Plain answers to practical questions.

Dutchtown Research Laboratories studies how people move from information to judgment, then builds software that makes that process easier to inspect, challenge, and improve. Shield™, Cite™, and Steward™ are separate products built for different kinds of uncertainty. Sentinel™ is the shared reasoning architecture beneath them, not a standalone app.

Reasoning is how we get from information to a conclusion. In plain language, it means separating what you know from what you assume, weighing the evidence, considering other explanations, noticing what is still unknown, and deciding what follows.

No. Shield is currently being built for moderation work generally, not for one specific platform. It can fit volunteer moderators, small community owners, creator communities, forums, Discord servers, Reddit communities, gaming communities, marketplaces, and professional moderation or Trust & Safety teams. The common role is moderation: someone is responsible for reviewing reported or contested content and deciding how community rules should be applied.

Shield is currently formatted around online moderation workflows. A moderator reviews a report, the surrounding context, relevant rules, and available evidence. Shield helps structure that material, surface uncertainty and competing interpretations, produce The Finding™, and preserve the human moderation decision and its reasoning.

During moderation of an online community or platform: for example, when a reported post, comment, message, account, or pattern of user behavior needs review. Shield is useful when context matters, moderators disagree, the rule is not obvious to apply, or the decision should be documented for later review.

Not at present. Sentinel's underlying reasoning concepts may be adaptable to other governed decision environments in the future, but the current Shield product is being designed and described as a moderation system. DRL does not present future adaptability as an existing capability.

Use Cite when the important question is not just ‘what information can I find?’ but ‘what does the evidence behind this claim actually support?’ Examples include checking a viral claim, tracing a quotation, reviewing sources for a paper or article, comparing conflicting reports, investigating whether several sources are genuinely independent, or determining whether repeated coverage all comes from the same original source.

Cite is intended for people doing evidence and source analysis: researchers, students, journalists, writers, analysts, investigators, educators, creators, policy teams, and people doing serious personal research. The depth of the interface may differ by user, but the core job remains evidence, provenance, corroboration, contradiction, and uncertainty.

Cite does not declare ultimate truth and is not an autonomous investigative authority. It evaluates the connection between claims and available evidence, shows where that evidence came from, exposes weaknesses or conflicts, and states what the record can and cannot support.

Steward is being designed for private reflective reasoning. A person uses it when something happened in their life and they want to understand the situation before deciding what it means or what to do. It helps separate facts, interpretations, emotions, assumptions, unknowns, values, and options while leaving the decision with the user.

After an argument, before an important conversation, while making a difficult work or relationship decision, when journaling, when angry or uncertain, or whenever someone wants to slow down the jump from ‘this happened’ to ‘this must mean X.’ Steward is not meant to replace the user's judgment; it is meant to make that judgment more deliberate.

No. Steward remains in development. Public descriptions of Steward explain its intended direction and design principles, not currently available functionality.

No. They are purpose-built reasoning environments. Shield is currently centered on moderation, Cite on claims and evidence, and Steward on personal reflection. They may use conversational interfaces where useful, but their purpose, workflow, safeguards, and outputs are more specific than a general question-and-answer chatbot.

Yes, within the boundaries of each product's purpose. DRL can adapt reading level, explanation depth, workflow detail, interface complexity, professional terminology, language, and accessibility without quietly changing the underlying reasoning standard. Adaptability does not mean pretending the product already supports every possible domain.

Sentinel is DRL's shared intelligence and reasoning architecture. Its systems preserve evidence, structure reasoning, control how lasting learning occurs, and protect the boundaries between analysis and human authority. Shield, Cite, and Steward use governed access to the Sentinel capabilities appropriate to their purpose.