Budget Category: Core Development

Proposal Kickoff: November 1st, 2020
Budget Amount: 50,000,000 SC per month, from block subsidy
Treasury Kickoff: 240,000,000 SC, from treasury
Timeline: Active until Jan. 2023
Stockpile Cap: 960,000,000 SC
Leftovers: to be returned to treasury

This proposal is to carve out a significant budget for the core development of the Sia protocol. This budget would be intended to cover things like transaction pool maintenance, wallet maintenance, explorer maintenance, testing frameworks, scalability enhancements, testing, auditing, bug fixes, etc. Anything related to code on the network that is necessary for the functioning of the Sia protocol. This category is also to include technical support and documentation.

The kickoff sum is intended to be set aside as a stockpile to be used to pay the foundation members. A big chunk of it should be sold rather than kept as SC, so that the foundation’s critical employees have job security even in the event that SC falls in price.

Similarly, new hires are expected to be made with such that there is a stockpile (in cash) ready to provide them with job security as well.

I expect a significant amount of the monthly budget for core development to go to Skynet Labs over the next 18 months (more than half, totaling greater than 450 MS in 18 months). After the foundation spinout, Skynet Labs will be employing 8 out of 9 full time core developers of the Sia protocol. Though many of the hours of those 8 will be dedicated to Skynet related tasks, a significant amount of core development work is still necessary, and Skynet Labs has both most of the standing knowledge about Sia core and also most of the active talent related to Sia core.

Skynet Labs should also be contracted to help with onboarding and employee bring-up as the Sia Foundation hires new developers. New developers are going to be the most successful if they can learn from the best, and the best includes the long time engineers at Skynet Labs.

The exact spend and deliverables to be broken down in proposals within this budget category.

3 Likes

I expect a significant amount of the monthly budget for core development to go to Skynet Labs over the next 18 months (more than half, totaling greater than 450 MS in 18 months). After the foundation spinout, Skynet Labs will be employing 8 out of 9 full time core developers of the Sia protocol. Though many of the hours of those 8 will be dedicated to Skynet related tasks, a significant amount of core development work is still necessary, and Skynet Labs has both most of the standing knowledge about Sia core and also most of the active talent related to Sia core.

So how long until The Foundation has all of the core protocol developers in house? I get that that can take a while, but if there are no plans, then the Foundation would pretty much be a money pump for Skynet Labs for forever.

1 Like

The intention is for the foundation to build up a set of core developers over the next 2-3 years. It may be the case though that Skynet Labs will always have a percentage of the most talented developers for Sia.

Depending on how things grow over the years, it also may end up that getting more developers becomes less pressing with time. Similar to Bitcoin, once many of the major core features are built out, it may be less pressing to keep adding new features.

Speaking from personal experience from working at Nebulous (now Skynet Labs), the recruiting game is a long and arduous process. It takes quite a bit of time and effort to build a solid lead funnel, screen potential hires, and onboard new employees.

We’re in the process of hiring full-time roles at the Foundation, but I’d expect that process itself to take some time to solidify and ramp up. In the meantime, I think it makes sense that a fair portion of the budget is allocated to Skynet Labs, the team that is most active and experienced in building and engaging with the core Sia protocol today.

@Taek - a few thoughts on my end for this specific proposal:

  • What is the reasoning/rationale behind a 2 year clause? Can understand that a longer contractual period is desired for stability purposes, but wondering if this should be shortened and renewed on a year to year basis.

  • The community will need to determine a framework for audibility and accountability of spent funds. Ideally, we want to maximize transparency without the process being a major PITA for the grant participants.

  • How should we be thinking about coin price volatility? It completely makes sense that Skynet Labs will be liquidating a large chunk of crypto funds into cash to ensure cash-flow stability. That said, is the Foundation leaving the responsibility of how the grant receiver wants to hold said funds? Should we (the Foundation) be suggesting / enforcing target ratios (e.g. 50% crypto 50% cash)?

4 Likes

I feel that the Sia protocol still needs a significant amount of development at nearly every level, and I wanted to ensure that there was a significant budget set aside over a long term period to achieve all of the development.

I believe we will be able to come up with frameworks for dev work completed in future proposals. I agree there should be a high degree of accountability for money spent on development, but I don’t think we need to figure out the shape of that accountability here.

I think the foundation should encourage people to take as many siacoins as possible while at the same time realizing that people have real world cash flow needs, and of varying degrees. Asking a single 22yo fresh out of college to go 50 or 60% siacoin may be reasonable. Asking a parent with 3 children and limited savings to go 50% siacoin on their salary may not be reasonable, unless the remaining 50% is quite substantial. I don’t think specific targets make sense given the volatile nature of siacoin. I do think it is okay to ask for incremental lockups, however.

3 Likes

(post withdrawn by author, will be automatically deleted in 24 hours unless flagged)

That is a good question, how specialized are the developers that the Foundation are looking for? From what I understand Dingles is correct in that that kind of hourly rate is pretty high. An average dev makes $50 an hour, and I’d assume that you’d go for a more select group so probably more like $100 an hour. But I guess if you’re going for the creme de la creme it may be more? I’d like to find out.

$200/hr is not an unreasonable rate for contracting engineer time from a company. I’d say it’s probably on the low side of what I’ve seen for specialized engineers. There’s a lot more that the rate includes besides just the developers salary. Full time employees cost a lot more than just their base salary.

I’d like to see at least 2 new engineers being trained 3 months after the launch of the foundation, but I also expect 6-8 months before any meaningful contributions are made by them. Contracting Skynet Labs makes full sense.

2 Likes

Good devs are hard to come by, and even the best will require significant training. It will take a long time before I feel as comfortable entrusting consensus code to an in-house dev as I am entrusting it to someone like David or Chris. I think we’d all like the Foundation to become less dependent on contracting over time, but realistically it could be years before we have a team that’s big enough, skilled enough, and experienced enough to tackle big projects without any outside help.

I haven’t given a ton of thought to specific salaries yet, but they’ll have to be competitive if we want to retain top talent. 50MS/month should be more than enough to compensate a small team very well, even if a good chunk is spent on contracting. Long-term, I’m aiming for 6 to 10 in-house devs, with enough left in the budget to contract Skynet whenever we need experienced engineers on short notice.

True — rule of thumb is to multiply by 1.5. The tax situation could be different for non-profits, though.

Historically I think Skynet Labs has had a multiple around 1.3x the base salary as the full cost of an employee. We are also pretty well known for being incredibly lean for the work we do.

I think the foundation hiring 2-3 engineers in the next 2-3 months is a reasonable goal, and I think that the hiring process should kick off as soon as the entity is incorporated.

2 Likes