About Dostupno

Dostupno helps teams turn accessibility requirements into work that can be planned, checked and repeated. It connects user needs, design and development practices, assessment protocols, evidence and repair tasks.

Three people review one interface using a keyboard, speech output and magnification.

Scope

The methods cover websites, electronic documents, mobile applications, desktop software, digital services and self-service terminals. They also include clarity and cognitive accessibility, because technical conformance alone does not prove that a process can be understood and completed.

Common core and national profiles

Web Content Accessibility Guidelines (WCAG 2.2) at levels A and AA provide the international web baseline. Relevant platform guidance and established testing practice supplement that baseline. GOST requirements and Russian regulatory or procurement references are one national profile. Other countries and organisations should apply their own legal, language and procurement context.

Readiness and evidence

The materials are working methods intended for discussion and piloting. A result is meaningful only when it names the tested version, sample, tools, manual checks, people involved and known gaps. An automated result is evidence about the tested sample, not certification of an entire product.

Where to continue

Use the guide to choose a workflow, the sources to understand the basis of the methods, and the evaluation method to plan an assessment.

Purpose and public value

The goal is to help a person independently complete a practical task: find information, read a document, fill in a form and obtain a result. Service owners get evidence of a barrier; developers get a requirement and an acceptance check. For the project team, the open methods form a reusable base of skills and requirements that can be applied to its own products and shared with other developers.

Support and the next pilot

Baseline development and operation use the team's own available resources. Maintenance covers fixes, repeat checks, source updates and feedback. Independent assessment, session moderation, participant support and remediation are planned separately for the next stage.

The first iteration is planned as 6–8 individual sessions lasting 30–45 minutes. Participants will carry out concrete actions in a safe copy of the form with a prepared website address; no real requests will be submitted. Each session will use 3–4 suitable tasks: find the “Check website” button, find the website-address field, identify the “Submit” button and press it in the safe copy, understand the error and confirmation messages, read an on-screen label or find a contact. Participants will not analyse WCAG criteria, methods, reports or technical specifications.

Recruitment aims for these profiles: 1–2 blind participants or screen-reader users; 1–2 deaf or hard-of-hearing participants; one low-vision participant; one keyboard or alternative-input user; and two people who benefit from clearer wording and attention support. Needs may overlap, and no diagnosis is required.

A prepared researcher from the team or an external specialist will moderate; no individual has been selected yet. A second observer records task completion, barriers and assistance. Before recruitment, the accessible communication format will be agreed according to each participant’s preference, for example text chat or a Russian Sign Language interpreter. A blind participant will use a screen reader to find the button and field; a low-vision participant will read a magnified label and check whether the error message is noticeable; a keyboard or alternative-input user will perform the scenario without a mouse. Deaf and hard-of-hearing participants will check whether visual and text-based messages are understandable without sound.

Participants may stop at any time. Recruitment and sessions have not yet taken place. Report clarity for a service owner and whether a developer or AI can implement a technical specification are separate checks and are not part of this user pilot. The pilot also does not validate all protocols or provide a statistically sufficient assessment of the entire site.