I’ve written up a proposed process for dealing with RFPs in this docs pull request. If there are no strong objections I would like to merge it and try the process out with some initial RFPs for stuff we’ve already identified that needs doing.
3 Likes
I like the concept and specially having always a clear ownership model of proposals.
I do think however that not all RFPs should be DAO wide. Funding a swag store should be coming from the ecosystem group, and only if that is not happening should there be a wide snapshot vote to tell them to do it.
3 Likes
DAO Docs
Tasks that needs to be done are published in, for example Clarity, like we do now, except a base and maximum reward is attached to it. The price will gradually increase from the base reward to the max reward until someone accepts the task and starts working on it. Then that task will be “off the market”.
The task has a set deadline by which it needs to be changed to “In Review” status by which point stewards would look over the quality of work and if satisfactory mark it as finished and pay out the reward, or send it back for further edits if there are smaller things that needs to be corrected.
In the case of work being done very poorly or not at all, stewards could change the ownership of the task to put it back on the market and/or assign someone else.
–
Does this explain it more clearly? It’s not intended to encompass things like running a swag store in and of itself, but rather to act as an incentive to get work done through tasks.
By open/closed to the public I’m referring to how anyone on the internet can’t just log into the task platform and start accepting tasks. There are gating mechanisms (albeit not intended as such) in that you need to request access to the DAO, speak to someone to get added to the workgroup etc.
While anyone can do that, it eliminates most of the issues with “open to the public bounties” in practice.
I’d like to add that in my experience working in workgroups, the main issue isn’t tooling or communication, but over-managing and far too much bureaucracy for the size of the workgroups. It seems as if the top is designing systems to facilitate thousands of workers for global cooperation, while active workers are <10
The idea that multiple people would submit proposals to do the same thing for ENS to then sit back and cherry pick the best ones is pretty far removed from the reality of the DAO right now. We don’t have enough active workers for that.
I really think that the best thing to do, in order to make the DAO conducive to work being done, is to stop managing it from the top and instead let it grow naturally from the ground up via the workers. I fully expect that different workgroups will need different models, and the best way to figure that out is to just start the work with as little red tape as possible and then see what works and what doesn’t work.
Just my 1c.
Right, this is handled in the process docs - stewards can choose to pay for an RFP directly out of WG funds.
The problem with this is that it pretty much enshrines hiring the cheapest bidder. We could spend a lot of time waiting for people who are incapable of doing a job to fail, so that we can put the job back on the market and try again.
That’s an intentional feature in order to help with sustainability and avoid careless spending of public funds. It’s generally viewed as desirable when dealing with public funds in non-profits and most countries even employ laws like that for government contracts.
When that isn’t used you inevitably run into issues with nepotism and corruption.
The one worry I can see is that the lowest bidder would produce work that isn’t good enough, which is where final approval of the work before payment by the stewards comes into play.
The idea was made to solve exactly that. I came up with it because the docs workgroup stalled for months. The way it’s intended to solve it is by putting the reward as a direct incentive to complete the task on time. If you don’t complete it you’re not paid, and since it’s intended for bite-sized tasks the due date can be set pretty aggressively (say 7 days) and each user would be limited in how many tasks they can simultaneously accept.
You’d be able to have the task fail to be completed by 12 different users before you’d reach the same level of delay that currently exists.
However, I don’t think it would be a big issue right now. The workgroups are quite small and if someone repeatedly doesn’t finish tasks they could be excluded from the workgroup. If and when workgroups grow large enough to encompass thousands of users the idea might need to be modified or changed, but we’re not there yet.
1 Like
I don’t know if governments and non-profits are a shining example of good execution 
Given how large the ENS treasury is, and how slow the DAO has been to start thus far, I’d be much more concerned with making sure we actually start spending funds than being worried about being too inefficient, at the moment (it’s easier to go more restrictive later than the other way around, in my experience). This is a problem I see a lot in DAOs generally: massive treasuries, but skittish about spending any of it.
This is why a lot of DAO tools companies still raise from VC, even though many would rather get funded by DAOs, btw. DAOs make it too hard / too long to get funded, while VCs will write you $1M literally in a day in the current environment.
If we could come up with a more predictable model, I think you’d see more groups coming in for smaller amounts, more frequently, because they have a reasonable confidence they can get the proposal through in a reasonable timeframe and get what they need, as long as they have proven execution ability.
The other issue (and I hear this a lot from really prominent groups in the space) is that often a group will make a proposal to do some work for a DAO, go through a whole proposal process, and then another group will basically fork that work and get their own proposal through instead before the first one gets approved. If this can be addressed in this RFP model @nick.eth , I think that would be really good!
Basically some sort of guarantee at some point that if the DAO wants X work done that you are proposing, that after a certain amount of time spent going through the process, they won’t just choose another identical proposal without a good reason why the first proposal wouldn’t be good for the DAO.
Not getting paid is not enough of a disincentive to not grab the work if you’re not confident you can do it, imo. If you want something like this to work, I think workers would need to stake and have that slashed if they don’t complete the work (or get an extension approved, etc.). That could put undue burden on the workers though (and lock out anyone that doesn’t have the initial capital to spare), so still think the RFP model would probably achieve better results.
That point is exactly why I’m focusing on sustainability. People have a tendency to be careless with spending while the coffers are high, only to engage in extreme cost saving measures when it’s too late and it ends up causing damage.
By trying to be sustainable while the coffers are high means that ENS doesn’t have to suffer through periods when the coffers are low. The more we save now, the more ENS has available to spend later, if necessary.
Just because one is rich doesn’t mean it makes sense to spend $1k on a newspaper. It makes sense to save for a rainy day.
@nick.eth Would you still want to reserve 50% of the treasury funds if the ENS treasury was managed sustainably?
Where those laws exist, they generally require accepting the lowest qualified bidder. And they often lead to substandard work that has to be redone at great expense. I don’t think they’re a model we should strive to imitate!
I’m happy for WGs to experiment with models like this for smaller tasks. It’s pretty much how GitCoin bounties work, and they can work very well in certain circumstances. I just don’t think it’s viable for the larger or longer-term things that RFPs can be used for.
This is a good point. One way to handle it would be to specify that bids have to be submitted privately and are only revealed at the end of the submission process (but before a decision is made).
The goal of the endowment is to ensure there are stable operating funds regardless of revenue, so yes.
1 Like
I already answered that concern in the actual post you’re omitting from your quote:
If DAO funds are managed sustainably, that implies that there are operating funds regardless of revenue. There isn’t a reason aside from this to set aside 50% of the treasury funds that you’ve explained.
It’s not actually an answer, though - we’d still be accepting the lowest bid, and just trying to fix things after the fact.
This is probably a subject to discuss in the endowment thread instead of here.
It does answer it, well.
I just did.
I’ve now merged this PR.
2 Likes
I tend to agree with @nick.eth here re it likely defaulting to cheapest bidder - which would not be ideal. Best practice in RFPs that require expertise/specialized knowledge (at least in nonprofits/governments) - tends to select - not on the cheapest/lowest price - but on "Best Value Determinations (aka “Best Value for Money”) often in conjunction with “Life-Cycle Cost Analysis” in a fair and transparent manner.
It is not, in fact, typically viewed as desirable/best practice to go with cheapest re public spending, bc in the long run on more complex requirements, going with the cheapest often means going with less experience and lower quality, or higher long-term cost - which in turn, often leads to defaults/breaches - resulting in delays, additional $ for corrections or other unanticipated expenditures, etc. The exception to this is when purchasing really basic items that do not need complicated, knowledgeable entities or individuals completing the work (i.e., purchasing paper or pens for an office, etc.)
I would anticipate that almost all RFPs in the ENS-sphere will be complex and require significant expertise/knowledge - which would in turn make me lean towards utilizing some type of best value determination (generally a combination of technical qualifications, cost, and other factors [e.g., past performance]) to decide who is best suited (not cheapest) to deliver quality work, on time. The downside of this is - it is harder to do versus simply saying - minimally qualified bidder, and lowest cost.
3 Likes
So what is a proposal if not a feature/epic story in a sense of regular product management?
If you agree it basically is the same thing, the proposal has a crucial flaw. Nobody knows how expensive the work is at the beginning. People signing up for a job are not oracles and I think it is clear from some other comments that the result of this could be either an unfinished job or poor quality.
Same with asking people to work for free to show their quality. Best people never will. There is almost 1-1 analogy in a world of design agencies - bad ones do the tenders and accept fix scope fix price, good ones don’t.
I’d propose to turn RFP to the “Bet” as described in Shape Up. In a nutshell, workstream itself will invest into the discovery process (they will need to cover the expenses, not sure how to do it in DAO terms) and present the Bet to working group stewards (or any decision making part of DAO). At this moment, the stewards would have way more information to make the decision, including the investment needed and the working group responsible for delivering the whole thing can be assembled.
In some of the situations (bigger things) this may require turning a workstream into Sub-DAO, but it still feels like a better and more scalable solution than following the fix scope fix price model.
I’m not quite sure how this would work - can you elaborate with an example?
I’ll do my best. To pick an example of a proposal close to my area of knowledge, I’ll transform the Swag Store to a community build project (not outsourced, in the original case is the RFP probably the best way to do it).
Swag Store
ENS needs someone to run a swag store, including design of the merchandise, fulfillment and shipping, promotion etc. Someone write a detailed problem statement including the value for the community. Community WG agrees to adopt it and creates a discovery group (IDK how to process this tbh, in a regular company this is a group of more experienced people with a proven track record). They will work on “shaping up” the idea, to design key elements and deliverables, ending with some PoC or any kind of low level prototype (diagram, document). In this case it would have to include some elementary research of logistics API providers, assuming that the shipping itself is not something that DAO want to do.
They may, in the end, recommend to outsource the project, and this is a great example where they perhaps should recommend this. As a result, they will present their findings to the WG allowing WG for more informed decisions. Let’s say the outsourcing will require no initial investment and there will be 25% of margin to the operator. Another option will be to invest 10 ETH to have the app built internally based on nextjs shop. I honestly don’t know what solution will be better, personally I usually follow the path of least resistance, so I will perhaps vote for outsourcing the thing as a member of WG, so take this just as an example (where we will essentially “waste” effort of discovery team and money spent on it, but I hope I explained before why this is not a waste).
Important to say here is that the decision shouldn’t be based on money/time alone (or any kind of ROI) - part of this should be “the appetite” - how much is this aligned with what we do as ENS. For important things, like having public SIWE running, I expect that appetite will beat all the odds. I think that for most of the big decisions, this should be turned into a formal proposal, to understand what is the appetite of governance token owners. For smaller things I can imagine running this proposal only within WG (basically turning them into the subdao).
So, now we have the “budget” approved. Discovery group outcome should be good enough to create tasks in something like Clarity. One of the tasks actually could be the project governance, because assessment of individual contribution alone could be problematic and there will be needed some established communication line between the contributors. Aside from that, contributors send PR and are getting paid, that’s the core concept I consider super valuable.
I know there will be feelings about this post 
My experience is like 25years managing software engineering and a month or two participating in DAOs, so take it with a grain of salt.
Also, it may look like I am attacking the core concept of DAO where smartcontract rules. I know, reason is not that I don’t like DAOs or anything, on the contrary, I think this is the future of work. But the human level of coordination would be needed IMO for running things, on top of the core meritocratic concept built in smart contracts. These discussions prove it very well I’d say
1 Like