Following the recent infrastructure proposal, several thoughtful questions came up in the comments and in community discussions. This post consolidates the responses and addresses the framing gaps that were rightly pointed out - particularly around revenue, sustainability, and what's actually concrete in the proposal.
On the proposal feeling "vague"
A few people noted the original proposal led with high-level themes (improve onboarding, optimize infrastructure, continue product development) rather than concrete deliverables. Fair criticism.
The concrete work being funded for this period:
Infrastructure: Full Hive RPC node, dedicated image hoster (serving even years-old images that are missing on other instances), Hivesearcher indexing, Hivesigner, Hivexplorer, Hivekeeper, and the eSync pipeline powering edit history and immutability, ePoints, eOnboard and few other servers/services powering them.
Mobile app: Continued development, performance improvements, new features. Recent release include 3.5.5.
Vision-Next: Self-hosted infrastructure for community-run frontends with custom domains (configurable-app branch). Recent release include 4.3.8.
eSync: Migration toward a HAF-based unified approach.
MetaMask/EVM marketing and outreach: The technical integration is already shipped - direct MetaMask login via Hive Snap, multi-chain wallet (BTC, ETH, BNB, SOL), and EVM onboarding flow are live. Focus now is on reaching the MetaMask and broader EVM user community to drive adoption.
hive-x402: Deeper integration of the AI agent payment facilitator (HBD as settlement layer, zero fees, 3-second finality).
All open source at https://github.com/ecency. Monthly delivery posts on the Ecency blog track what actually ships.
On the ask amount (333 vs 300 HBD/day)
The previous reduction from 396 to 300 HBD/day was announced here - we cut team and expenses to bare minimum. Shortly after, Hetzner increased prices roughly 11.6% and load continued growing. We brought back our full Hive node to handle that load while also contributing to Hive node decentralization. The 333 HBD/day reflects those realities.
For a detailed cost breakdown, see https://trust.ecency.com.
On codebase activity (corrected, with proper credit)
A community member raised concerns that recent activity is mostly maintenance and translation noise. After running the numbers more carefully, this is the honest split.
A note on translation work: Crowdin commits aren't all "automated" in the dismissive sense. Crowdin auto-commits when our team adds new strings to the codebase due to new features, and it also syncs translations contributed by human translators from the Hive community (https://ecency.com/contributors). That's real contributor work - making Hive accessible in many languages is part of Ecency's value. But it's translation work, not code work, and the two should be reported separately.
The proper split, over the last 12 months:
Vision-Next:
Real-code commits (locale-only stripped): 3,858 total
Human authors: 3,748
Bot commits (dependabot etc.): 110
Separate locale-only commits (mostly translator contributions): 2,144
Ecency Mobile:
Real-code commits (locale-only stripped): 1,165 total
Human authors: 1,137
Bot commits: 28
Separate locale-only commits: 3,672
Across the two flagship repos, that's 4,885+ human code commits in 12 months, plus thousands of locale contributions from translators across many languages.
The earlier "1,012 commits in March" framing I shared in comments was misleading because it didn't separate locale-only commits from code commits - thanks to @lordbutterfly for flagging it. The corrected numbers are smaller but still substantive: this isn't a finished product on autopilot, it's continuous active development plus a real localization pipeline. We are constantly finding ways to make product sticky and improve retention, not only work towards easy onboarding but also keeping users engaged.
Beyond Vision-Next and Mobile, ongoing work continues across the mobile SDK, hive-x402 package, eSync, ePoints, eOnboard, Hivesigner, Hivesearcher, Image hoster, and Hivekeeper. All open source. There are few private repos as well related to some sensitive services like Points, Onboard, authorized APIs.
For us number of commits are not much important, what's important is to build and deliver value that helps users and community to do everything Hive and stay engaged.
On revenue and sustainability
The biggest framing gap in the original proposal: revenue and sustainability weren't explicitly addressed. They should have been. This is the honest picture of where things stand.
What's being explored:
hive-x402: Live facilitator at https://x402.ecency.com, @hiveio/x402 npm package published, CAIP-2/CAIP-10 profiles merged upstream. Revenue thesis: Ecency operates the facilitator and takes a margin on volume from AI agent micropayments. This is not proven yet - it's a path being built and explored, not a guaranteed revenue stream.
Ecency Points / AI Credits: Points system was introduced to improve engagement and (modestly) revenue via gamification. Honest assessment: not substantial enough to cover infrastructure cost in any meaningful way today. AI Credits builds on the same foundation - users earn credits through engagement and spend on AI tools. Deeply integrated into the platform, harder to fork.
Vision-Next self-hosted: B2B/SaaS path for communities and projects running their own Hive-powered frontends. In active development, not yet a commercial offering.
Two real constraints we work around:
Hive users mostly come with earning expectations, not spending tendency. Subscriptions or paywalls layered on a rewards-trained userbase tend to push users to free forks.
Anything monetized at the frontend layer can be forked open-source within days. So revenue paths have to be infrastructure-level, deeply integrated, or services-based.
These are constraints, not excuses. They explain why the revenue paths look the way they do.
Worth noting: hive-x402 wasn't promised in any previous proposal. We built it because we're actively thinking about revenue models and where agentic payments are going. That's the pattern - under-promise, over-deliver, and a meaningful chunk of what we ship doesn't appear in proposal bullets until after it exists. The effort is there; the framing in proposal documents has been lacking, and that's the gap this post is addressing. We are always transparent about expenses, deliveries, codebase.
On the focused-tool comparison
Direct comparisons between Ecency and focused tools (wallets, extensions) don't quite hold because the cost structure is different. A focused tool maintains one or two products. Ecency maintains a mobile app, web platform, new browser extension (Hivekeeper), plus full Hive RPC node, dedicated image hoster, Hivesearcher, Hivesigner, Hivexplorer, eSync, ePoints pipeline - services some other apps and users depend on. Different problem space, different surface area. The principle of pivoting toward sustainability still applies; the route just has to fit the scope.
What's next
Revenue paths above continue to be developed and integrated. As specific milestones become measurable (hive-x402 volume, first AI Credits paid users, first Vision-Next licensee), they'll be reported transparently as we explore their use cases and implementations.
Monthly delivery posts continue on the Ecency blog.
Code is open at https://github.com/ecency for anyone wanting to audit directly.
Cost breakdown is at https://trust.ecency.com.
Acknowledgments
Thanks to everyone who engaged with the original proposal - @stoodkev , @davideownzall , @lordbutterfly , @encuentro , @seckorama , @igormuba , @abatae , and others. The criticisms on framing and on the codebase audit were fair and this post exists because of them. And special thanks to the Ecency translator community at https://ecency.com/contributors who make Hive accessible across many languages - your work is real and it counts.
Proposal vote: Hivesigner, PeakD, or Ecency wallet links in the original proposal post.
Thanks, that's definitely a better post!
Thanks, your initial feedback is what prompted this post. Appreciated.
So like I said in the last post, Do you actually make any revenue at all?
Yes, we have small revenue from Points purchases and premium account signups.
Seems like that would be something that should be in the transparency report if you are asking for funds.
Fair point.
https://trust.ecency.com shows expenses but doesn't currently break out the small revenue streams (Points purchases, premium accounts) as separate line items. We'll add that breakdown in the next transparency update. The amounts are modest - not material to covering infrastructure cost today - but you're right that "asking for funds" comes with an obligation to show all sides of the financial picture, including the small income side. Thanks for the push.
You can def up the numbers a bit with more AI assistance. In my analysis
We at Magi do 8x the output that Ecency does on what was 3.5x the budget on basically state of the art greenfield tech thats generously speaking 10-20x more complex with 100s of tests and architecture design needed for each feature that doesnt exist elswhere.
And to note, Ecency was in second place behind us, which should tell people that others need to pick up their efficiency quite a bit as well.
We are not competing with serious blockchains at the current rate of development in the ecosystem.
Glad to see some stupid take every once in a while to cheer up the comment section 😄
Analysis based on what? If you had built at 8x the output in the past 12 months you would have a rocket ship about to land on Mars by now.
Is this the same tech you (a non developer) have been able to pick up and move forward in 2 months using AI?
I wouldn't want to be on that rocket ship built at 8x with AI
I don't mind the AI part. I just don't like people zeroing out on other teams like that for no reason.
It was for a very good reason. Go read the mattermost discussion.
But its great people sticking together throwing insults not knowing context.
I actually saved him from presenting publicly, false information he repeated multiple times in Mattermost, that was off by 90%, not 5%, 10%, 90... which he obviously had the sense to address here.
But great to show support, dont let facts get in the way.
Ugh, love it. Its kind of why I didnt publish it. People dont like others looking. Analysis based on DORA, SPACE, Martin Fowler, Pragmatic Engineer, CHAOSS methodologies and a custom built DME index.
No, not 12 months, as I wasnt leading the team at the time, but for the last 3 months yes, what we delivered is akin to landing a fleet of rocket ships on Mars.
Im a part of a team of 7 people, I did not build Magi.
Feel free to counter what Im claiming.
Love the enthusiasm. Keep it up 🤣
Absolutely.
Bro 🤦♂️
Yes, you can code 8x faster with AI, but can you review 8x faster? No. Not without being sloppy.
Oh, let me guess—you'll use AI to review too?
I use AI to work, and I've worked for two US tech companies (ex-Uber, current Pinterest) where we had access to all AI tools plus direct training from frontier AI labs. AI is unavoidable for our job. It's deeply integrated into our workflow, and without it we'd take longer to ship or even find internal docs.
Yes, we can code 8x, 10x, even 100x faster. But the bottleneck is reviewing. The more code AI writes, the harder it becomes to understand what it did, why it did it, and how it integrates with the rest of the codebase.
In real terms, if you write code 10x faster, you're maybe 2x more productive because of the time you'll spend reviewing and wrapping your head around it.
8x code delivery is dangerously "vibe-coded startup SaaS" level of "productivity."
Your comment makes me skeptical of Magi's safety. You responded well to the proposal, but now you're doubling down on AI to ship even more. That worries me.
If you think 8x code delivery comes for free, that's concerning. Combined with your earlier comment about those three vulnerabilities Claude found being known but unaddressed—by the time they're fixed, there will be many more. And I'm afraid you'll address them with even more AI. Why would you ship code with known vulnerabilities? Why not fix it first? I know why: because that requires reviewing, and it sounds like you really want to ship 8x more code, regardless of quality. That scares me.
I do agree AI can produce better, faster code. But if AI makes you 8x more productive, you're not actually 8x more productive—you just look 8x more productive. That's dangerous, and it makes me skeptical. You almost had me with the proposal, but you just pushed me away.
I really hope Magi works, everything gets fixed, and your approach is right. I hope I'm wrong. But as it stands? What you're saying is scary. It sounds a lot like my friends who just discovered vibe coding.
Edit: ran ecency AI to fix typos and grammar. Sorry for the excessive dashes it adds, I will take them over grammar mistakes.
Appreciate the thoughtful breakdown, Igor. The "AI makes you code faster but the bottleneck shifts to review" point is exactly right and applies to anyone shipping code, including us. Real productivity gain is much smaller than raw output speed, and assuming otherwise gets people in trouble. Good context to put on the public record.
Bro, lets not be absurd. I was being more than fair and generous because not only is this code related but we do hundreds of tests on devnet and then testnet and have to prove the code is safe and working. Its all new tech. Not only does "faster" include code theres multitudes more work going into it.
These absurd assumptions are laughable. We run hundreds of audits monthly, penetration tests, spend probably a 100+ hours a month on safety and bug hunting alone. So we use Ai assistance on code but dont use it extensively on security?
These are the kinds of bad faith accusations that are simply emotionally driven due to liking the project that you feel is attacked.
What are you talking about? Who is shipping code with known vulnerabilities?
Those are all comments I have made on your proposal. Nothing I said here is new except my surprise to you mentioning again very high multiple X coding speeds.
Who is shipping code with known vulnerabilities? You said you are. Maybe we have different opinions on the word "shipping", but to me if it was merged it was shipped even if not in prod. Let me rephrase then using your definition , you are not shipping code (because it is not on mainnet) with known vulnerabilities but you are merging code with known vulnerabilities . You acknowledged on the proposal that the 3 vulnerabilities claude found on my side with a single pass were identified May 2 (so merged before that) ago but not patched by May 5th, not on mainnet, granted, but merged and vulnerable
https://ecency.com/@lordbutterfly/re-igormuba-teks9y
This is what worries me when devs delivery 8x code. I am not defending Ecency because I like them, it is because I commented my worries about security in a proposal asking for over 300k and the guy who made that proposal came to another proposal brag about the development speed of a team that asks for 1/3 of their budget.
If your proposal asked for 1/3 of what Ecency is asking I would actually be impressed and my view on the whole "8x" would change completely, but it is the opposite, Ecency are the ones asking for 1/3 and they are not trying to push their dev methods on anyone else's proposal let alone on smaller proposals, you are the one asking for more (3x more) and promoting your questionable (questionable in my eyes, which I questioned directly at your proposal) methods on a smaller proposal (1/3 the size of yours)
Edit: to make it clear, accountability is all about perspective, ask for more and get questioned more
Obviously.
Yeah not even close to mainnet, going up for testing and undergoing audits.
Oh yes, the "shut the hell up and look the other way" approach. Yeah no, not only will I not abide but this mentality has been present for ages, if you know whats best for you keep it to yourself. If youre running a proposal, its better that we all support each other, right. Buddies. See how @asgarth popped in? On accident? 😂
Well no, I will not be silent and not speak my mind and accept you telling me:
"Oh if you asked 1/3 of their ask then you can speak."
Seriously, what kind of logic is that?
You shouldn't be silent but you should also not expect others to be silent or to not express disagreement. You are free to speak your mind but he who talks is also talked to.
The logic that projects and leaders who ask for more budget get asked more questions and confronted more, and by asking for more and being more socially active you open the door to what you may feel is invalid or unfair criticism.
Youre projecting.
That has literally nothing to do with you telling me not to speak because Magi is asking for more money than Ecency.
I never objected to any questions.
Keep up the great work guys 👍 !BBH !PIZZA
$PIZZA slices delivered:
@day1001(1/10) tipped @ecency
Send $PIZZA tips in Discord via tip.cc!
keep winning and saving
Excelente tecnología vamos avanzando
Hey, ecency! It’s been a while. When I reinstalled the app today, I was truly amazed by all the updates I observed while trying to scroll what’s up. Yay! It’s indeed a job well done.
Good follow up, although I believe that DAO funding should be more agile and restrictive nowadays. The world is changing and reshaping with AI, thus crazy budgets need to be revised.
Agreed - AI is reshaping cost structures across the board, and DHF funding norms should evolve with that reality. Worth a broader community conversation on what "right-sized" looks like in 2026 with proper transparency reports
The system has been so abused that it's now difficult to get proposals approved. Apologies for this personal reflection.
I think more emphasis should be placed on being self-sustaining or even contributing to the overall ecosystem. Advertising should be an option, even if it contributes a minimal amount. It's a scalable system; if we attract more users, we'll earn more from advertising.
The costs seem very high to me. Hive is global, so I don't see the reason to pay Western programming or other prices.
In short, we need proposals that contribute, not those that consume resources. If Hive surpasses one dollar, then we can reconsider our vision.
At Hive Open Days I have met so few people that were against ads on Hive, if the choice is withering away or getting ads the choice is obvious, they would choose ads.
I am actually yet to meet someone, apart from Hive frontend devs, who oppose ads.
It is not as if we are not giving away our data for free with Hive. Anyone can train any AI or run a node and get all the onchain data for free.
We post where we are and what we are doing and about our hobbies and any advertiser outside Hive can cross reference and target us without giving back anything at all to Hive. Why not get rewarded while at it?
Since SEO came up again.
Try to fit in for bigger updates some cheap press release.
Add a " Blog or Update" section on ecency ( only for SEO, static pages).
And well maybe some basic guest posts on Blogs/ websites that are crypto or Freedom of speech related ( and that cost 0 if done right with the right way to outreach and be nice, only time).
This way one front end could show up in search since nobody does here Offpage work.
I know nobody cares and people say it for years and i bacame tired too to mention it.
Remember the marketing proposal? years ago? " we hire expert agency top 10 worldwide".
Just for Info, top 10 agencies are private and do marketing for 1 client only and next it was a waste. Not a single mention from other websites to hive or what hive does. Something evergreen.
So sad it is, this is untapped long hanging fruit potential bigger as the waste of all proposals that never attract 1 new user.
Appreciate this and you're right that offpage SEO is the biggest underexploited lever on Hive.
On our side, the onpage/technical foundation is in good shape - CDN caching for content delivery (fast page loads, consistent
crawler response times), clean SSR markup on Vision-Next, canonical tags, sitemaps. That's the easier half.
The harder half - which is what you're pointing at - is external authority. We've started something there: the weekly Ecency news
digest gets cross-posted to Medium specifically as a backlink play, slowly building external signals over time. It's not fast, but it compounds. Guest posts on crypto and free-speech blogs is a logical next step, fully agree the right approach is patient outreach and being genuinely useful, not paid placements.
On the static "Blog/Update" section - good idea, we can look at a dedicated update/changelog page on ecency.com that's clean markup for crawlers and links well from the homepage. Most of our update content lives in Hive posts which are indexable, but a consolidated static page would help.
Marketing with big agency spend rarely beats consistent low-cost ongoing work. The path you're describing (cheap, consistent, evergreen, relationship-based) is the right one. Will keep pushing on it.
@ecency The infrastructure transparency is the part I appreciate most — it's rare to see a project break down actual operating costs vs. proposal ask. One gap I'd be curious about: how does the mobile user growth curve track against infrastructure cost growth? If onboarding accelerates 2-3x in the next year, does the current ask still hold, or does the unit economics change at scale?