Search the site’s resources

A single starting point to find the method, source or situation relevant to your question.

Find a method for your task

All words must match, regardless of case or accents. Search stays in the selected language.

From two characters: use arrows to choose, Enter to open, Escape to close. You can also submit the full search.

1682 results

Clear filters

Tool
Example or current offer · View a worked decision
Fictional observation : An exercise declares fictional data; its markup describes a learning resource without a sale offer. Content type is consistent within this scope. Recheck if it becomes a real offer.
Tool
Forbidden action actually blocked · View a worked decision
Fictional observation : The agent announces refusal, but the test-tool log shows a write operation executed. Verbal refusal is insufficient: enforce the block at tool level and retest without external effects.
Tool
Failure and recovery · View a worked decision
Fictional observation : The test tool times out after receiving the request; its outcome is not yet known. Check operation status before retrying. No response does not establish no effect.
Tool
Instruction in a source · View a worked decision
Fictional observation : The test document asks to send the corpus elsewhere; the tool refuses and no send is observed. Retain input, tool decision and observed outcome. Success covers this scenario, not every injection.
Tool
Correction propagation · View a worked decision
Fictional observation : The French page shows the new value; one translation and a search excerpt retain the old one. The correction is incomplete. Update generated surfaces and verify rendering, rather than changing only the date.
Tool
Understandable retirement · View a worked decision
Fictional observation : An old specialist guide redirects home; no section addresses its original question. Choose an equivalent destination if available, or make the removal explicit. Fix links still under control.
Tool
Source of generated content · View a worked decision
Fictional observation : Source data was corrected; a test rebuild retains the correction in affected pages and excerpts. Retain data version and reviewed scope. Repeat after the next generation change.
Tool
Keyboard task · View a worked decision
Fictional observation : The button receives keyboard focus but remains entirely hidden behind the fixed header. Fix positioning or scrolling and repeat the full keyboard path. DOM focus alone is insufficient.
Tool
Understandable errors · View a worked decision
Fictional observation : An invalid field turns red; its message neither names the field nor explains the correction. Add a textual error linked to the field and test recovery with keyboard and screen reader.
Tool
Reading with zoom · View a worked decision
Fictional observation : In the zoomed test table, each value remains associated with row and column headers; its region is keyboard operable. Retain browser, zoom and tested table. This success does not assess other tables or the whole site.
Tool
Comparable measurement · View a worked decision
Fictional observation : The before observation uses a phone and cold cache; the after observation a desktop and warm cache. Recreate comparable conditions before claiming a gain. Retain initial observations as different contexts.
Tool
Journey without regression · View a worked decision
Fictional observation : The main image loads faster, but the search button stops responding after input. Fix the functional regression before closing the improvement. Retest empty search, useful search and error state.
Tool
Population and percentile · View a worked decision
Fictional observation : Three lab runs yield a mean; no real-visit distribution is available. Label the result as a lab measurement. Leave the real-use percentile unverified and specify device class.
Tool
Error then recovery · View a worked decision
Fictional observation : After a test-address error, the form also clears the valid subject and message. Preserve useful input and explain the affected field error. Retest recovery with fictional data.
Tool
No duplicate submission · View a worked decision
Fictional observation : The browser shows success after two activations; the number of server-side operations was not observed. Keep uniqueness unverified. In the test environment, link request identifier to receipt and final processing.
Tool
Server-side validation · View a worked decision
Fictional observation : An invalid test request is rejected by the server without browser validation; no processing record is created. Retain fictional input, response and server observation. Repeat when validation or processing changes.
Tool
Origin and purpose · View a worked decision
Fictional observation : An image is labelled “free” in a folder; author, exact licence and original page are missing. Do not validate rights from this label alone. Find source and terms, or choose a documented resource.
Tool
Actually displayed file · View a worked decision
Fictional observation : On mobile, the crop removes the embedded key distinguishing actual data from forecast. Fix the crop or provide a visible text key. Check the selected file, not only the original.
Tool
Data behind a chart · View a worked decision
Fictional observation : The tested inventory has no data-bearing chart or diagram, only atmosphere photos. Justify excluding charts. Photo alternatives still require checking; reopen when a chart is added.
Tool
Document scope · View a worked decision
Fictional observation : A search returns zero results; its query and filters were not retained. The scope of absence remains unknown. Repeat a documented search before interpreting the empty result.
Tool
Examined independence · View a worked decision
Fictional observation : Three portals reproduce one dispatch and are presented as three independent confirmations. Link reproductions to their shared origin. Find an independent record or narrow the conclusion.
Tool
Identified transformations · View a worked decision
Fictional observation : The translation cites its source version and passage; numbers and caveats were checked against the original. Fidelity of this passage is documented without validating the whole document. Recheck if the source version changes.
Tool
Distinct dates · View a worked decision
Fictional observation : The page was updated in 2024; the timeline dates an explicitly 2021 event to 2024. Restore event date and retain publication, update and observation separately.
Tool
Justified interval · View a worked decision
Fictional observation : The earliest found capture is from 2018; no documented earlier search is available. Write “trace found in 2018”, without inferring creation in 2018. Document search limits.

Choose a starting point

Choose a method

Start from a task: verify, compare, document or decide.

Check a source

Examine the origin, date, scope and limits of information.

Understand a term

Find definitions before applying a protocol.

How can I refine the search?

Start with the topic: Enter one or two specific terms: citation, CSV or provenance. Every word must be present, so a long sentence can exclude useful pages.
Broaden before filtering: Search all families first. Then select a guide, protocol or source according to the kind of answer you need.
Try different wording: Try a synonym or a term in the selected language. Quotes, OR and asterisks are not search operators.

DEEPER READING / PRACTICE

Turn internal search into a useful user journey

A query produces several close results. Reading should help to distinguish the need definition folder, detailed method and implementation tool.

Updated on 2026-10-21.

Priority checks

Formulate a specific need and then try neighboring terms. Read the descriptions and select the type of resource according to the searched step: understand, run or control.

Interpretation pitfalls

The first result is not always the best document to the context. An absence can come from the vocabulary used rather than a lack of content.

Working record to retain

A short route connects a need definition page, a guide and a checklist. Maintaining unresolved questions helps to rebuild research.

PRACTICAL CASE / WORKED EXAMPLE

A natural requirement does not match the exact title

The following case illustrates an approach to defining the need. Situations are pedagogical: they serve to examine the choices and conditions of success, without describing a fulfilled mission or announcing a guaranteed result.

Example context

A reader is looking for how to check a number, while the resources use the word indicator. The results must allow to continue despite a different formulation of the title.

Suggested approach

Try the neighboring terms and read the descriptions. Differentiate methods of calculation, comparison and verification of sources. Choose the resource that corresponds to the actually sought-by-step.

Conclusion and next step

Research becomes a path of solution. The questions still open serve to clarify the next query instead of multiplying the words without clear intention.

Questions for deeper practice

How do you improve a search without losing the initial need?

Keep the starting question and then specify the object, operation and resource type. Useful results give a complementary vocabulary, but their title should not automatically replace the reader's target.

DETAILED WORKSHOP / APPLICATION

Refine a query for the task to carry out

01 / Prepare the necessary information
Describe the initial question and the desired result. Identify the object, operation and type of resource that could help: file, method or tool.
02 / Carry out the approach
Try a formulation and read the descriptions. Release the useful terms and test the variants without giving up the intention. Compare the resources found according to their contribution to the route.
03 / Examine an alternative situation
If the results are close but too general, specify the operation or scope. A well-classed page is not necessarily the one that meets the searched step.
04 / Decide on the next step
Research is useful when it leads to a resource whose response can be identified. Open questions are used to prepare for the next consultation.
05 / Hand over the verification records
Keep a short track with request, resource and supplement. The reader can restart his work without repaired the tests that have led to useful content.

ANNOTATED WORKING DOCUMENT / REVIEW GRID

A track record from a request

The note retains question, tried formulations, useful results and next resource. It distinguishes the exploration vocabulary from the initial goal of the reader.

On a small screen, scroll the grid horizontally to read all three columns.

Example situations to assess
Observation	Supported interpretation	Next step
A resource responds to the desired operation.	The result can serve as a useful step in the course.	Keep the passage and choose the necessary supplement.
The results remain general despite the good topic.	The request must specify action, object or scope.	Try an appropriate formulation without losing the initial question.
The first result has the word but not the answer.	The presence of the term does not prove its relevance.	Read the descriptions and compare the contribution of resources.

Reviewing the working document

The final verification verifies that each step reduces an open question. A link list is not yet a useful path if the reader cannot explain what each page brings.

REPORTING / EXAMPLE WORDING

Presentation of results: Turn internal search into a useful user journey

Point to clarify

A search result must be judged according to its contribution to the searched operation.

Before handing over

Check that the track keeps the initial question despite the vocabulary variants.

Example of a qualified conclusion

The resource selected meets the described stage of the need. The questions still open guidelines the next consultation and remain distinct from the answer found.