HOW THE CATALOG WORKS
Methodology
The public catalog is built from selected, immutable releases of reviewed records. Only released taxonomy information appears on these pages.
What belongs in the taxonomy
Each record represents one reusable technology concept with an immutable canonical ID. The scope includes vendors, products, frameworks, standards, tools, platforms, languages, databases, libraries, and runtimes. Categories and domains provide a browseable hierarchy around those records. Our aim is a useful reference grounded in observed hiring demand, rather than an inventory of every GitHub repository.
Inclusion and retention
Our editorial policy uses a rolling 12 months ending on the review date:
- New technologies: a candidate must appear in at least two distinct job advertisements during that period to qualify for inclusion. Maintainers find candidates with full-text searches of job-ad titles and descriptions for names and aliases; identity, scope, and supporting sources still require human review.
- Existing technologies: at least one verified mention during that period is enough to meet the retention threshold. One or two mentions do not trigger retirement.
- No recent mentions: an existing technology with zero verified mentions over the full period is retired from the active taxonomy after human review of source coverage and matching.
Count a distinct advertisement once for each technology, including when it mentions multiple aliases. Duplicate copies and reposts of the same advertisement do not provide additional evidence. A repository, star count, or vendor announcement alone does not meet the job-ad threshold.
Retirement describes inclusion in this taxonomy; it does not mean the technology itself is discontinued or unused. Preserve its canonical identity and historical releases, and explain the reason for retirement without recycling its ID.
Job-ad evidence and coverage
The evidence comes from the database maintained by quea.st, which collects IT job advertisements from Europe. It reflects the advertisements Queast observes, not every vacancy or all technology use in Europe or worldwide.
Reviewers search job-ad titles and descriptions directly for the candidate name and aliases, then verify the technology, distinct-ad count, dates, European IT relevance, and coverage of the full 12-month window. New candidates are searched this way because they are not yet in the technology-association index. Missing or incomplete data is not evidence of zero mentions: uncertain cases stay pending review rather than being retired on that basis.
This policy guides new additions and ongoing review; it is not a claim that every existing record has already been checked against the threshold. Inclusion and retirement decisions enter the public catalog through reviewed releases.
Identity and authority
The taxonomy team maintains stable public IDs and slugs, aliases, descriptions, classifications, source links, relationships, revisions, and releases. A technology can change its name or slug without losing its identity; redirects preserve older links.
Sources and review
- Maintainers collect source links and record which claim each source supports.
- Changes are reviewed field by field, with a reason and the release that contains them.
- A human reviewer approves each revision. Automated suggestions can assist research but cannot publish themselves.
- Each public release is deterministic, checksummed, and kept immutable after publication.
Historical releases remain available. Corrections create a new revision and release; slug changes and merges retain redirects.
Popularity from job advertisements
Opening a technology record loads weekly counts of distinct, non-hidden Queast job advertisements mentioning that technology in Finland, Sweden, Germany, and the United Kingdom. Results are cached per technology for six hours. The graph ends today and begins on January 1, 2026, or later when Queast data for the technology begins later. Older and future-dated source observations remain outside the displayed graph. Every available weekly observation is included without a minimum-sample filter. The two-ad inclusion threshold applies to editorial review, not to individual chart points. The displayed chart is not a complete 12-month eligibility check. This signal describes observed job-ad demand, not general adoption, revenue, or market share.
The four popularity labels use quartiles of positive distinct-ad technology-association counts across the same four markets since January 1, 2026. A missing association stays unclassified. A Requires context marker warns that the name is an ordinary or ambiguous term whose text matches need surrounding evidence; it does not by itself validate an advertisement match.
Freshness and missing data
Record pages show the review date available in their taxonomy release. Missing descriptions, classifications, source links, relationships, mappings, or job-ad observations are labeled as unavailable.
Corrections and disputes
Public intake is not open yet. The contribution page explains the current status.