In this article, the requirements and ideas of what a Guild on WAX is supposed to do is explained. As well as an estimation on the costs of operating a guild, such as hardware and personell.
Operating a Guild on WAX, is something larger than only validating blocks. One of the main tasks of Guilds on WAX is ofcourse to validate transactions and blocks. However, ontop of that, they are also required to provide services, tools and/or some other type of value for the community. Community in this sense is, and not limited to, developers, node operators, websites, games, dAPPs, content creators, artists and users.
Someone with intentions of starting and operating a validator on WAX, will be required to have a solid base setup of at least Block Validation node(s) (BP Node), API RPC node(s) and Peering node(s) that are publicly available for the community to utilize. On top of that they need to add additional value to the ecosystem. It is difficult to specify exactly what value means in this case. As it will be different for different people.
How does guilds add more value than validation?
This additional focus outside of block validation, greatly helps to reduce the overall cost to develop, deploy and maintain applications, websites, tools and games in the WAX ecosystem. Since the teams build these tools can rely on the publicly available infrastructure to get going, test their idea and build a userbase. It can be seen as a way to decentralize the focus on WAX by having a lot of small to medium sized teams that can focus on what they think the ecosystem will benefit from.
To ensure that all guilds on WAX provide value for the ecosystem, WAX has a decentralized team which are elected by the community, whos main task is to review and rate all the work done by guilds. This task is not always simple annd straight forward, as they somehow need to validate and compare different type of creations in somewhat of a standardized system. This team is called The Office of Inspector Generals, and consist of 3 community elected entities, which individually rate each guild and their work. These rating are then presented to the community and used as a guideline for small and huge token holders, proxies and teams when they vote for the top 21 Guilds.
The team that created WAX, holds a large chunk of tokens, they have so far always voted exactly as the OIG rate guilds.
More info on the OIG
Guild updates: https://github.com/wax-office-of-inspector-general/waxguilds/tree/main/reports
Guild guidelines: https://wax-oig.notion.site/Versioned-Guidelines-a83f99e6acd0444ab7cec9419ca4eff0
WAX Governance: Found in the WAX Course
Article on OIG election: https://academy.anyo.io/elect-new-ruler-of-wax-governance/
Rough estimate on hardware costs
I will use hetzner as the example, as that is the most used service provider and also the best value with unlimited bandwidth and good easy to understand options. There are plenty of other options, such as OVH, Leaseweb, IONOS etc.
In the best of worlds, the Guilds on WAX would use a vast variety of options and datacenters. Some of the guilds on WAX operate their own datacenters, others rent private Racks and put in their own managed machines, while the majority rent Bare Metal machines from bigger server service providers like those mentioned. The obvious benefit of this is a lower fixed monthly cost, and full control and choice of hardware. The additional benefit is that the chain is less affected by the choice of one or a few external companies or regions of the world. The power of decentralization.
These numbers and prices are not exact, and each guild will have a different setup. So don't take this as a blueprint, but more like an idea of what likely is required and the estimated cost, which probably can be pushed lower, and likely higher. The estimation on quantity is on the lower end, some service might run on the same machine(s), depending on hardware specs, node type and much more. Some services, like Full History require multiple physical machines for scaling and redundancy.
The price also depends on when you got the machine, the specs (CPU, RAM, Drives), and location.
est. € Per machine
Mainnet, Testnet, Backup
Peer Blocks & Transactions
Line: EX, AX, PX
Line: EX, AX, PX
Feed Live Data
Line: EX, AX, PX
Load Balancer, Firewall, proxy...
Depends on setup
Line: EX, AX, PX, Cloud
Hyperion, Dfuse, Roborovski
Line: EX, AX, PX
Atomic API, IPFS, LightAPI etc
Line: EX, AX, PX
Database, storage etc
Elasticsearch, Postgres, Snapshots etc
Line: EX, AX, PX
~ €4600 / $5000
Total avg* - Estimated from collecting data from some of the long-term guilds on WAX. The lower end seem to be for Standby Guilds, or teams that own their hardware.
n=11 median=€4600 avg=€4845low=€3000 high=€8500
The rough estimated monthly costs depends on the services and tools that they operate, how they set it up, and how much traffic their services recieve. It's impossible to know exactly what all teams run, but those offering non filtered History APIs, and those operating highly used services will have a way higher cost than average. They see infrastructure expenses close to, or even above €10000 a month.
The Guilds tech team
Now, each guild will have different focuses on what they think the ecosystem needs and how they can provide that. Some cater more for users, other for artists, developers or WAX infrastructure. It's impossible to decide what is most important, as it all depends on the observer, and what situation the ecosystem is in.
During smooth periods, teams focusing on understanding and developing solutions to help the infrastructure side, might be seen as not adding value, while during a crisis, those teams are essential to smoothly get through it.
And teams building foundational tools and contracts, that other dAPPs tap into and utilize are essential for developers, but might not seem to add a lot of value from a user perspective, since they use the tools that is built ontop of these core functions.
Or teams providing accurate history of transactions, might seem less important in the everyday use of the chain, but essential to analyse issues, patterns and help to provide users and projects with the information they need
Personell - The team
With that said, there's a set of core work that all guilds do, and we will focus on that aspect. For this, a Guild is required to have at least one highly efficient and knowledgeable Fulltime system administrator. Plus at least one, probably two additional team members that understand and is capable of managing the systems and softwares.
The WAX Guilds had a Median of 3 team members on the tech side.
Operating Infrastructure for WAX requires 24/7 availability. Where at minimum one in the team is on standby in case they are required at any time and point of the day. This might be baked into their salary, or is something additional.
Now, we can't know what each individual employee earns, most individuals and teams tend to want to keep this information to themselves. So what we can do is to check market price for the required talent, which of course is just an estimation, since it will depend on location, company, person, job, benefits and a lot of other aspects.
Data from glassdoor.com, Yearly salaries in $USD - Germany selected as EU base price
As stated before, we don't know what guilds pay their team, how they are employed etc etc, so above numbers will not be an accurate estimation for all teams. But it is the avg salaries according to glassdoor, of what the employees could earn if they worked elsewhere. Keep in mind that avg means that half the people earn less than that in that industry.
For that reason, let's make two estimations, one estimating the lowest income, junior dev in Australia, and one using the median of all the above salaries.
Based on the above, Each tech member costs somewhere around $4000-6000 per month. The actual expenses for a company may differ, as depending on where they are based they will have different fees, taxes and other expenses for each employee. So let's take the number in the middle and estimate a cost of $5000 per month and employee.
The actual estimated cost for a guild
Once again, I want to press the importance around that these numbers are estimations, and doesn't represent the actual costs of any team, but at least they are based on self-reported numbers, plus market salaries. Some teams will likely be below, around or above these numbers. Many teams also have other sources of income, and these employees may also do other work than only WAX related tasks.
Hardware: ~$5000 / month
Team: ~3, ~$15000 / month
Estimation on costs: $20000 per month
Other employees and costs not taken into account
Now, I'm not trying to imply that the income of an employee is the cost of the company, nor that everyone working in Guilds earn salaries in the lower brackets. But we have to make an estimation somewhere, and some teams will have higher costs, and others lower.
The above calculations also only take into account the tech side of the team, these employees might have other tasks, or the guild also have employees or hire services to enable other tasks. Such as Graphic designers, content creators, social media manager, accountants, frontend developers etc.
But these jobs or tasks are rather individual, and we don't have enough insight in each guilds operation to do a full estimation. And to do this properly, we should also look at other sources of income for these teams. The larger teams (4-5 techies) likely have several sources if income, while the teams with 1-2 techies probably don't have much time or capacity for that.
What type of person or team should consider operating a guild?
The short answer is, a person, or a team with linux sys admin experience, that has a lot of eagerness to learn and believe in the future of WAX, NFTs and blockchain.
To get started, you have a lot of softwares, tools and systems for the foundational guild infrastructure, and even more as you add other services. This will require time, research and persistence.
I've been working in the DPOS ecosystem for the last 5 years, I've seen many teams come and go. Those that are still around tend to be started by, or quickly acquired at least one passionate and driven system administrator willing to learn and put in the required time. Teams that don't expend, tend to get burned out, and slowly withdraw from the ecosystem, or change focus.
Expect months of research and testing.
You should expect 2-6 months to learn and set up the initial infrastructure and monitoring. Also in that time, your should preferably also start to work on what you believe would be a value-add for the ecosystem.
As in any new business venture, you need some type of capital to pay for the growing expenses when set up the infrastructure. You also need to buy yourself enough time to get a foothold in the ecosystem.
There's a lot of documentation, guides, articles, videos and chats that will help you in this process. And a lot of knowledge that frequently is shared by other guilds.