The hardware and bandwidth for this mirror is donated by dogado GmbH, the Webhosting and Full Service-Cloud Provider. Check out our Wordpress Tutorial.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]dogado.de.
Hybrid telemetry objects for backward
compatibility. bid_ingest_telemetry() now returns
a hybrid bid_issues object that behaves as a list
(maintaining legacy compatibility) while providing enhanced
functionality through new methods: print.bid_issues(),
as_tibble.bid_issues(), and
bid_flags().
Modern tidy telemetry API. New
bid_telemetry() function provides a clean, tidy approach
that returns organized issues tibbles (bid_issues_tbl
class) for better integration with dplyr workflows and data analysis
pipelines.
Bridge functions for telemetry-to-BID integration. New functions seamlessly connect telemetry issues to BID stages:
bid_notice_issue() - Convert individual issue to Notice
stagebid_notices() - Batch process multiple issues to Notice
stagesbid_address() - Sugar function for quick issue
addressingbid_pipeline() - Process first N issues with
limitsTelemetry-informed structure selection.
bid_structure() now accepts telemetry_flags
parameter to adjust layout selection and suggestion scoring based on
real user behavior patterns (e.g., avoiding tabs layout when navigation
issues are detected).
Enhanced validation with telemetry references.
bid_validate() now includes optional
telemetry_refs parameter to append provenance information
linking improvements back to specific telemetry findings.
Corrected stage numbering. Fixed stage
progression to proper BID framework order: Interpret (1) → Notice (2) →
Anticipate (3) → Structure (4) → Validate (5). Added temporary migration
support with stage_number_previous attributes and
session-level migration notices.
Consolidated suggestion system. Unified
suggestion rules for Interpret, Notice, and Validate stages in
R/suggest_rules.R, reducing code duplication and ensuring
consistent messaging across the framework.
Removed duplicate warning-suggestion pairs. Cleaned up verbose output by preferring actionable suggestions over redundant warnings where content overlapped.
Enhanced Structure suggestions organization.
Consolidated structure-specific suggestion functions into
R/structure_suggest.R with documented schema and improved
concept-based grouping.
Improved telemetry flag extraction. New
.flags_from_issues() helper function creates comprehensive
boolean flags from telemetry patterns for better layout and suggestion
customization.
Better suggestion theory integration. Factored
out .suggest_theory_from_text() for reuse across Notice
stage functions, improving consistency in theory
recommendations.
Migration support utilities. Added
.show_stage_numbering_notice() and
.format_telemetry_refs_for_validation() helper functions
for smooth version migration.
Updated Getting Started vignette with new telemetry workflow examples and corrected stage numbering throughout.
Enhanced Telemetry Integration vignette showcasing the new API with hybrid objects, bridge functions, and tidy workflow examples.
Comprehensive test coverage for new bid_issues class methods, bridge functions, and edge cases. Also, expanded coverage for previously missed lines across bid_anticipate, bid_concepts, suggest_rules, and utility functions.
For users upgrading from 0.3.0:
# Legacy telemetry code continues to work
legacy_notices <- bid_ingest_telemetry("telemetry.sqlite")
length(legacy_notices) # Still works as before
# But now also provides enhanced features
as_tibble(legacy_notices) # New: get tidy issues view
bid_flags(legacy_notices) # New: extract telemetry flags
# New modern API for tidy workflows
issues <- bid_telemetry("telemetry.sqlite")
critical <- issues |> filter(severity == "critical")
notices <- bid_notices(critical, previous_stage = interpret_result)Note: bid_ingest_telemetry() will be
soft-deprecated in 0.4.0 in favor of bid_telemetry().
Current usage remains fully supported.
Fixed concept-based bias mitigation generation.
bid_anticipate() now correctly generates concept-specific
biases (e.g., “attention bias” and “belief perseverance” for Visual
Hierarchy) instead of falling back to generic biases.
Improved stage flow validation. Updated validation logic to support the flexible BID workflow where inner stages (Notice, Anticipate, Structure) can occur in any order, while maintaining proper validation for entry and exit points.
Enhanced previous_stage handling. Fixed validation errors when passing data.frame objects to BID functions by improving stage name extraction logic.
Fixed stage progression validation warnings in tests by correcting test sequences to follow proper BID framework order
Updated suggestion pattern matching in tests after DRY refactor consolidation
Layout auto-selection is deprecated. The layout
selection feature in bid_structure() will be removed in
bidux 0.4.0 to reduce complexity and focus on concept-based suggestions.
Existing code continues to work with deprecation warnings.
Layout-specific bias mitigations are deprecated.
The layout-dependent bias mappings in bid_anticipate() will
be removed in bidux 0.4.0 in favor of concept-driven bias mitigations.
Existing code continues to work with deprecation warnings.
BID stage order updated to: Interpret → Notice → Anticipate → Structure → Validate. This reflects the most natural workflow while maintaining flexibility for iterative usage.
bid_notice() no longer accepts
target_audience parameter. Audience information
should now be managed through the bid_interpret() stage
using the data_story or user_personas
parameters. The function will warn if the deprecated parameter is
provided.
bid_anticipate() no longer accepts
interaction_principles parameter. This parameter
has been removed in favor of the new include_accessibility
parameter (default: TRUE). The function will warn if the deprecated
parameter is provided.
Field name changes for consistency:
previous_question →
previous_central_questionprevious_story_hook → previous_hookuser_personas → personas (in bid_interpret
output)bid_structure() no longer accepts a
layout parameter. Layout is now automatically
selected based on content analysis of previous stages using
deterministic heuristics. The function will abort with a helpful error
message if the deprecated layout parameter is
provided.
Enhanced bid_validate() with
experiment/telemetry/empowerment flags. New optional
parameters: include_exp_design (default: TRUE),
include_telemetry (default: TRUE), and
include_empower_tools (default: TRUE) for context-aware
suggestions.
Accessibility-focused bias mitigation in
bid_anticipate(). New
include_accessibility parameter adds layout-specific
accessibility recommendations to bias mitigation strategies.
Improved context propagation. All
previous_problem, previous_theory,
previous_audience, and previous_personas now
properly propagate through the entire pipeline to the Validate
stage.
Automatic layout inference for
bid_structure(). Uses sophisticated heuristics to
analyze content from previous BID stages (problem, evidence, data story
elements) and automatically select the most appropriate layout
(breathable, dual_process, grid, card, or tabs).
Ranked, concept-grouped actionable suggestions.
bid_structure() now returns structured UI/UX
recommendations organized by UX concepts (e.g., Cognitive Load Theory,
Progressive Disclosure) with specific Shiny/bslib component pointers,
implementation rationales, and relevance scores (0-1).
Telemetry-aware suggestion scoring. Layout
selection and suggestion ranking now consider telemetry data,
automatically avoiding problematic patterns (e.g., avoiding tabs if
nav_dropoff_tabs is detected).
Enhanced CLI feedback. Clear messages show the
auto-selected layout with rationale, plus a helpful tip about using
bid_concept("<concept>") to learn more about any
detected concepts.
Field name normalization. Added
normalize_previous_stage() helper function to handle legacy
field names and ensure consistent data flow between stages.
Enhanced helper functions. Updated utility
functions with time stubbing support (.now()) and improved
safe column access patterns.
Comprehensive test coverage. Added extensive
tests for all heuristic branches, suggestion structure validation, CLI
message verification, telemetry integration, and error handling for the
deprecated layout parameter.
To update existing code for bidux 0.3.0:
# OLD (0.2.x)
notice <- bid_notice(
problem = "Users confused",
evidence = "High error rate",
target_audience = "Analysts"
)
# NEW (0.3.0)
notice <- bid_notice(
problem = "Users confused",
evidence = "High error rate"
)
interpret <- bid_interpret(
previous_stage = notice,
data_story = list(
audience = "Analysts" # Move audience here
)
)
# OLD (0.2.x)
anticipate <- bid_anticipate(
previous_stage = structure_result,
interaction_principles = list(hover = "Show details on hover")
)
# NEW (0.3.0)
anticipate <- bid_anticipate(
previous_stage = structure_result,
include_accessibility = TRUE # New accessibility focus
)
# Field name updates in downstream code:
# Access previous_central_question instead of previous_question
# Access previous_hook instead of previous_story_hooktarget_audience parameter in bid_notice()
(deprecated, will be removed in 0.4.0)interaction_principles parameter in
bid_anticipate() (deprecated, will be removed in
0.4.0)Telemetry-driven UX friction detection. New
bid_ingest_telemetry() ingests
{shiny.telemetry} data (SQLite or JSON), applies
configurable thresholds, detects friction patterns (e.g., unused inputs,
delayed interactions, error clusters, navigation drop-offs, confusion),
and generates BID notices and reports. Includes robust input validation,
format auto-detection, and clearer CLI summaries. (#18)
Telemetry integration vignette. A new vignette
shows how to set up {shiny.telemetry}, run
bid_ingest_telemetry(), interpret indicators, and customize
thresholds within the BID workflow. (#18)
Modular printing and display rules.
print.bid_stage() refactored into
print_stage_header(), print_stage_content(),
and print_stage_footer() with reusable field printing and
stage-specific display rules for clearer, more consistent
output.
Consistent data loading. Introduced
load_external_data(); concept, theory, bias, layout, and
accessibility loaders now share one code path with sensible
fallbacks.
Suggestion generation & validation
utilities. Consolidated helpers:
generate_stage_suggestions(),
get_default_suggestion_rules(), and
evaluate_suggestion_condition(); standardized parameter
checks with validate_character_param(),
validate_list_param(), and
validate_bid_stage_params().
Stronger parsing & normalization. Telemetry JSON handling now supports both JSON arrays and JSON-lines; column normalization validates and standardizes inputs; percentage/label formatting made more consistent.
Safety & style passes. Safer subsetting
(drop = FALSE), safer indexing (seq_len()),
tightened conditionals/bracing and indentation, aligned function
signatures and lists, and general formatting cleanup for readability and
maintainability.
Docs & guides. Added CONTRIBUTING guidelines and a Code of Conduct; standardized GitHub issue templates for clear, consistent reports. (#16)
README overhaul. Reorganized overview, features, installation, and a concise quick start; added a telemetry section; improved examples; suppressed noisy messages during README generation; added a CRAN downloads badge. (#15, #16)
Reference updates. Exported and documented
bid_ingest_telemetry(); refreshed man/ pages
and package index to reflect new modules and function locations.
(#18)
File organization. Standardized dash-separated
filenames (e.g., bid-classes.R,
bid-concepts.R); centralized telemetry
ingestion/analysis/notice creation into telemetry.R; moved
shared helpers into utils.R.
Namespace & dependencies. Added
DBI and RSQLite to Imports, and
importFrom(stats, complete.cases); relaxed version
constraints for cli, dplyr, and
jsonlite; expanded NAMESPACE with explicit
imports.
{shiny}, {bslib}, {DT},
{plotly}, {reactable},
{htmlwidgets}.bid_report().bid_notice().These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.
Health stats visible at Monitor.