{"slug": "building-a-knowledge-network-in-public", "title": "Building a knowledge network in public", "summary": "A developer built Stadan, an open-source knowledge network software that integrates facts, opinions, and intentions into a single graph. The project also includes SteveCare, a commercial SaaS that analyzes public web data to uncover trends and relationships. The goal is to transform information into understanding rather than monetizing attention.", "body_md": "Over the last few years, I've spent a lot of time thinking about search engines, social networks, AI, and how organizations discover information. The more I thought about it, the more I realized that we have become incredibly good at storing data, but we still struggle to understand it.\n\nSearch engines help us find facts. Social networks expose opinions. Companies accumulate enormous amounts of internal knowledge, while AI models are becoming increasingly capable of generating answers. Yet all of these systems remain largely disconnected. Information is abundant, but understanding remains scarce.\n\nLet me explain...\n\nI believe there are really three kinds of knowledge. Facts answer the question \"what exists?\", opinions answer \"what do people think?\", and intentions answer \"what do people want?\". Most systems specialize in one of these categories. Search engines focus on facts, social networks expose opinions, and support systems or CRMs contain intentions. Unfortunately, these intentions usually remain trapped inside organizations.\n\nI believe all three belong in the same graph. That idea eventually evolved into three projects that reinforce one another.\n\nStadan is the open-source software that provides the building blocks required to understand facts, opinions, and intentions. Through API connectors, crawlers, embeddings, graph databases, vector databases, plugins, and AI models, it provides the infrastructure required to transform information into understanding.\n\nThe project is designed to be self-hosted, extensible, and open. Communities, companies, and individuals can use it to build their own knowledge systems while contributing improvements back to the project. Rather than being a source of data, Stadan is the software that enables knowledge systems to exist.\n\nSoftware alone is not enough. To create value, data must be collected, transformed, enriched, and accumulated over time. That is the role of SteveCare, a commercial SaaS built using the Stadan open-source software.\n\nSteveCare contains data and historical intelligence. Its purpose is to continuously analyze public information coming from websites, RSS feeds, sitemaps, GitHub, Reddit, YouTube comments, forums, blogs, podcasts, job postings, and many other sources.\n\nOver time, SteveCare accumulates knowledge about companies, technologies, products, communities, and discussions. By connecting these relationships together, it becomes possible to discover trends, opportunities, competitors, partnerships, candidates, and emerging markets. A blog post mentioning a technology may lead to a GitHub repository. A YouTube video may generate discussions on Reddit. A podcast episode may reveal a trend months before it appears in surveys. Job postings may indicate which technologies companies are adopting before they officially announce them. In that sense, the web itself becomes a graph.\n\nTo make this possible, the platform relies on APIs whenever they are available. RSS feeds and sitemaps become the primary discovery mechanisms, while pages are crawled only after being discovered through those structured sources. The objective is not to blindly scrape the web, but rather to leverage the information websites already expose about themselves.\n\nAnalyzing websites reveals much more than their content. Technologies, frameworks, infrastructure providers, analytics tools, payment systems, and authentication platforms can often be inferred from the pages themselves. Combined with information coming from job postings and public discussions, this creates a surprisingly rich picture of companies, products, and ecosystems.\n\nSearch engines and, to some extent, the web ecosystem itself have become broken. Much of today's internet revolves around buying eyeballs in order to display more advertisements. Websites compete for clicks, platforms compete for attention, and users are increasingly overwhelmed by noise.\n\nMy goal with SteveCare is different.\n\nRather than monetizing attention, I want to monetize intelligence.\n\nBy understanding the relationships between companies, products, technologies, communities, and discussions, SteveCare can help organizations discover new leads, identify potential partners, uncover emerging markets, and better understand their competitive landscape.\n\nOver time, I envision SteveCare helping B2B companies automatically match their products and services with organizations that could benefit from them. It could assist enterprises in creating better proposals, identifying pain points, and discovering opportunities that would otherwise remain hidden.\n\nIn other words, I believe the transformed knowledge is ultimately more valuable than the raw data itself.\n\nBecause SteveCare is monetized through the intelligence it generates for businesses, I also hope to build a search engine that contains no advertisements. If the platform is already sustainable through its B2B insights and reports, there is no need to optimize for clicks or attention. Instead, the search experience can focus entirely on helping people discover relevant information.\n\nOver time, I believe the reports, recommendations, and relationships generated from this accumulated knowledge may become more valuable than the underlying data itself, because they transform information into understanding.\n\nWebX Protocol: Trust and Federated Intelligence\n\nWhile SteveCare focuses on public knowledge, WebX Protocol represents the long-term vision for private knowledge.\n\nOrganizations possess enormous amounts of intelligence hidden inside support tickets, emails, internal forums, chat rooms, documentation, and knowledge bases. In many cases, these signals appear months before they become visible to the public. People rarely complain on Reddit before complaining to support. They do not ask Hacker News for features before asking the vendor, and they do not write blog posts about their frustrations before opening tickets. Support systems often contain signals that foreshadow future trends.\n\nOrganizations participating in the WebX Network may run Stadan privately on their own infrastructure. This allows them to analyze their internal knowledge while keeping raw conversations, customer information, and confidential documents entirely private. Stadan could be connected to virtually any internal system, including ticket systems, ERPs, CRMs, CMSs, emails, chat platforms, documentation portals, and knowledge bases.\n\nInstead of sharing data, organizations could choose to share anonymized and aggregated intelligence. Rather than exposing that a specific customer requested a feature, they might expose that demand for HIPAA support is increasing in healthcare, that interest in AI workflow automation has accelerated significantly, or that migration away from a competitor is becoming more common. In effect, organizations would exchange intelligence rather than data.\n\nFor example, a company like Salesforce could ask the network questions related to the experience users are having with its software. Organizations running Stadan privately could analyze the support tickets, documentation, internal discussions, and CRM interactions available within their own environments and contribute anonymized insights back to the network. Salesforce would not receive customer conversations or confidential information. Instead, it could receive aggregated intelligence indicating that users are struggling with a new feature, that sentiment toward a recent release is improving or deteriorating, or that a specific workflow is generating a growing number of support requests.\n\nThis would allow software vendors to discover problems and opportunities far more quickly than waiting for complaints to appear publicly. In many cases, organizations using the software know about issues long before they become visible on social networks, forums, or review websites.\n\nWebX Protocol provides the trust layer required to make this possible. The blockchain itself is not intended to store knowledge. I believe organizations running their own self-hosted Stadan instances are far better suited for that purpose. Knowledge remains under the control of the organizations that generate it, while WebX Protocol focuses on enabling trust and value exchange between participants.\n\nOrganizations establish trust relationships with one another, and reputation, complaints, votes, and historical accuracy become part of a trust graph that evolves over time. If Salesforce trusted Microsoft at fifty percent, and Microsoft trusted a startup at fifty percent, then Salesforce might indirectly trust that startup at twenty-five percent. Companies could also choose to share trust relationships with one another, allowing webs of trust to emerge naturally throughout the network.\n\nIn many ways, WebX Protocol is less about cryptocurrency and more about credibility. The token itself does not exist for speculation. Its purpose is simply to simplify micropayments between organizations exchanging intelligence. Rather than requiring thousands of bank transfers, a single transaction could compensate an entire ecosystem automatically.\n\nUltimately, WebX Protocol is not designed to decentralize knowledge itself.\n\nIt is designed to decentralize trust.\n\nAlthough the long-term vision spans three projects, I have no intention of building everything at once. In reality, much of the underlying work has already been done. Many of the ideas, components, and architectural decisions have been evolving for years through previous projects and experiments. I plan on continuing to develop Stadan, SteveCare, and WebX Protocol simultaneously, but in successive layers.\n\nSince Stadan provides the open-source software foundation, it needs to exist before anything else. My initial focus is therefore to release the core platform and make it usable, both for myself and for anyone interested in self-hosting it.\n\nOnce the platform reaches a usable state, I will release SteveCare, which is built using the Stadan software. Initially, SteveCare will focus exclusively on the technology ecosystem. By continuously analyzing technology-related websites, blogs, communities, GitHub repositories, podcasts, videos, forums, and job postings, I hope to build intelligence around technologies, companies, products, and trends.\n\nMy goal is to use SteveCare to promote SteveCare. By publishing interesting findings, reports, articles, videos, and newsletters generated from the intelligence gathered by the platform itself, I hope to demonstrate its value publicly.\n\nInitially, I intend to sell access to the SteveCare API and reports to B2B startups operating in the technology sector. Over time, as customers begin using the platform, I expect its reach to naturally expand toward the industries and markets they care about. Rather than attempting to crawl the entire web from day one, I want the platform to evolve according to the needs of its users and customers.\n\nBecause SteveCare is monetized through intelligence rather than attention, I also intend to release a free search engine with no advertisements. Instead of optimizing for clicks and impressions, the goal will simply be to help people discover information and relationships using the intelligence already accumulated by the platform. If the business is sustainable through B2B insights, reports, and APIs, then there is no need to monetize users through advertising.\n\nAs the communities around Stadan and SteveCare grow, I hope to form the WebX Protocol with organizations that are already self-hosting Stadan. Participation in the network would always remain optional. Organizations that wish to exchange anonymized intelligence and benefit from a trust network could choose to participate, while others could continue using Stadan entirely independently.\n\nIn that sense, I view the three projects as layers built on top of one another.\n\nStadan provides the software.\n\nSteveCare provides the intelligence.\n\nWebX Protocol provides the trust.\n\nTogether, they form a single ecosystem whose purpose is to help transform information into understanding.\n\nI'll be releasing what has already been built, along with the project websites, over the coming days. From that point on, the remaining journey will be built in public.\n\nAs much as the software itself, I believe the ideas, architectural decisions, and lessons learned along the way are worth sharing. That is why I intend to publish articles regularly, explaining the concepts behind the projects, the technical choices that drive them, and the documentation required to understand and use them.\n\nEvery day, I will work on a piece of the platform and then write about it. Those articles will become Dev.to articles and part of weekly newsletters. At the same time, interesting trends and insights discovered through SteveCare will become their own articles, reports, and newsletters.\n\nMy hope is that these articles will help others understand the reasoning behind the architecture, challenge some of the assumptions, and perhaps even contribute new ideas. Whether people agree or disagree with the vision, I believe the discussions themselves are valuable.\n\nOver time, I expect Dev.to to become one of the main places where I share the evolution of the ecosystem. I intend to write frequently about the concepts behind Stadan, SteveCare, and WebX Protocol, as well as the technical choices, experiments, successes, failures, and documentation that accompany them.\n\nMy goal is simply to create useful software, share everything I learn, and allow the project to compound over time.\n\nMost people choose between building, writing, and selling.\n\nI'm trying to do all three simultaneously.\n\nBecause I suspect that, in the long run, trust compounds just like software.\n\nAnd perhaps the most valuable asset we can build isn't code.\n\nIt's credibility.", "url": "https://wpnews.pro/news/building-a-knowledge-network-in-public", "canonical_source": "https://dev.to/steverodrigue/building-a-knowledge-network-in-public-2plg", "published_at": "2026-06-15 05:54:53+00:00", "updated_at": "2026-06-15 06:10:41.907474+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "large-language-models", "ai-infrastructure", "developer-tools"], "entities": ["Stadan", "SteveCare", "GitHub", "Reddit", "YouTube"], "alternates": {"html": "https://wpnews.pro/news/building-a-knowledge-network-in-public", "markdown": "https://wpnews.pro/news/building-a-knowledge-network-in-public.md", "text": "https://wpnews.pro/news/building-a-knowledge-network-in-public.txt", "jsonld": "https://wpnews.pro/news/building-a-knowledge-network-in-public.jsonld"}}