I have a few objections, both to the aim of the proposal and the specification proposed:
This isn’t in alignment with the plan of action that found wide support at the engineering workstream retreat – namely, that a default working configuration should be specified which is to be used in the absence of, but superseded by an individual workstream charter. This allows every workstream the option to define a set of terms that is appropriate for its configuration, while not requiring any workstream to do so explicitly in the event that the default is appropriate. I do not think that the specification outlined here is suitable for all workstream configurations, and there is no language in the proposal that allows this definition to be superseded by an individual workstream charter. In the context of the engineering workstream, for example, it is the majority opinion among the workstream participants that this proposal is entirely unsuitable. In contrast, I think that this proposal would be suitable for a workstream configuration wherein a leader hired and delegated work to an external agency as in the example given by Erik in his recent blog post.
To ’ point – I agree that this places entirely too much authority and responsibility in the hands of the workstream leader. The intention here is to maximize value to the DAO, and doing so requires that the DAO’s best interests are prioritized. Those are likely not to be in alignment with the best interest of any particular workstream leader, and there are insufficient checks and balances to ensure that the DAO’s best interest is prioritized in all decisions in this model. While the risk of impropriety might be low given the current DAO roster, we have to build robust governance, and that means striving to put in place policy that would maintain the integrity of the DAO should it someday in the future be infiltrated by bad actors.
I have some issues with many of the points of view expressed in the specification section; I’ll address them in order:
Single Workstream Leader - I don’t see the motivation for this as part of the definition. The flexibility afforded to workstreams by having multiple leaders can be highly advantageous to productivity, and we should allow workstreams to arrange themselves to optimize productivity so long as it is capital-efficient to do so. This is evaluated by the community at the time each workstream proposal is put before governance.
Resourcing – A primary consideration of each workstream should be to maximize productivity – optimizing productivity on a team requires maintaining a careful balance between competence and culture. Happy teams are properly motivated, and motivated teams are productive. All other things being equal, individuals tend to be happier working on teams where they enjoy spending time with their coworkers. Properly assessing culture fit requires buy-in from the rest of the team. Centralized ShapeShift was very good about this as part of the hiring process, and for that reason in part, the quality of the working environment was extremely high. Delegating hiring responsibility and authority to any singular individual is highly unlikely to result in an optimal team configuration. If the intent is to optimize a working team configuration, the only sensible strategy is to evaluate candidates as a team. Likewise, when an individual needs to be fired, it has been my experience that the motivation for doing so is abundantly clear to the team. Requiring buy-in from the team should not slow down necessary firing decisions. Making these decisions democratically also avoids potential resourcing pitfalls common to centralized companies such as those where a middle manager takes personal issue with or has an incorrect perception about a particular employee. Finally, as said above, in order to attract and retain talent, it will be necessary for us to provide workstream contributors with some degree of security around their employment arrangement with the DAO. All other things being equal, why would any talented contributor with multiple career opportunities pursue an employment arrangement that could be terminated at the discretion of a single individual for any or no reason at all when other options are available? At the very least, some document approximating an employee handbook that outlines how performance disputes are handled and the terms under which an employment agreement can be severed should be provided. Finally, though it has been stated to be the case several times, I have yet to hear an argument as to why the ability to make unilateral resourcing decisions is necessary to perform the function of a workstream leader in all cases. It seems instead that this is more about wielding authority or establishing the perception of authority, and as far as I’m able to see, that carries a high risk for little reward given the inadequate system of checks and balances in this model.
Budget – Not to hold workstreams accountable for the details of budgetary agreements made with the DAO makes no sense to me whatsoever. I have always seen the budget approved for the workstreams as amendable contracts with the DAO, and never held the idea that the line items were fungible. If, as a workstream leader, I had a budget approved that allocated $50,000 monthly, $10,000 of which I took as salary and the remainder was to be issued as bounties, then decided after community approval that I would instead pay myself $40,000 monthly and use only the remaining $10,000 for bounty work, are you of the opinion that this is acceptable? You can make the argument that this would be discovered quickly, but there are many ways to hide how the funds are spent. I’m happy to be proven wrong here, but I would very much like to poll community sentiment on this hypothetical example.
Maintaining the requirements that workstream leaders are held accountable to budgetary agreements with the DAO and amendments must pass governance will require workstream leaders to spend more time carefully considering budgets up-front, which is exactly what we want. This does not have to be overly restrictive. To account for unanticipated expenditures of reasonable size, it has become a common practice for contingency funds to be added as a line item to each budget. Adopting a practice in the DAO where the approved budgets carry no meaning other than in sum total seems exceedingly foolish, and this would be the first place I’ve ever heard of such a policy in practice.
Benefits – I think there may be a fundamental difference in perception about the role and expectations of a workstream leader. I’ve always taken issue with the “leadership” designation because I don’t believe it to be quite accurate. I have always considered the role of a workstream “leader” to be exactly as mentioned above – a combination of secretary, coordinator, and spokesperson. As one of the KeepKey workstream leaders, this is what I consider my position to be. I and many others do not and should not look to the workstream “leader” for literal leadership and direction, and so I have always thought something like “coordinator” or “representative” to be a more appropriate designation. While singular leadership in the literal sense may be appropriate some workstreams, in others it makes no sense at all and is much more likely to be harmful than helpful in my opinion.
Drawbacks – I do agree; there is a strong perceived centralization risk that arises from delegating authority to workstream leaders. These concerns are not unfounded as this has indeed proven to be problematic in recent times as the perspective of what the role of a workstream leader is has drifted in the minds of some such that two distinct schools of thought on the matter have arisen. I do not think that external governance is a sufficient check to the unilateral authority that this proposal would grant to workstream leaders, and for one reason – context. As a contributor to the engineering workstream, I have no context at all about what is going on in the marketing or operations workstreams, for example. Unless a proposal to remove the leaders of either of those workstreams made a verifiable accusation of some egregious action, I would be unlikely to vote on the proposal because I would not feel sufficiently informed to give a meaningful opinion on the issue. This proposal is a great example of the problem I’m identifying here. Out of context, what has been written above sounds fairly reasonable, but the context of the dynamic on the engineering workstream gives color to this proposal that isn’t externally visible. While the specifications above might seem acceptable to community members without the relevant context, it is currently the majority opinion of the engineering workstream participants that this is a very bad idea.
0xdef1cafe:
theoboldfrazier:
elmutt: