Compare two file lists →
Find missing, extra and repeated names after extraction using lists pasted into the page.
The comparison checks supplied names and paths, not file contents or integrity.
Read the full methodExamine a web address, ISBN or ISSN, compare image licences and use calculators with clear methods and limits.
Choose a task. Each workshop explains what it calculates or records and links to the full method.
Put two or three directory records side by side. Compare their purposes, publishers and source documentation.
Open the directory comparison →Identical displayed times can refer to different instants. Enter the date, time and offset stated by each source.
Fictional example: 02:30 at +02:00 is 00:30 UTC; 02:30 at +01:00 is 01:30 UTC. The second instant is 60 minutes after the first.
Method: convert each time to UTC, then subtract the first instant from the second. Use the offset applicable on that date; this tool does not determine a city’s daylight-saving rules.
Local calculation to the minute. Entries are neither sent nor saved. A time difference cannot establish the actual order of events if their source dates are wrong. Dates and time zones explained · W3C ↗
Distinguish a research object from a person identifier. This workshop prepares a link and checks the ORCID check digit. It contacts no registry and verifies neither object existence nor author identity.
Local processing, with no submission or automatic storage of the input. Opening the prepared link is a separate action contacting the registry and possibly the DOI destination.
Documentation consulted on 9 October 2026: DataCite — DOI Basics, ORCID — Identifier structure. Examples come from this documentation; they are not publications or profiles created by the portal.
Identify the right publication
A book edition and a serial title are different bibliographic objects. This workshop checks transcription arithmetic locally. It does not query a catalogue, identify a rights holder or establish that the number was assigned.
| Identifier | Structure | What to compare with the record |
|---|---|---|
| ISBN-10 | Legacy 10-character format; final key can be X. | Title, edition, language and format. |
| ISBN-13 | 13 digits; ISBN prefixes 978 or 979, excluding the ISMN prefix 979-0. | The particular edition and product form. |
| ISSN | 8 characters, conventionally shown as two groups of four; final key can be X. | Serial title and medium, rather than an individual article. |
The last character is excluded from this sum. The expected key is the complement of the remainder, modulo 10 for ISBN-13 and modulo 11 for ISBN-10 and ISSN. In modulo 11, 10 is represented by X.
| Position | Digit | Weight | Product |
|---|
A matching check character detects some transcription errors; it does not validate publication quality, catalogue availability or reuse rights. A mismatch is a prompt to check the source, not permission to replace the last character.
Documented calculation examples: ISBN 978-92-95055-12-4 in the international manual; ISSN 0317-8471 in the Library of Congress method. The numbers illustrate arithmetic, not new records created here.
Sources consulted on 25 October 2026: International ISBN Agency · ISBN manual · ISBN.org · Library of Congress · ISSN · ISMN. No entries are sent or saved.
Separate the contacted host, path, parameters and fragment. A familiar name in the path or before an @ can obscure another destination. This workshop does not verify site identity or redirects.
HTTPS describes the protocol, not reliability. The registered domain is not inferred automatically: suffixes such as .co.uk require an appropriate suffix list. A shortened link requires a separate check of its final destination.
Fictional example: https://example.org/a and https://example.org:443/b share an origin because 443 is the default HTTPS port. http://example.org/a or https://example.org:8443/a have a different origin. Path, query and fragment do not define the origin.
Origins and default ports: WHATWG — URL Standard · .
Documentation consulted on 8 October 2026: MDN — URL.hostname, URL.username, URL.hash.
One chart, two views
Fictional example: an index rises from 98 to 99 to 100 over three equally spaced periods. Compare the line on a full and a narrow scale. Its values and change stay the same.
| Period | Index, arbitrary unit |
|---|---|
| A | 98 |
| B | 99 |
| C | 100 |
From A to C: +2 units, approximately +2.04%. A line encodes positions: a clearly labelled narrow scale can reveal a change. For bars whose length encodes the value, use a zero baseline.
Documentation: Office for National Statistics — Axes and gridlines · Consulted on 15 October 2026. Display changes locally without saving.
A combined rate uses the total number of cases divided by the total population. Both groups must use the same definition and period, and must not overlap.
| Group | Cases / population | Rate |
|---|---|---|
| A | 10 / 20 | 50 % |
| B | 90 / 100 | 90 % |
| Combined | 100 / 120 | 83.33 % |
The simple average (50 + 90) ÷ 2 = 70% gives equal weight to the two groups. The combined rate gives equal weight to each observation: (10 + 90) ÷ (20 + 100) × 100 ≈ 83.33%.
Calculated only in this browser. No entries are sent or saved. Rounding to two decimals does not measure uncertainty; this tool does not validate group comparability.
From an image to its reuse terms
Find the licence on the file’s source page, including its version. This comparison covers the six international 4.0 licences; it does not assign a licence to an image.
| Licence | Commercial use | Sharing adaptations | ShareAlike |
|---|---|---|---|
| CC BY 4.0 | Permitted under the terms | Permitted under the terms | — |
| CC BY-SA 4.0 | Permitted under the terms | Permitted under the terms | For shared adaptations |
| CC BY-ND 4.0 | Permitted under the terms | Excluded by ND | — |
| CC BY-NC 4.0 | Excluded by NC | Permitted under the terms | — |
| CC BY-NC-SA 4.0 | Excluded by NC | Permitted under the terms | For shared adaptations |
| CC BY-NC-ND 4.0 | Excluded by NC | Excluded by ND | — |
All six require appropriate attribution, a licence link and an indication of changes. SA governs the sharing of adaptations; ND prevents sharing adapted material. NC restricts commercial uses. Other rights, including privacy and personality rights, can still matter.
Local reading aid: no entries are sent or saved. It neither verifies the rights holder nor decides whether a modification legally counts as an adaptation. If these points are uncertain, consult the source or obtain permission.
Keep the work’s title when supplied, creator, source page, licence name and version, licence link, and changes made. Example of a record structure: title — creator — source — CC BY 4.0 — resized. Fill it only with the actual file’s information.
Official licence summaries linked in the table, consulted on 25 October 2026. Read the applicable legal code and notices for the exact terms.
Save the initial question, units, inputs, date and limitations with your result. An illustrative example is not a measurement of your situation. Recalculate when inputs change.
Calculations and worksheets run in your browser without sending or saving entries. Sending a prepared query to Google is a separate action. The address reader does not contact the pasted destination.
Choose with a method
A search moves forward when you know what document could answer the question. A portal, a search engine, a database and a public service can complement each other while serving different purposes.
Choose a route to display its checkpoints.
Write the question in one sentence and specify the expected answer: a text, date, definition, dataset or procedure. A precise question helps you choose between general search and a specialist resource.
When you find a claim, look for the document it cites. Record the title, author or organisation, period and relevant passage. Another page repeating the same source is not necessarily independent confirmation.
Check that both values use the same unit, period and comparable populations. A difference can come from definitions, coverage or revisions rather than a real change.
Keep limitations with the result. If the scope of a value is unknown, record that gap instead of building an average or ranking from incompatible inputs.
Distinguish an information page from a service where you can take action. Read the host name and check that the link belongs to the intended service, especially before sharing data or making a payment.
A familiar domain mentioned within a page does not establish that page’s identity. The address reader helps separate the parts of a link; syntax checking cannot verify the service.