Skip to main content

Overview

Bowrain is a governed, multi-user content platform served from one place — the Bowrain server. Everyone reaches the same workspace through two clients: a browser and a native desktop app. They are equal real-time clients, so the editing experience, terminology, content memory, presence, and reviews are shared across both. Writers and reviewers work here entirely; developers meet the same workspace through the same server runs — what a run parks arrives in the review queue, and what gets approved here feeds the next run. This section describes what the platform does, organized by what you do with it; client-specific differences are noted inline, and the desktop app page covers what is different about the native client (offline editing).

A quick tour

The editor works on content from any connector — a content platform, a design tool, a repository, or a developer's checkout — held on the server, shared across a team and governed over time. These short tours run against a live workspace.

The translation editor

Open a file and it becomes a side-by-side workbench: source and target, a live preview of the page, and — beside each block — the matches from the workspace content memory and the approved terminology. One editor serves every target locale.

Outdated wording

The narration uses terminology that has since been retired. The interface and the behaviour shown are unchanged.

Shared terminology and memory

Terminology and a content memory are often per-person files or spreadsheets. Bowrain keeps them on the server, shared across every project and every translator in the workspace — one governed source of the team's approved wording, searchable and versioned.

Outdated wording

The narration uses terminology that has since been retired. The interface and the behaviour shown are unchanged.

Review and approve

Projects in Bowrain are reviewed by the team together, whichever connector the content arrived through. The Review surface steps through a file one block at a time — source, target, and the status that tracks each block from draft to translated to reviewed — so the team always knows what is left, and approvals are recorded against the workspace rather than a local file.

Checks that learn from corrections

Every correction a team makes is a signal. Bowrain aggregates the phrasings people keep fixing into candidate rules, shows the blast radius of each across real content before it lands, and — on approval — promotes it into a versioned, enforced check. A correction made once becomes a check that stops the mistake from recurring, shared across everyone and every AI tool. See the correction-learning loop.

Collaboration and streams

A workspace is shared by a whole team. You invite teammates by email and give each a role, and streams keep parallel work apart — like branches, a rebrand or a risky pass runs on its own stream, compared and merged back into the main line when it is ready.

Outdated wording

The narration uses terminology that has since been retired. The interface and the behaviour shown are unchanged.

Screenshots

Login

Bowrain authenticates through your identity provider. Depending on configuration, sign-in options include GitHub, Google, LDAP, or other OIDC-compatible providers.

Login page with identity provider optionsLogin page with identity provider options

Workspace Selector

The workspace rail on the left lets you switch between workspaces. Each workspace is an isolated environment with its own projects, content memory, and terminology.

Workspace rail with multiple workspacesWorkspace rail with multiple workspaces

Project Dashboard

The dashboard shows all projects in the active workspace. Each card displays the project name, source and target languages, and file count. Create new projects or open existing ones from here.

Project dashboard with project cardsProject dashboard with project cards

Project View

The project view shows files in the selected project with format detection, word counts, and block counts. Upload new files by dragging them onto the drop zone, or click "Add Files" to browse. Access the content memory and terminology from the header buttons.

Project view with file list and upload zoneProject view with file list and upload zone

Content memory

The content memory browser provides full access to the workspace content memory. Browse, search, add, edit, and delete entries. Filter by source or target locale.

Content memory browser with search and filteringContent memory browser with search and filtering

Terminology Explorer

The terminology panel provides concept-oriented term management. Browse, search, add, edit, and delete concepts with multi-locale terms. Import from CSV or JSON, export the full terms store. Each term has a lifecycle status and domain classification.

Terminology explorer with concept browserTerminology explorer with concept browser

Settings

View workspace information including name, slug, description, and your role.

Settings page with workspace informationSettings page with workspace information

Features

  • Translation editor with Visual and Table views, inline tag support, block status tracking, and keyboard navigation
  • Context panel with per-block memory matches and terminology suggestions, one-click apply
  • Terminology management — concept-oriented, with multi-locale terms, lifecycle statuses, and CSV/JSON import/export
  • Content memory with content-aware tiered matching (generalized, structural, plain), fuzzy matching, and a browsable explorer
  • AI translation using configurable providers (Anthropic, OpenAI, Ollama)
  • Machine translation via DeepL, Google, Microsoft, ModernMT, or MyMemory
  • Pseudo-translation for layout testing and QA
  • Multi-workspace environments with role-based access control
  • Single sign-on (SSO) via GitHub, Google, LDAP, and other identity providers
  • Format-agnostic editing supporting HTML, XML, JSON, YAML, PO, Properties, Markdown, CSV, XLIFF, SRT, VTT, and more
  • File export in original format with translations applied
  • Drag-and-drop file upload with format auto-detection

Deployment

The platform is served by bowrain-server. Users authenticate through your identity provider and can belong to multiple workspaces with different roles. See For developers → Self-hosting for deployment instructions.

Browser and desktop clients

The browser and the desktop app share the same core surfaces (translation editor, content memory, terminology explorer) through the @neokapi/ui shared package, and connect to the same server and workspaces — so multi-user presence, SSO, terminology, and the content memory are shared across them. The differences are about where and how you run, not what you can do: the desktop client adds offline editing and runs as a native app. See The desktop app for the full comparison.