Price captures exchange value. The infrastructure a civilization depends on often sits outside that ledger.
That is the starting point of a long TechFlowPost essay by Fugui, published on Sept. 14, which argues that crypto in 2026 is split down the middle. On one side, real-world assets have climbed into the tens of billions of dollars, Wall Street money is lining up for tokenized Treasuries, stablecoins keep expanding, ETFs have been approved, prediction markets are busy, and market participants can still rely on familiar scoreboards such as TVL, fees, cash flow, P/E, NVT and DAU. On the other side, a different class of assets and projects is quietly breaking down: open-source tooling, security work, public goods infrastructure, DeSci experiments and protocols that generate value for ecosystems without cleanly turning that value into cash flow.
A market with money, but not for the things it runs on
The essay opens with the Ethereum Foundation’s own retrenchment. In February 2026, the foundation published a blog post titled 「This Is Fine (Until the Grant Runs Out)」 and introduced Project Odin, a plan aimed at teams that had previously received large Ethereum Foundation grants but no longer had a clear path to follow-on funding. In June, the foundation said it would cut about 20% of staff, with 54 people leaving. Vitalik also stepped back from the core decision-making layer, while the foundation said it would shrink and stop trying to act as the ecosystem’s central manager, recasting itself as a more ordinary participant focused on CROPS: censorship resistance, openness, privacy and security.
The point in the essay is blunt. One of the most important blockchain ecosystems in the world is struggling to sustain the public goods it relies on.
It treats that not as an isolated event, but as part of a wider pattern. Security researchers may save ecosystems from losses worth hundreds of millions of dollars by finding critical bugs, yet the people who find those bugs do not necessarily earn rewards on that scale. WEBCAT, an open-source tool that helps browsers verify code integrity, received funding from the Ethereum Foundation in July. It is difficult to answer a simple investor question around a project like that: how exactly does it make money? The closer a tool gets to pure public infrastructure, the harder it becomes to charge each beneficiary.
The article then runs through a string of examples. Gitcoin Grants has distributed tens of millions of dollars, but GTC has fallen more than 99% from its historical high. Helium once had more than 1 million hotspots, covered more than 170 countries, saw Helium Mobile expand quickly into the hundreds of thousands of users, and recorded a noticeable rise in Data Credit burns, with net deflation appearing in certain quarters. Yet by mid-2026, Helium Mobile had been acquired by Noble Mobile, and the revenue mix shifted toward carrier offload, with consumer subscription revenue no longer flowing directly back to HNT.
VitaDAO has funded 31 longevity research projects and holds IP assets worth tens of millions of dollars in its treasury, but VITA holders have never received dividends. Friend.tech packaged attention into Keys, generated tens of millions of dollars in fees early on, and later saw both TVL and activity collapse, leaving many Keys effectively worthless.
The essay’s summary is simple: real-world assets are trying to find cash flow, meme tokens are trying to find attention, and almost nobody is trying to establish a price discovery framework for public goods.
The problem is not just funding. It is missing measurement and incomplete disclosure.
The piece argues that blaming all of this on market cycles misses the point. Even in the 2021 bull market, GTC was still falling, and open-source developers still relied heavily on mission-driven work rather than durable funding. Nor does the author put the blame on venture capital. Venture capital is supposed to seek risk-adjusted returns. That is its job.
The essay says crypto is facing two more fundamental failures.
First, it has no measurement tool built for public goods. Traditional finance asks projects for revenue, EBITDA, cash flow, TAM, CAC and LTV. Those questions can get a market close to a valuation range for a company. Public goods projects often fail every one of those screens without being low-value. An open-source cryptography library may discover a bug that could have prevented $1 billion in losses and still show zero revenue. A DeSci project may fund 30 longevity studies, publish papers, open data and improve life expectancy over the long term, while showing zero protocol income today. A decentralized social protocol may refuse to sell ads or take a cut of user relationships and also report zero revenue. In those cases, familiar models like P/S, P/F, DCF or MV=PQ are not merely underestimating the project. The formulas break because the underlying economic flow is missing by design.
Second, the ledger itself is incomplete. The article argues that many projects that claim to build public goods provide surprisingly weak disclosure. Once a grant is received, where did the money go? How much do core contributors earn? How large are the off-chain assets, including lab equipment, clinical datasets or IP licenses in DeSci? What stage has the research reached? Were funded papers actually produced? Were contributor rewards distributed fairly or captured internally? These are often impossible to verify on-chain, and if teams do not publish them voluntarily, outside capital has little basis for comparison or follow-up support.
DeSci is presented as the sharpest example. On-chain governance and IP-NFTs may look transparent, but lab operating costs, data quality, labor inputs and failure rates all sit off-chain. An investor or donor may want to keep supporting the work, but without even a usable quarterly report, caution replaces conviction.
The essay also warns against over-trusting on-chain metrics. TVL can be inflated. Active addresses can be manufactured through Sybil activity. Volume can loop back on itself. A high NVT does not automatically mean a bubble, and a low NVT does not automatically mean undervaluation. If an open-source library is called by 100,000 downstream projects but those calls do not create on-chain transactions, NVT sees none of it. If a security tool protects the ecosystem by preventing failure, its value may remain invisible until something breaks.
The result is a deeply awkward disconnect. Meme projects are easy to score because attention can be counted, even if social value is close to zero. DeFi is legible because TVL and fees are visible, while the harder question of financial accessibility is barely measured. DePIN can show revenue, but coverage in remote areas and service resilience attract less scrutiny. DeSci tokens can trade every day while real research outputs such as papers, datasets and clinical milestones go largely untracked. DAOs can expose treasury balances while governance quality and coordination efficiency remain hard to measure. Decentralized social networks can post DAU growth without proving that data sovereignty has truly moved to users. Open source is hit hardest of all: revenue may be near zero, while system-wide dependency is the metric that actually matters.
The article’s core claim is that markets have developed a habit of treating price as proof of value. Whatever is countable and monetizable enters valuations quickly. Shared social value that does not pass directly into cash flow is usually left outside the frame.
Traditional finance has frameworks. Crypto has barely imported them.
From there, the essay turns to the traditional world and argues that tools for this problem already exist, even if crypto pays little attention to them.
It starts with ESG. The author notes that Digital Asset Research expanded the idea into ESGN by adding a Network Health dimension tailored to digital assets. That framework covers more than 250 indicators, including energy use, developer distribution, token holder concentration, code quality and decentralization. ChainScore Labs is cited for another idea, a Sustainability Premium. After Ethereum’s merge reduced energy consumption by 99.9%, the essay says, the network also reduced climate-related regulatory and blacklist risk, which could translate into lower capital costs and better liquidity.
Still, the author says ESG is only an entry point.
GRI, the Global Reporting Initiative, pushes farther by separating financial materiality from impact materiality. A project may not directly boost corporate profit and still have a major effect on people, the environment or the economy. Under GRI logic, that effect should be disclosed. IRIS+, maintained by GIIN, is presented as a more practical toolkit. It offers more than 500 standardized indicators and frames impact around five questions: What happened, who benefited, how much impact occurred, what was the project’s contribution, and what risks remain. The essay argues that this maps neatly onto Web3, especially for DeSci, where claims about scientific output, beneficiaries, scale and causation need to be tested rather than narrated.
SROI, or Social Return on Investment, asks a more direct question: how much social value is created for each dollar invested? For many crypto public goods, financial ROI may be close to zero while social ROI is clearly above zero. The UK Green Book is cited for a broader definition of social value that includes prosperity, justice, safety, climate, environment, health, well-being and distributional effects, without forcing every category into a dollar-denominated output.
The article also mentions the Digital Public Goods Standard used by the Digital Public Goods Alliance. That framework looks at nine dimensions including relevance to the SDGs, open-source licensing, ownership, platform independence, documentation, privacy and security.
What matters in the author’s view is not just that traditional systems have metrics. They also have disclosure rules. Public companies publish annual reports, ESG reports and social responsibility reports. Auditors sign off. Fraud can be punished. By contrast, crypto public goods projects often do not even have a standard reporting format. Teams that got grants write a blog post after the money is gone. Teams that did not get grants may not even do that. The article stops short of demanding public-company-style audit burdens for every project, but it insists on a basic format for financial, impact, governance and risk disclosure. Without that, neither donors nor investors can make side-by-side comparisons.
Social value is not token value
The article then adds an important warning. A high ESG score or a high social impact profile does not mean a token should rise.
Its transmission chain is longer than that. Social value is not the same as enterprise value. Enterprise value is not the same as protocol value. Protocol value is not the same as token value. Break the chain at any point and the value does not arrive.
The DeSci example is used again. A project may create $100 million in social value in a year by opening research, shortening research cycles and increasing collaboration. But if users do not need to hold the token, researchers do not need to stake it, the token is not used in service payments, the treasury does not capture the value, governance does not control budget rights, and the token is not a gate to data access, then there may be almost no relationship between that social value and the token’s market capitalization. The essay labels that situation as one where α, the Public Value Capture Ratio, is near zero. A project can be a strong public good and still be a weak token investment.
The reverse can also hold. In a DePIN network, rising social demand can lead to more service usage, more staking demand, more token burn, stronger security needs and a larger treasury. In that case, social value can actually travel into token demand. The article says that chain can be observed in functioning DePIN systems, but it can also be structurally severed. Helium Mobile’s acquisition by Noble Mobile is the example given: once subscription revenue stopped flowing back to HNT, both α and β were damaged.
That leads to the article’s three key questions: how much social value does a project create, how much of that value is internalized by the protocol itself, and how much of the protocol-level value is actually borne by the token? The second question is α. The third is β, the Token Capture Ratio.
The essay is careful not to overstate precision. It says α and β cannot be calculated exactly in the real world. Shadow pricing choices, attribution boundaries and counterfactual assumptions can push estimates apart by 2x to 3x, and token design details can move β significantly. Anyone claiming to have measured α at 0.37 and β at 0.62 with precision is either confused or selling something. The right output is a range, adjusted for risk and updated against real data.
VitaDAO is offered as one case. The essay says VITA holders have never received dividends, and the DAO’s own guidance says VITA is not an economic instrument. Scientific research operates on decade-long cycles, while token liquidity lives on six-month cycles. That mismatch keeps α low. Friend.tech is another case. Its β looked high for a period because attention had been turned directly into a tradable object, but the mechanism eventually undermined the social layer by making relationships zero-sum. Farcaster and Lens, both of which have chosen not to issue tokens, are used to support the point that social graphs as public goods can be damaged rather than strengthened by forced tokenization.
Filecoin appears as a more constructive example. In September 2026, Filecoin launched ProPGF Batch 3 and awarded about $1.4 million to 19 projects. The essay treats the key innovation not as the grant amount alone, but as the Filecoin Kernel framework, which divided public goods into Essential and Important categories and centered evaluation on a cold, practical question: if this project disappears, does the Filecoin network actually get hurt?
That, the author argues, is what mature protocol thinking looks like. Public goods are not a moral side issue. They are a budget issue, a governance issue, a prioritization issue and a disclosure issue.
A six-layer framework from cash flow to market pricing
The article’s proposed answer is not another ESG scorecard. It is a full-stack framework for turning non-economic value into something engineered, auditable and potentially transmittable into tokens. The name is described as flexible, whether PIVF or PVATV, but the architecture is fixed: six layers in strict sequence. If one layer fails, the next one should not be modeled as if the chain were intact.
Layer 1: Economic Value
This layer tracks cash flow, revenue and usage value: annualized revenue, protocol fees, treasury income and PSRF, or Protocol Sustainable Revenue Floor. PSRF is defined as the annual funding needed to keep a project alive, including core contributor compensation, key infrastructure spending, security audit reserves and inflation buffers. If annual income exceeds PSRF, the project can sustain itself. If it does not, the project faces supply risk. The article says every project should publish quarterly revenue breakdowns, spending breakdowns, treasury balances and PSRF coverage ratios. That is the minimum standard for transparency.
Layer 2: Public / Social Value
This layer captures externalities, public goods output and social impact. The article asks how many downstream projects depend on an open-source library, how many real research programs a DeSci protocol is funding, how many underserved areas a DePIN network reaches, and how many users’ real data a privacy tool protects. It suggests using Theory of Change to map the line from inputs to activities, outputs, outcomes and impact, then tailoring auditable metrics from IRIS+ by sector. For DAOs, that can include active contributor addresses, proposal execution rates and funding use reports. For DeSci, funded projects, IP-NFT counts, open-access papers, experiment progress reports and off-chain asset inventories. For DePIN, active nodes, geographic coverage, real data utilization and service-level metrics. For open source, downstream dependency counts, critical bug fixes and code reuse. Every metric needs a source. If it is on-chain, point to the contract. If it is off-chain, provide raw documents or independent audits. No source, no score.
Layer 3: Impact Attribution
The article calls this the easiest layer to skip and the easiest one to game. Value may exist, but was the project responsible for it? Five questions have to be answered one by one: what happened exactly, who benefited exactly, how much remains after deadweight is removed, what share of the outcome belongs to this project relative to other actors, and what negative side effects came with it? Sybil-heavy activity should be discounted. Outcomes that would have happened anyway should be removed. Value displaced from another project should not be counted. Cooperative outputs should be split by contribution. Serious negative effects should trigger hard red lines instead of being offset by positive public-good claims.
The essay warns that if token buybacks or treasury allocations are tied directly to counts of funded projects, paper counts or active addresses, projects will optimize for quantity and manipulation. The safeguards it lists include multidimensional cross-checks, independent third-party audit sampling and dMRV, with raw data anchored on-chain and open to challenge and slashing. Without this layer, the framework becomes a narrative machine wearing a measurement costume.
Layer 4: Protocol Capture
This is where α enters. Once social value has been identified and attribution has been made, the next question is how much of that value can be internalized by the protocol. The article says α depends on mechanism design: whether service payments must use the token, whether validators must stake, whether the treasury has recurring revenue, whether burn mechanisms are actually active, and whether governance can change real economic parameters. Sequencer revenues, MEV distribution and protocol fee switches are all described as α-adjusting levers. α is not a fixed constant. It is a range that moves with protocol parameters.
Layer 5: Token Capture
The fifth layer addresses β. Even if the protocol captures value, how much of that reaches the token? The answer depends on token design. Does the token have explicit value-capture rights? Does treasury income flow back to holders? Are buyback-and-burn mechanisms hard-coded in contracts? Are staking yields driven by real revenue rather than inflation subsidies? Can governance tokens actually decide budget allocations? Here again, the article rejects point estimates. β is a range. Many projects appear to have token capture, but if treasury multisig control sits with the team, buybacks can be canceled at will, or staking returns are mostly inflationary, β should be marked down sharply.
Layer 6: Market Calibration
Only after the first five layers are complete does the market enter as a calibration step. This layer keeps native crypto metrics such as NVT, MVRV, treasury NAV, market cap per active developer and net deflation status. If the value range produced by the framework differs from market capitalization by more than 50%, the article says there are usually three explanations: the market has not recognized the value yet, the shadow-pricing assumptions are too optimistic, or the project carries a critical risk the model missed, such as regulatory attack or team failure. The market should be respected, the author says, but not worshipped.
The final output should therefore be a range, not a precise price. The article also suggests three diagnostic indicators: PSRF coverage ratio, impact-to-market-cap ratio, and a composite risk discount rate. All data inputs should be sourced, assumptions should be disclosed, and confidence intervals should be included.
How to build the bridge
The article’s next question is practical: if non-economic value is real, how is it injected into token value? The answer is not automatic repricing. It requires institutions and mechanisms.
The first bridge is public-goods funding flows. Quadratic funding and RetroPGF are already operating at scale, the essay says. Gitcoin has distributed tens of millions of dollars through quadratic funding. Optimism’s RetroPGF, before its pause, went through multiple rounds and allocated funding on the scale of hundreds of millions of dollars. Open Source Observer, or OSO, unified GitHub, npm and on-chain deployments into an auditable impact pipeline and, before the pause in RetroPGF, had already been used to allocate tens of millions of OP. Once funding reaches project treasuries, it can support buybacks, contributor rewards or product adoption, which can then improve economic value. In that path, the token acts both as a funding currency and as a value scale.
But the article says there is a harder obstacle: the free-rider problem. Even if every institution agrees that an open-source security library created $1 billion in social value, no single user has a strong incentive to pay for its upkeep if the tool remains free. Let someone else fund it, and everyone else can keep using it. The essay frames this not as a moral failure, but as a classic prisoner’s dilemma.
The proposed answer is forced internalization through protocol-level taxes and Pigouvian subsidies. A simple example is given for layer-2 ecosystems: a fixed share of sequencer revenue, for example 10% to 20%, could be routed automatically into a RetroPGF funding pool through a preset contract, without waiting for votes or goodwill. EIP-1559 burns, sequencer fee-sharing and MEV smoothing are all described as examples of protocol-level taxation logic. The upkeep cost of public goods, the article says, should be borne proportionally by the economic activity that depends on them, rather than by individual donors stepping in by choice.
The second bridge is impact certificates. Hypercerts, developed by Protocol Labs, are described as ERC-1155 semi-fungible tokens that record who did what, for whom, when and with what impact. That turns completed beneficial work into a tradable, divisible and attributable on-chain asset. In the essay’s framing, this is the first time social value starts to take on a recognizable asset form. dMRV and dynamic oracles are essential here because impact cannot be trusted as a self-reported number. It has to be fed by ongoing, auditable data.
The third bridge is ESG compliance premium. If a project scores well on ESG-related dimensions, the author argues, it gains access to institutional capital, stronger liquidity and a lower discount rate. With the global ESG asset pool exceeding $30 trillion, that can matter especially for low-margin projects. The sustainability premium discussed by ChainScore Labs is used as the supporting example rather than a rhetorical flourish.
The fourth bridge is attention-led demand, but only when attention is verified as real adoption rather than turned directly into a security-like object, as the article says happened in Friend.tech. The path runs from real attention to user and developer adoption, then to network effects, and finally to token demand through gas, governance or staking. dMRV, oracles and OSO are presented as the tools that can help distinguish real adoption from empty rotation.
The essay also adds a technical point on monetization. Shadow pricing should not be pegged to static fiat values. If impact is always priced in dollars, fiat inflation and crypto market cycles can break the link between measured social value and the actual capital available in token markets. In a bear market, tokens can fall 80% while the dollar value of social impact remains unchanged, leaving builders unable to raise matching support. A more stable method, the author says, would rely on dynamic algorithmic oracles anchored to baskets of real-world purchasing power, such as one hour of mid-level developer labor, the cost of a standard security audit, or the marginal cost of 1 GB-year of decentralized storage. That makes shadow pricing move with productive inputs rather than a frozen fiat figure.
Token design matters, and some projects should not issue one at all
The article argues that tokens themselves need redesign. Governance cannot be ceremonial. It has to affect real economic variables. As an example, the essay cites Uniswap’s UNIfication proposal in December 2025, which turned on the fee switch and burned 100 million UNI with a 99.9% approval rate, with the first burn worth roughly hundreds of millions of dollars in equivalent value. In that framing, governance is not decorative. It is part of the value mechanism.
On utility, the article says the strongest support comes when DePIN payments for compute, storage or networking must use the token, validators must stake it, and burn mechanics create deflation. It also sketches a newer category, impact utility: retrospective impact distributions, impact-based buybacks, and on-chain certification of third-party audits. The article stresses that if buybacks are tied to impact outputs, the metrics must be multidimensional and independently audited. A single gameable metric should not be enough.
Yet the essay is equally clear that many projects should not issue tokens. It points to Farcaster, which remains active with 40,000 to 60,000 DAU and no token, Lens, which still has not issued one, and VitaDAO, where VITA was intentionally designed as a non-economic tool. In the author’s view, there is an internal tension between strong public-good characteristics and token value capture. Where the project is highly public-good in nature, the revenue path is unclear, and the free-rider problem cannot be solved through mechanism design, grants, donations, treasury support or protocol taxes may be the more honest structure.
Force a token where the mechanism is wrong, the article says, and the result is often a meme, a broken team, or a new vehicle for impact washing.
The framework can be gamed too
The last major section turns to the failure modes of the framework itself. The essay says every new system in crypto attracts a second wave of innovation from people trying to exploit it faster than builders can defend it.
The first failure mode is impact washing. Once social value feeds token valuation, projects will have an incentive to buy certifications, fake users, inflate scientific citations, stuff DAO voting, overstate DePIN devices or stage fake social-good activities. If funded project counts, paper counts or active addresses are tied directly to buybacks or treasury distributions, gaming becomes nearly inevitable. The proposed response is to treat impact scoring itself as a protocol under attack. Sybil-heavy activity should be tested with identity clustering and sampling. Read metrics should count only interactions with behavioral consequences. Academic citations should be filtered through DORA-style principles and expert sampling. Airdrop-driven participation should be filtered across 30-, 90- and 365-day windows. Score inputs should come from multiple oracles with stake, challenge and slashing. Double counting needs clearly bounded contribution chains. Severe negative effects should trigger separate red lines instead of being netted away.
The article invokes Goodhart’s law directly: once a metric becomes the target, it stops being a good metric. That is why attribution has to stand alone as its own layer, and why the output should be a range rather than a point estimate.
The second failure mode is subjective shadow pricing. The author says experience from SROI and IWA shows that different valuation methods applied to the same externality can produce differences of more than 3x. Open-source public goods do not yet have reliable monetization coefficients, and dynamic oracle parameter choices create new subjectivities. The realistic goal is a narrower estimate band, not the removal of uncertainty.
The third is attribution difficulty. Public-goods impact is often jointly produced. One open-source tool may be integrated by many projects. A security upgrade to a base library may protect an entire ecosystem. It is inherently difficult to assign that value cleanly to a single protocol. Attribution discounts can reduce overstatement, but not solve the problem completely.
The fourth is market recognition lag. Crypto markets remain heavily speculative, and non-economic indicators may take years to be priced. The essay cites research by Jin and others suggesting that environmental dimensions are priced only after specific events, while social and governance dimensions have little clear event-study support. It also notes the gap between practitioners and academia. Frameworks from people such as Burniske, Woo and Samani circulate mostly in blog posts and industry reports rather than peer-reviewed journals.
The fifth is governance capture. The definitions and weights of impact metrics, the selection of third-party auditors, and the mechanisms for challenge and slashing can all be captured by whales or core teams. The article suggests non-token-weighted voices such as citizen juries or DID-based reputation systems, auditor rotation and supermajority plus timelock requirements for impact-parameter governance.
The sixth is fake transparency. Even if raw data is anchored on-chain, off-chain data can still be falsified. Labs can fake images, nodes can simulate activity in virtual environments, and users can be bought through traffic farms. dMRV can prove that a record on-chain has not been tampered with. It cannot prove that the uploaded data was true to begin with. That still requires real-world verification points, random checks, reputation systems and harsh penalties.
Even so, the author argues these risks are not reasons to do nothing. ESG data systems in the 2010s were also weak and developed over time. There were greenwashing scandals there too. Crypto, the article says, has not even really started the process.
The missing ingredients are a language and a ruler
The essay ends by reducing the problem to two missing pieces: a common language and a common ruler.
Traditional finance has a vocabulary that investors, analysts, auditors and regulators all understand: revenue, EBITDA, DCF, P/E, WACC and beta. It also has a ruler: accounting rules, disclosure requirements, audit processes and regulatory penalties. Those two things let capital move.
Public-goods-oriented Web3 projects lack both. Builders say they are creating something important. Investors ask how much revenue they have. The conversation stops because there is no shared language. Even where a new language begins to emerge, projects often fail to disclose in a standardized way, disclosures are not audited, and audited disclosures are not comparable. In that vacuum, fundraising falls back on narrative, vision, memes and endorsements. The essay says narrative cannot be audited, compared or properly reviewed after the fact. In bull markets, almost any story can attract capital. In bear markets, narrative-led projects are often the first to fail.
The author closes by listing tools that already work in pieces: Gitcoin, Optimism RetroPGF, Hypercerts, OSO, dMRV, SourceCred, ESGN, IRIS+ and SROI. OSO helped allocate tens of millions of OP using impact metrics. Hypercerts turned impact into a tradable asset. Gitcoin’s quadratic funding proved that 100 people donating $1 each can outweigh one person donating $100. Filecoin Kernel, debates around L2 sequencer revenue sharing and EIP-1559 burns all point in the same direction.
What is still missing is the layer that strings those pieces into a standard pipeline: an integrated framework that starts with economic value, identifies public value, verifies attribution and anti-gaming, measures protocol capture through α, measures token capture through β, and then calibrates against the market. The purpose is not to eliminate financial metrics. It is to fill the gap when financial metrics are zero or distorted. The purpose is not to replace audits. It is to embed disclosure standards from day one, including what must be disclosed, in what format, with what data source, under whose audit, and with what penalty for fabrication.
The essay also reaches back to older theory. It references Coase’s 1960 argument that when transaction costs are too high for private markets to internalize externalities, new mechanisms are needed. Blockchain, in the author’s telling, is itself a device for lowering transaction costs. Quadratic funding was mathematically shown by Buterin, Hitzig and Weyl to optimally aggregate public-good preferences. Gitcoin has run versions of that logic for years. A crypto version of a Coase-style theorem only works, however, if externalities can be measured, assigned, traded, disclosed and audited.
One last number appears near the end. Citing Hoffmann and others, the article says the demand-side value of open-source software is about $8.8 trillion, and that every $1 invested in MLOSS corresponds to at least $100 in global economic value. Those gains are currently ownerless in market terms: consumed by society, weakly maintained, thinly funded and poorly priced.
The article returns to the concrete problems of 2026. The Ethereum Foundation is shrinking. Security researchers are living grant to grant. DeSci projects operate on 10-year scientific cycles while token liquidity runs on six-month cycles, and their ledgers remain opaque. Decentralized social networks hesitate to issue tokens at all. Behind each example is the same break: value is being created, but it is not returning to the people who create it in a form the market can recognize and trust.
That is why the essay finishes with the question raised by Filecoin Kernel: if a given project disappears, does the network actually get hurt? The author says the broader crypto industry should ask the same thing of its public goods layer. If those projects die, Web3 gets hurt. Real-world assets need secure base protocols. Meme markets need functioning chains. ETFs need live nodes. AI agents need decentralized storage and compute. Every polished commercial layer rests on open code, basic infrastructure, security research and open data that do not always monetize directly.
The final claim is not that sentiment should replace valuation, or that a new ESG rhetoric should replace old meme rhetoric. It is that crypto needs verifiable data, auditable indicators, comparable frameworks and enforceable disclosure standards that can translate social value into a language capital understands, force the free-rider cost back into protocol design, and lock gaming risk inside mechanism design rather than leaving it in marketing.
Price records exchange value. Open-source code, security research, open science, decentralized communication and privacy protection are worth more than what price alone can record. In the article’s view, a mature industry is not defined by how many multiples it can rally, but by whether it can honestly support the systems that keep it running and whether it can build a mechanism that prevents everyone from pretending those systems do not exist.