly. In an open allocation system these two mindsets will self-organize into different teams, and will avoid working on each others' projects. The former group will find that the latter are cowboys who throw together messes that they then have to clean up. The latter will find that the former are slow killjoys who result in mediocre products delivered late. The latter group will get to market first, which makes the former believe they're marginalized and gets most of them to quit the company. When the product inevitably starts creaking at the seams, there will be nobody around to fix it and introduce robust development processes. In a formal org, these would likely be two job functions (call them "SRE" and "SWE", or "Infrastructure" and "Features", say) who are forced by upper management to work on the same codebase. Disagreements between them become explicit negotiations where all the technical issues and trade-offs are laid on the table, and then people can come to an agreement after looking at all the considerations. Example 2: You have a brilliant developer who is terrible at working with other people and doesn't particularly want to deal with them. In an open-allocation world, he will end up shunned; nobody will want to work with him. In a managed org, it's often possible for his manager to pick off a certain really difficult technical challenge, have him go off and find a solution, and then interface with the rest of the group via some technical API. Example 3: You have a technical lead who can't let go of the technical aspects, and insists on taking all the most interesting work for himself. In an open-allocation world, nobody will want to work with him, which means that he won't be a tech lead for very long, and eventually he'll probably get frustrated and leave. A manager could gently explain to him that taking all the most interesting work is incompatible with being a team lead, and have him choose one or the other. Example 4: You have a project where consistency is the whole benefit of the project. For example, you're doing a company-wide visual redesign to make the company's product offerings more consistent for users, or you're introducing a language styleguide, or you're defining standard APIs for communication between components. These efforts typically never succeed unless they have top-down support from a strong leader. Everybody sees the need to agree on a standard, but when it comes to deciding what that standard is, everybody has their own agenda and nobody has authority to decide in favor of one alternative or the other. Standards committees are good public examples of this: officially, it's "open allocation" where every company and member of the public who wants a voice has one, but unofficially they're dominated by the actions of people who can afford to implement and bring to market a concrete solution.

nostrademons 13y ago

I don't think open allocation works either once you get past Dunbar's Number. Instead of making politics go away, it makes it worse. All the political considerations that are out in the open when you're managing a formal org structure become hidden in the backroom deals of who wants to work with who else and how they want to go about doing it. Here's another observation for you: there is a system where there is no org chart and nobody telling anybody else what to work on. That system is the startup ecosystem. And the end result of that system is what you term "VC-istan": some players become more powerful than others, and they freely choose who they want to associate with based on reputation, or how well their personalities mesh, or pattern-matching on previous successes, or who they went to college with. The only "solution" I could come up with is that there is no solution. If you give people agency and freedom of association, you will eventually run into situations where their freedom to act impinges on someone else's freedom to act, and it is impossible to let both of them work on what they wish without forcing one or the other to make tradeoffs. Or, alternatively, they could both choose to work alone, in which case it is impossible to build anything larger than what one person can do by himself. Once you see that tradeoff, it's possible to make choices that stake out various parts of that space, eg. another thread currently on HN [1] describes how your options are management, niche development, or consulting as you get older. [1] https://news.ycombinator.com/item?id=7373301 https://news.ycombinator.com/item?id=7373301

thecage411 13y ago
Can you give an examples political considerations that are out in the open in a formal org compared to open allocation? I think you could make a case for the opposite.
- nostrademons 13y ago
  Example 1: You have a group of people who care deeply about performance, reliability, maintainability, and long-term health of the code base. You have another group of people who care deeply about user experience, development velocity, and pushing features to market quickly. In an open allocation system these two mindsets will self-organize into different teams, and will avoid working on each others' projects. The former group will find that the latter are cowboys who throw together messes that they then have to clean up. The latter will find that the former are slow killjoys who result in mediocre products delivered late. The latter group will get to market first, which makes the former believe they're marginalized and gets most of them to quit the company. When the product inevitably starts creaking at the seams, there will be nobody around to fix it and introduce robust development processes. In a formal org, these would likely be two job functions (call them "SRE" and "SWE", or "Infrastructure" and "Features", say) who are forced by upper management to work on the same codebase. Disagreements between them become explicit negotiations where all the technical issues and trade-offs are laid on the table, and then people can come to an agreement after looking at all the considerations. Example 2: You have a brilliant developer who is terrible at working with other people and doesn't particularly want to deal with them. In an open-allocation world, he will end up shunned; nobody will want to work with him. In a managed org, it's often possible for his manager to pick off a certain really difficult technical challenge, have him go off and find a solution, and then interface with the rest of the group via some technical API. Example 3: You have a technical lead who can't let go of the technical aspects, and insists on taking all the most interesting work for himself. In an open-allocation world, nobody will want to work with him, which means that he won't be a tech lead for very long, and eventually he'll probably get frustrated and leave. A manager could gently explain to him that taking all the most interesting work is incompatible with being a team lead, and have him choose one or the other. Example 4: You have a project where consistency is the whole benefit of the project. For example, you're doing a company-wide visual redesign to make the company's product offerings more consistent for users, or you're introducing a language styleguide, or you're defining standard APIs for communication between components. These efforts typically never succeed unless they have top-down support from a strong leader. Everybody sees the need to agree on a standard, but when it comes to deciding what that standard is, everybody has their own agenda and nobody has authority to decide in favor of one alternative or the other. Standards committees are good public examples of this: officially, it's "open allocation" where every company and member of the public who wants a voice has one, but unofficially they're dominated by the actions of people who can afford to implement and bring to market a concrete solution.