Quick wins for a faster PC:
Repair Windows errors before they cause bigger problemsFix Now →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →In Routebase’s September 2026 study, just under half of 493 self-contained public API specifications that could be compared had changed structurally since the directory’s 2023 copies. Among those changed contracts, about two-thirds included at least one change the study classified as breaking. And among 161 contracts with a breaking change, 143 did not raise their major version. The findings are a warning not to treat a version number—or a missing specification URL—as a complete account of an API’s status.
What happened to the 1,927 specification URLs?
Routebase re-fetched 1,927 origin URLs for Swagger 2.0 or OpenAPI 3.x specifications listed in the APIs.guru directory. The fetch took place on September 14, 2026, using one attempt per URL, a 20-second timeout, and no retries. The directory copies served as the study’s spring 2023 comparison point because its update job had last refreshed a specification in April 2023.
As an Amazon Associate I earn from qualifying purchases.
| Outcome of the 1,927 URL fetches | Count | Share |
|---|---|---|
| Returned HTTP 200 | 1,230 | 63.8% |
| Returned 404 or 410 | 521 | 27.0% |
| Did not answer | 122 | 6.3% |
| Returned 401 or 403 | Not stated by Routebase | 2.0% |
| Other outcomes | Not stated by Routebase | 0.8% |
These are URL outcomes, not a census of whether the underlying services still existed. A provider may have moved a specification file while leaving its API available. A one-time fetch without retries can also be affected by temporary network failures.
Recommended Free Tools
How many API contracts could be compared?
The headline contract results use a much smaller group than the URL count. Of the fetched documents, 838 could be parsed and paired with a 2023 directory copy. Routebase then used 493 self-contained pairs for its main consumer-visible contract comparison. Pairs that referenced external documents were excluded because the comparison parser treated those references as empty objects.
#1 Best Overall
- API Design Patterns
- ABIS BOOK
- Manning Publications
The 1,927 figure therefore describes selected origin URLs, while the structural-change percentages describe 493 comparable pairs. Neither denominator represents every public API, and the results do not cover internal APIs. The old side was the directory’s rendered and sometimes corrected copy, not necessarily a byte-for-byte copy of the provider’s original specification.
How often did comparable contracts change?
Routebase reported structural changes in 48.9% of the 493 self-contained pairs. It aimed to count changes clients could observe, excluding edits to descriptions and examples. The comparison accounted for path placeholders by position and inlined local references before diffing.
Of the contracts that changed structurally, 66.8% had at least one breaking change under Routebase’s classification rules. The study’s reported breaking categories show the kinds of changes that can matter to clients:
Rank #2
| Reported breaking-change category | Share of specifications with a breaking change |
|---|---|
| Endpoint removal | 45.3% |
| Response-field removal | 44.1% |
| Parameter removal | 36.6% |
| Response-type change | 36.0% |
| Newly required request-body field | 17.4% |
| Endpoint-path change | 15.5% |
These categories can overlap: a single specification could, for example, remove an endpoint and change a response type. The percentages should not be added together.
Did breaking changes come with a major-version increase?
Among 161 comparable contracts with at least one breaking change, 143 (88.8%) did not raise their major version; 18 did. In 86 cases (53.4%), the full version string remained unchanged. These figures describe version strings in this sample and under this method; they do not establish that semantic versioning itself failed.
The practical point for API consumers is narrower: a major-version number alone was not a reliable signal that a contract had remained compatible in the specifications Routebase examined. Teams should inspect changes to the contract rather than using the version label as their only alert.
Did a few large providers drive the pattern?
Routebase reported that four providers accounted for more than half of the directory, so it also recalculated the headline results using alternative weighting and exclusions.
Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Repair Windows errors before they cause bigger problems3Scan for outdated or missing drivers - takes under a minute| Calculation | Structurally changed | Changed contracts with a break | Breaking contracts without a major bump |
|---|---|---|---|
| Raw results | 48.9% | 66.8% | 88.8% |
| One vote per provider | 42.0% | 70.8% | 87.7% |
| Excluding the four largest providers | 50.3% | 66.7% | 88.8% |
The pattern persisted across these calculations, although the percentages vary with the weighting. This check does not make the directory representative of all public APIs; it shows that the result was not confined to the raw count’s provider weighting.
What does the longer history show?
In a separate analysis of 106,083 consecutive revisions of existing specification files from 2015–2023, Routebase reported that 26.7% changed the contract structurally. Of those structural changes, 30.8% broke a client under its rules. This is a different analysis from the 493-pair, three-year comparison, so its percentages should not be combined with the headline results.
What did the study find about specification quality?
Routebase also assessed 2,481 specifications in the 2023 snapshot. These quality figures use that separate population, not the 493 pairs in the contract-change analysis.
- 47.6% defined a 4xx error response for at least one endpoint.
- 27.8% attached no security requirement to any endpoint.
- 20.5% defined no security scheme.
- 31.8% used a version string matching the x.y.z shape Routebase checked.
- Routebase’s own neutral-core style rules flagged 98.1% of the specifications, with a median of 13 findings.
The final figure reflects Routebase’s conventions and lint rules, not an external standard or independent quality verdict.
PC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minuteHow should API teams use these findings?
For API consumers
- Review contract diffs during upgrades, especially changes to endpoints, parameters, required request fields, and response shapes.
- Do not infer that a service was retired solely because its specification URL returns an error; check the provider’s current documentation or status information.
- Treat a version number as useful context, not a substitute for checking the contract and testing client behavior.
For API publishers
- Compare each proposed specification with its previous version before release, so reviewers can see potentially breaking changes.
- Check live behavior against the published contract; a clean specification diff cannot establish that production responses match the document.
These are sensible safeguards suggested by the study’s findings, not workflows whose effectiveness Routebase tested in this corpus. The study did not compare tools across release checks, live-response validation, reference handling, configurable rules, or reporting of false positives.
How strong is the evidence?
The study was published by Routebase, which used its own parser, style guide, and breaking-change rules. The team says it manually reviewed ten randomly selected breaking-change verdicts. Its companion repository provides pseudonymized data, tables, checksums, and an aggregation script that can recompute the published aggregates without network access. That makes the aggregates inspectable and reproducible; it is not an independent replication of the parser’s classifications.
- The corpus was limited to public APIs represented in the APIs.guru directory, not all public APIs and not internal APIs.
- The re-fetch covered Swagger 2.0 and OpenAPI 3.x origin URLs; it did not include other source formats.
- Files rejected by the parser were excluded from diff and lint results, and the companion repository says documents larger than 30 MB were skipped.
- External-reference pairs were excluded from the main comparison.
- Breaking-change labels are Routebase’s operational definitions, not an independent standards body’s ruling on each provider’s compatibility policy.
The companion data and tables are identified as CC BY 4.0, and its aggregation script as MIT. Routebase’s article is the primary source for the study’s methods and reported findings; the companion repository documents the data and aggregation method. The APIs.guru directory project provides background on the corpus, but does not independently verify Routebase’s measurements.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
