Guild types and onboarding thoughts

@ALuhning Love the thoughts! Some ping-backs for you –

Success – you make good points about quantity vs quality/activity/uniqueness. I think good metrics do need to be a bit blunt to be useful but you always sacrifice something in the attempt.
I actually see Guild metrics as a progression where, at each step, we learn more about both the guilds and how to measure them.

For example, in the earliest days the metrics that interest me are indeed blunt:

  1. How many guilds are there?
  2. How many on-chain accounts are linked with each guild?

I don’t think this is a mistake because you have to start somewhere and bluntness is ok to begin with as long as you don’t get overconfident in the numbers. Crawl → walk → run.

Then, presumably (as you indicate), people will take advantage of this (if incentives are provided) and/or we’ll notice abnormalities in the data which help us better classify different guilds and/or we’ll notice that it’s too broad to accurately reflect behavior of “good” communities. Then the second round of metrics will probably need to classify guilds differently, perhaps subdividing into categories like {mass community (size matters), service guilds (quality matters)}.

Those categorizations also have to come back to first principles of “what actually makes this guild successful?” since they should really be grouped based on that – eg. a service guild may ultimately be like a consultancy and thus use something roughly like “billable hours” whereas a marketing guild might be “bounties earned” (from a third party, presumably, to avoid self-dealing). A developer guild could manage towards number of active developers (which we already measure but don’t correlate with guild memberships) or they could manage towards $NEAR earned via hackathon prize referral incentives. A BD guild could manage towards maximizing their referral fees for refer-and-earn for grants.

And we can also look more at what engagement actually means. I’d argue that, at the early/high-level stage, the halo provided by an “active, engaged account” should extend to any guild that account has chosen to associate itself with. For example, a developer actively creating an app and winning hackathon prizes should give “success juice” to the Dev Guild that they’re associated with via how that guild’s metrics are tracked, and that an account like that is worth much more than an account that is idle or just auto-retweeting things. Over time, we’ll be able to quantify that better.

Regardless of the way we sub-divide categories or classify engagement, my major point is that we should start by collecting the blunt metrics first and then interrogate the data to make better decisions about the specific metrics over time. It may mean that it over-glorifies Guilds that have low membership requirements to start with. As long as we know that, we can keep evolving the approach to be more specific. But, as experience tells me, trying to over-engineer such things up front leads to constipation and lack of shipping and lack of learning.

Global Onboarding vs Guild Onboarding

I’ll differentiate global onboarding from guild-specific onboarding because they’re two pretty different sides of the same problem. Global onboarding means everyone who touches the NEAR ecosystem (eg visiting, using a NEAR-based app, watching a video on YT, retweeting a thing… whatever) ideally finds their way as quickly and easily as possible to joining a Guild since Guilds are a fantastic retention and engagement mechanism to keep people engaged with the project and its facets.

The goals of global onboarding are thus supply-driven – take the firehose of traffic touching NEAR and sort them as quickly/easily as possible into the best Guild for them. In reality, this likely means optimizing for them to choose between the ~3 most popular Guilds by default but giving them the option to find more specialized guilds later in the process too. Think of this as the “If you get 7 followers on Twitter then they consider you properly onboarded” loop… if you just join a guild and post a thing, we’ll assume you’re in good shape and, even if you later would join 1-2 other more specialized Guilds, at least you’re not going to go wandering off into the woods and go check out some other ecosystem instead.

Then the other end of that funnel is the demand side, or Guild Onboarding, where the Guilds need to be ready to receive new members and help guide them home (at least those Guilds which actually want to receive unvetted new members). This process will be different for each Guild, but it does need to pass through some checkpoints either way so we can all acknowledge when someone has actually “joined” that Guild (ie an on-chain linkage).

It’s sort of like Club Rush or Fraternity Rush from school days – lots of people trying to rapidly sort themselves into the communities that are right for them.

The success of onboarding to me shows up in the success of members being more engaged (yay for on-chain activity!) and referring members, which ultimately means more people in the Guild. So, while we should always track metrics around onboarding in both directions, Global and Guild-specific, I still think its success materially feeds into the previously-mentioned metrics.

DAOs / on-chain tools

Yep, the goal here is to dogfood everything community. You can see in other ecosystems (esp Ethereum) that communities are iterating their tooling extremely quickly because they’re actually trying to build communities and those communities need tools. So the more we try to use this stuff, the more of a favor we’re providing to every builder who will need them later by helping to create better tools along the way.

Other stuff

every Catalyst community (DAO) is also an account and has a decentralized identifier (DID) making it unique. Every person joining a community gets their near account linked to a DID as well making it possible to create and store account specific data (all of this is done through Ceramic Network integration). Thus, for our communities, we know when someone joins or is affiliated because a proposal is passed that signals that.

yassssss… and doing something similar in spirit to this (though ideally even simpler in practice) is what I hope to see from the Guilds program, so every human who’s touched NEAR is a member of at least one Guild, which is verifiable on-chain.

I also hope that every hackathon that a developer wins or grant recipient a BD person refers or investment deal an investment Guild member refers will provide some referral incentive back to their guild automatically so all guilds are aligned with their members to drive actions which drive us all forward. The mechanics of this are still in need of production but the vision has been there from the beginning – Guilds are a program where anyone touching the NEAR ecosystem should find a community and those communities should be directly incentivized to support their members doing things that help our ecosystem.

1 Like

Roles work for sure, but don’t think they need to be used depending on how one perceives ownership of the membership. One can look at membership ownership as either belonging to the member or the community. If we take the perspective membership, represented by the NFT, is owned by the community, then the community can mint an NFT which contains the account name of the member. We can then query the community NFT holdings to find out if someone is a member. If membership is revoked, the community burns that NFT.

Lots of goodness in here - appreciate the elaboration. Completely believe in the vision and ultimate end state in driving NEAR adoption. Community didn’t bring me to NEAR (was random luck), but the community is one of the things that has kept me here. Hoping I’m contributing to that for other people as well.

Fair enough - just question utility of data you know has flaws in it other than an intermediary step to something useful. So, if it’s internal and the basis of a subsequent round of analysis - sure. Don’t think it’s overengineering to address what you intuitively know needs to be considered up front.

Using them for external communications would still be concerning in my opinion. Not saying that’s what it would be for, but, for example, marketing a number like on-chain accounts linked to each guild published as some campaign without explaining what that means to the average person might be considered by some to be potentially misleading depending on how it is used.

This is excellent - reputation by association (both ways). Probably going to steal that :slight_smile:

1 Like

I agree. Quality is definitely a big metric over just on chain member quantity. It can act as a baseline to start with. Ideally, we should get to a stage where we reward impact. Even if it’s a small guild of 5 people, doing incredible work, bringing on projects to NEAR, supporting projects building on NEAR etc.

I remember us discussing this. Definitely agree that we should start dog fooding our own tools. It’s always a balance between is this dog fooding our platforms and tools to get more people to use them vs is this easy enough for a web 2 person to come, join and interact with. If we’re waiting for the perfect owned platform with analytics, moderation and all the features we need, that would take time. But that’s where open source tools come into play, where we can start building on top of what’s already there. The team is already experimenting with functions there, including a NEAR login to Discourse:

I’d love to discuss more about this with you!

If anyone is interested in helping us out with the community platform, you can read more about it here and comment if you’d like to help: On chain Guilds platform- Come help us out!


All of this discussion around on-chain Guild members is awesome, and there are some great ideas, but I don’t think it’s clear the reason ‘why’.

By that, I mean the reason why we are fixated with having these Guild members on-chain. And even if they’re on-chain, what benefit do we gain by going through the significant amount of legwork required to add another level of ‘membership verification’ through NFTs etc?

I’d argue that the value the Community and NF gain from this is negligible.

Yes! :100: %!

How does that relate to supporting, organising, and mobilising communities of people on NEAR?

I don’t think being on-chain lends itself to enabling the above to any major extent.

I was speaking to a member of the NEAR Community recently in a feedback call for the survey. They’re involved in a project in the Creatives corner of the NEARverse and their focus is not on crypto, blockchain, or the like.

Rather, their focus is on adding value to artists and art projects off-chain, IRL.

They still add value to the Community and are still part of a Guild, but to what extent does
strongly encouraging Community members to be on-chain work to fulfill the original goals of the Guilds system?

If you’re part of an active Telegram group which supports artists building on NEAR, and you’re officially a ‘Guild member’, why do they then have to have an NFT or another hoop to jump through to verify that on-chain?

I think it’s important to understand why we want all of this on-chain verification mechanisms. If the value it adds is of little significance to the OG goals, are we just chasing a vanity metric?


Dear @erik.near first-of-all, we need to fight with favoritism in NF. Some guilds like Swine Guild who has great relationships with NF people have easily access to money , even if do their work bad.

I wrote my thoughts in their [September’s],(Swine Guild September Report🐷) October’s and November’s report, but unfortunately our councils didn’t put attention on it.

NF paid to this guild ~ $50 000 for last 3 months. The same money NF paid to Near Pad for instance.

Last payment:

Probably, NF think it’s a great idea pay $1200 for 5-7 memes and Facebook page management where 24 people?

Or for “cool” memes there children eats toilet paper?

Or May be it’s funny when Near memes make fun of other blockchains? @David_NEAR you approved these memes

Quickly friendly reminder:

And compare with memes which you approved today:

Some guilds like INC4 don’t do any reports. They just create proposal and include information about invoice #

Where is their report , dear @Grace ? @Primaveralina good day. Could you please show us November’s report? Thanks

At the same time I see how other guilds and people waiting for funding long months.

@illia told here (48:39) what Web3 idea is when people who use platforms, blockchains are able to control it.

Илья @illia очевидно , что консулам NF не нужно коммьюнити, чтобы кто-то там что-то спрашивал, узнавал, добивался соблюдения этикета, прозрачности. Все равно консулы делают так как им нужно и только то что они сами хотят. Для них коммьюнити это абуза и неудобные люди, которые лезут не в свое дело. Они говорят, что что-то будет меняться. Но ничего не происходит . Я здесь с июня

и наблюдаю как одни и те же люди получают деньги, голоса коммьюнити «выслушиваются» только для «галочки». Я отправил с десяток писем консулам NF, просил хотя бы немного прислушиваться к коммьюнити , но воз и по ныне там. Я не вижу от них никаких усилий направленных на то, чтобы наделить коммьюнити хоть каким-либо правами. Как они скажут так и будет, не прислушиваясь ни к чьему мнению. Грэйс даже не рассказывает никому почему она принимает те или иные решения, никаких стримов или ежемесячных отчётов от неё. При этом я хочу сказать, что здесь среди ребят очень много тех, которые действительно бъются за идею Web3, на 100% перекладывают твои слова о необходимой прозрачности и равноправного доступа в свои проекты и гильдии, создают ДАО.

Now, only councils from NF control all Marketing activities and Guild’s funding and don’t listen Near Community members.



@Dacha (not commenting for you, but for the newcomers who might believe occasional whining)

  1. It’s October payout request, not November
  2. Click on the purple link in the circle that you drawn on the screenshot with your own hand, it leads to the October report respectively ([Welcome] INC4 Guild Introduction - #28 by Primaveralina)
  3. November report and payout request have not yet been submitted - once done, you will be able to access the report link as well as on any other payout request I’ve been posting since I’m in the NEAR Guilds.

And stop tagging me in your posts. This is getting ridiculous :woman_facepalming:
Do smth useful instead.


1 Like

Did it once.

Ok, thanks. I’m ok for hate speech and bullying.

I Do lot of

Could you please include information in description. Thanks

1 Like

(Ignoring the back and forth because that part of this thread isn’t in scope)

Some good arguments above could be used to ask “why on-chain anything?” (not the worst question either). Let me hit that in the end, but I think the core question is actually higher level, which is that there are multiple types of things that are all being grouped under “Guilds” right now and it’s probably worth pulling them apart going forward because that tension seems like the root of our challenges.

IMO, there are two ends of the spectrum:

  1. Service Entities which basically act like consultancies, doing work or performing valuable activities in exchange for tokens. Some get those tokens via the community fund, others (like Silicon Craftsmen or possibly Legal Guild) get theirs directly from projects and are basically like consultants. These are analogous to many of the popular communities on Ethereum like RaidGuild. They should be judged based on their performance just like any consultancy is.
  2. style communities which are far broader and intended to be communities first and foremost. They should be able to expand as broadly as possible and optimize for member engagement and happiness so there are awesome places in the NEAR ecosystem to drop in, but TBH the members don’t have to directly care about NEAR itself. Ideally, someone who already runs a meetup should be able to turn that into a NEAR-based community with a few clicks using a tool like Astro.

The challenge is that they’re quite different but we treat both as “guilds”. Even more challenging, there are totally ways for one to become the other, so it’s a spectrum more than an absolute.

Which should be DAO-backed? Probably both, but in different ways.

Well, the consultancy-style service communities should be incorporated regardless of whether they use a DAO structure to organize their activities. A DAO just provides more flexibility to use a collaborative structure around it if the people want to open up their membership more broadly. But tbh isn’t actually necessary in order to run a successful service guild.

The broader communities are more complex but there should probably be multiple tiers of members:

  1. Lurkers – just hang out. Don’t have to be on-chain.
  2. Members – they choose the community by signing up with their wallet to get access to additional features, chats and possibly economic activity. This should ideally be permissionless (they select the Guild). This is the stage where the community benefits from being tokenized (and thus DAO-backed) since those tokens can be issued to provide these additional features and access.
  3. Owners – these are people who are given decision-making power in the DAO and must be selected by the community (the Community selects them, they must earn it). There should probably be a multi-level way to badge up, so it doesn’t have to be binary. The DAO is necessary here since it coordinates governance.

These should be fully autonomous – they shouldn’t rely on any support, whether financial or operational, from the community team or anything else. But they can still benefit by using incentives that we put out there (again, example being hackathon prizes).

I think if we can’t build the right model for “guilds”, the rest of it is actually premature.

Then we have to set pretty clear expectations on the difference between them. Even the word “Guild” is hard because it can be justified in both cases. Maybe then its’ more like “Service Guilds” vs “Guild Communities”.


Their work and results hidden from Near Community. Only NF and projects can judge their work. It’s not a problem, but due some confidential issues, they don’t need to follow transparency policy. Will be great to have list of these person and guilds to prevent misunderstandings. They are able to ignore questions from Near community and briefly describe open part of their work.

Absolutely. Unfortunately, I don’t know any DAOs on Astro DAO where council were elected by community choose. Even Community DAO has “pre-elected” councils from NF. May be only great team from Creativities DAO. All other have one -two-three councils (friends) who make all decisions. Community members cannot change anything. Only flat democratic DAO structure can involves community members in voting.

And yes, third category is “trusted”. These guys don’t need to write reports and have any DAO’s . Great example is Reddit team. Only one council . How Reddit community members can vote their leaders? How can vote for any initiatives? Members cannot even create any proposals.


I agree with @erik.near on the conceptual split of organizations that provide services and communities that are designed to activate and engage people.

I would vote to have different names to distinguish between these two types.

Given we are going to have more companies and partners contributing to building ecosystem - it make sense to call this kind “Partners”. Even if they are using DAO tooling for their business - they are not operating as a group of professionals or people with common goal (which is definition of a guilds).

Then the sub-communities of NEAR should keep the “Guild” naming.

If we all agree on this nomenclature, next step would be to classify the current Guilds into these two types.

For “Partners” it would mean clearly define their mode of operation and switch to contract based payment - where project would pay them for specific delivery.

For “Guilds” I see it more as defining what are major sections of NEAR community: languages and regions, themes (DeFi, NFTs, DAOs, etc), professions (devs, designers, product managers, etc). These sub-communities are evaluated by engagement of members and also how these people get activated by contributing to other projects and partners.

There will be Partners that provide services to NEAR Foundation directly (as a proxy to NEAR Ecosystem).

For example:

  • “NEAR Week”, which is an information and a marketing agency. In turn NF will be paying the marketing contract. Over time this contract would transition to Treasury DAO and it’s sub-DAOs that are responsible for that vertical in the ecosystem support.
  • A current guild that suppose to be providing services to projects, but actually not delivering due to lack of personnel or time - re-qualifying as partner would mean they need to deliver on the contracts they make with projects. E.g. they would need to either commit to working with projects or dissolve it.
  • For “DeFi Guild”, I would expect us to have Telegram or Discord channel where people are routed. Have an active discussions on the latest of DeFi both for users and developers. Learning and job opportunities, brainstorms and Twitter Space sessions. Community organizers and managers of such guild would then be the on the DAO council, managing the budget they have to facilitate engagement in their community, rewarding retroactively members who engage and bring value as well as receiving some compensation for their time.

PS. should be used a funding tool, to allow for streaming payments from one DAO to another. This would enable a control over the budget that Guild can spend, while not needing a constant confirmation and oversight.

Process of monthly reviews will be providing the visibility in the funds spent as well as understanding if this Guild is alive or should be cut from funding. Community members like @Dacha who are constantly on the watch may provide inter-monthly information if there is something weird happening in the guild’s community.


Attn : @Sofia_Alum

Thank You very much @illia . 24/7 here.


In a governance paper in 2018, we distinguished between 4 dimensions of communities you can deal within digital business models. We suggested, Dimension (1) Authority (stakeholder community), (2) Competency (peer community), (3) Enabling (supplier community) and (4) Capacity (market community). For each of these dimension you can use names. The problem with Guilds is obvious - it covers over peers, suppliers (partners, providers, enablers) and is part of the market. Due to the fact that you reach these dimensions with different approachs and languages I would value the approach and language higher than the naming itself.


Switching up naming only furthers confusion, it’s been hard enough to distinguish (and relate) Guilds and DAOs. I would prefer to keep referring to the community organizations as Guilds and referring to DAOs as their tooling.

Regardless of what value a Guild is providing (growth, service, or the other vectors @hitz mentions) they can be held to the “grant” payment model. Before doing work, a Guild needs to provide:

  • Desired funding amount (up to X, have heard 10,000usd in NEAR)
  • A meaningful metric and estimated numbers against it.
  • Plan of action to back up the estimate.
  • (if not the first time requesting funds) a report including the measurement numbers and what was learned, whether the numbers were met or not.

This simple but effective framework enables Guilds of all kinds. It’s flexible enough to accommodate wild experiments and rigid enough to ensure value is generated every month. It also provides managers and auditors a clear template to check.

Edit: I contrast the “grant” payment model where funding is provided ahead of time with a “bounty” payment model where work is rewarded after it’s complete. “bounty” model works in any increments including streaming.


Firstly thank you @erik.near for starting this discussion and to everyone for sharing thoughts, insights, concerns and suggestions. I largely echo the sentiment in this thread but would like to share a few thoughts too:

  • Whether it is a service Guild or a community building Guild (two broad categories as put forth by @erik.near above), every Guild should ideally have an on-chain presence in the form of DAOs. If not for receiving or disbursing funds then for governance or facilitating transparency and macro-level decision making.

  • I am not averse to the idea of streaming payments as suggested by @illia to Guild DAOs. If the application of similar tools already available in the ecosystem can help make Guilds more responsible and transparent, let’s use them!

  • We need to let the Guilds decide which organisational structure works best for them while making sure that it adds maximum value to the ecosystem. Agree with @shreyas that some Guilds are comfortable working in a closed team-like structure while others prefer opening themselves to the community to join and participate in.
    To assess the former (closed team-like guilds) we should look at the impact they make and experiments they undertake. While those Guilds may not have open membership, they can certainly undertake activities or do experiments that entail the involvement of the larger community to be considered successful.
    To assess the latter (open membership Guilds) we are already tracking the number of new members who sign up or join those Guild communities.
    In short, instead of just looking at the number of people who join a telegram chat or fill up membership forms to assess the impact of a Guild, we need to also look at the number of people engaged or opportunities/experiments initiated for the community to participate in. This in my opinion mitigates the “one size fits all” approach and allows more flexibility for people.

  • Agree with @starpause on retaining the “Guilds” naming for all guilds instead of getting too caught up in a nomenclature exercise UNLESS it is required to make things easier for the Foundation, Community and all ecosystem stakeholders in terms of compliance with laws or different regulations.

  • Guilds are NEAR’s competitive advantage over several other ecosystems in my opinion. What we need is the optimisation of existing guilds infrastructure instead of drawing hard straight lines of distinction between two broad types of Guilds, which to some extent dilutes the idea of Guilds we have now - which is inclusive and representative of all groups of community contributors now.
    That being said, I am not opposed to classifications if it brings more efficiency in oversight and reduces suboptimal dispensing of community funds WHILE retaining the inclusive nature of the existing guilds infrastructure where anyone with an idea can organize a community or build a team around.

  • There also needs to be a discussion around “onboarding” and what it means in different contexts. If a person is onboarded by a Guild, it must be made sure that the Guild follows through and encourages the person to use NEAR’s tools, explore the opportunities etc and most importantly engage. Lurkers in Telegram chats or Discord groups don’t add much value to the Guild or larger ecosystem if they are not actively participating, learning or contributing in some way.

  • The point @David_NEAR makes about taking into consideration the off-chain membership is very important. We don’t really need everyone to be on DAOs or holding NFTs to prove membership BUT we do need communities and contributors to have an on-chain presence (ideally DAOs) to make it easier for them to govern, organize and align their group while exploring the vast expanse of web-3 and the open economy.

  • I can see that the vast majority of opinions being shared here are by Foundation members or NEAR Core members and only a handful or even few by community members. I thus felt it necessary to voice my opinions. I am also open to jumping on calls with any and all participants in this thread to discuss all ideas in length. If a community representation is needed for such deliberations around community, I am always available!


Previously, we’ve classified guilds into

  • Theme based guilds- NEAR Music Guild, Cypherpunk Guild etc.
  • Service Guilds- Legal, Design, etc.
  • Regional/Language guilds- NEAR Russia, Vietnam etc.
  • Project guilds- Aurora, Ref Finance etc.

Agree that service guilds are similar to Raid guilds like communities- the pitch has always been that you come in with an existing skill or a real interest in learning and honing a specific skill and there will be a guild where you can do that while contributing to the wider NEAR ecosystem.

IMO, the others, except for project guilds, can be categorized into meetup like communities, based on their niche and audience. Communities of practise (COP) or communities of interest(COI).

Turning an existing community to a “NEAR based” with a few clicks- what would that look like?

I would frame this as “Contributors” as there can always be new contributors. The community doesn’t cease to exist if the owner exits.

What would that look like?

I feel like with most of the guilds, we’re a long way from getting here. Initially, it requires a lot of hand holding as even in the case of service guilds, most of them are collaborating and working together for the first time and they dont already have a working relationship. This means that there needs to be a place where they can rely on getting the support for a period of time, before they become independent. Most of the members in these communities haven’t worked together before and it’s really difficult for them to come together from day 1 and push forward.

Community commitment curve

We can think about more specific names that better reflect what they’re doing- like service, theme based guilds etc., but I think partners as a term can be confusing especially given that these are community members. We also don’t want the community to be misrepresenting what they’re doing.

We can set up an incubating model for service guilds and help them get “independence”. In this context, that would mean working with them on a timeline as well as steps to assess their value proposition, make introductions to potential “clients” who would pay them for their services etc. as the hand off stage.

Will also check out and review Roketo. cc @Jessica , @jcatnear

Completely agree.


The Guilds represent a strategic component of Ecosystem development and growth. I am proud to have seen the growth we have experienced in our Ecosystem in the last six months, and I realize that the conversations we are having now are triggered due to that growth.

I want to thank everyone for their input. However, I feel we are only scratching the surface by pointing out symptoms and going straight to the prescribed medication. I think we missed the problem definition and, more importantly, the source of the problem.

The problem is not the ‘naming’ but something else. @hitz input was guiding us in that direction. A way to better define the issues around the sub-communities, groups, or guilds. The problems and/or source of the problem could be pointing to funding, reporting, commitment, value-creation, trust, empowerment, opportunism, judgment, governance or lack thereof, etc.

The community team started working on a Tiering system to define the conditions better and support the development of the different types of Guilds. Such Tiering system is to be communicated and posted in this forum.

  1. Tiering these groups may help address problem areas that are more specific and relevant than a one-fits-all type of approach.
  2. After the tiering system is in place, funding will follow next, funding according to the tiering system.
  3. To implement an effective rewards mechanism, a Governance structure is required.
  4. A Governance will allow active community members to participate in the decisions that affect our Community.

All the above items are work-in-progress and will be shared with the Community for feedback and iteration. We hope that this multi-level approach to a diverse set of challenges resulting from the growth and maturity of our Community finds its way to bring some light and pave the way for further decentralization.

The Community team constantly measures the strategies and execution approach to see how they fit our values: Ecosystem first. Don’t forget what brings us here.


It actually seems like there are 3 categories since they’re meaningfully different:

  1. Service Guilds / Partners / Consultancies – as above, treat them like service providers that happen to be community-oriented/open in how they recruit members but otherwise are able to contribute in clear, accountable ways
    a. TAM: ~1k’s of people
    b. Who should be accountable and responsible: Ecosystem Success team.
    c. What they should do to support: Identify and test-drive high quality service providers (whether communities or consultancies, shouldn’t matter) to match with startups in ecosystem (including NF) who need services. Should be demand-driven not supply-driven.
    d. Current challenges: unstructured expectations around what these are and how they work means low quality work and high quality work aren’t really differentiated.
    e. Examples: Legal Guild, 4nts Guild

  2. Guilds – essentially crypto-aware communities that engage with NEAR or projects on NEAR as a way of doing the normal stuff communities want to do – learn, earn and/or play together. They would exist anyway without support but the support helps supercharge them.
    a. TAM: ~1M’s of people
    b. Who should be accountable and responsible: Community team
    c. What should they do to support: provide funding to bootstrap the communities (the equivalent of pizza and beer money for meetups), knowledge/workshops, recommended infrastructure/tooling, routing of new members into the communities,
    d. Current challenges: it’s quite unclear where communities are and where someone’s “home” could be to learn, earn or play together with like-minded people
    e. Examples: NEAR core discord community, Vietnam

  3. NEAR-based communities – web2 communities who don’t care about NEAR or blockchain but can benefit from being tokenized and DAO-ified. The full Open Web vision – how does every community on or facebook have a token and a DAO?
    a. TAM: ~100M+ of people.
    b. Who should be accountable and responsible: Astro team.
    c. What they should do to support: build the best and smoothest tooling out there to uplevel vanilla communities to slowly use crypto primitives to engage their members, provide necessary legal underpinning to do so (eg recommended legal paths when communities start hitting KYC needs), evangelize the heck out of web3-ifying your community and elevating success stories of communities that started using the (NEAR-based) web3 stack, major partnerships to drive this top down.
    d. Current challenges: This doesn’t actually exist at all because there’s still no basic way for web2 communities to tokenize.

The core question under all of this is how active of a role should the community team/fund or other funding orgs take in funding each of these stages. I’d argue the answer depends heavily on the goal:

  1. For funding communities to get them going, it’s “as little as possible to help develop and bring these communities to sustainability”
  2. For funding activities that should get done, for example running NEARCon 2.0 or driving 100k new social followers for NEAR or building a Udemy course to teach 10k people how to dev on NEAR, it’s “whatever we’d pay a consultant to do the same thing” (since this is pay-for-work-done and NOT a charity act for the purpose of bootstrapping a community)
  3. For funding orgs that give grants themselves to grow an area of value, for example Cypherpunk Guild, which is both a community and a grant-giving organization, it’s “whatever the Funding team would do normally if a nonprofit came in asking for the same ask to accomplish the same goal.”

It seems that these pieces are being all blended in ways that reduce clarity. I hope this helps create some clarity.


Thank you all for such a helpful discussion, which definitely brings clarity to me. I hope more community contributors like @chronear voice their perspectives.

I think NEAR-based communities can be called guilds if they want, and we all benefit from the structure of NEAR Guilds rooted in various languages and regions.

As a member of the @Community-Squad, I believe many squads of the NEAR Community Guild(s) are coordinating projects and contributors with Community DAOs. We will need more local and global community managers, forum sherpas, and moderators for Telegram, Reddit, Discord, wiki, etc.

I’m hoping to further experiment with a Community DAO that includes other Community DAOs / Guilds in specific groups, according to their proposed tiers and / or categories.

To summarize: I believe our Community Guild empowers communities to support ecosystem development by helping anyone find ways to get involved.


Thanks for the mention @jlwaugh

Looking forward to working closely with you and the NEAR @Community-Squad as we usher into the new era of “regionally-rooted, globally connected” communities of contributors. Count me in to contribute to all multi-faceted experiments pertaining to community governance, reward structures and inclusion impetuses.

One suggestion: Including a more diverse set of ecosystem participants, say Projects and Research Groups to the mix will expand the scope of the experiment/s exponentially and provide deeper insights on best practices, special cases, blockers, information exchange and engagement standards across the board.

Absolutely agree with @shreyas that this will be an evolving conversation and process as we continue to grow together and build our Community Commitment Curve. We can’t move processes too fast. What ignites quickly burns out just as fast.

I think the Proposed structure for Community-driven DAOs along with the thought-provoking discussion in this thread are significant steps in that direction! :trophy: