Privacy and site use

Browser calculations

The website’s calculators process the figures you enter in your browser. They do not send those figures to a server or save them in browser storage. Download any results you want to keep before leaving the page. Downloaded files stay wherever you save them.

The accounting-question search also runs in your browser. It does not send search text to a search service. It does place the search term and topic in the page address so a filtered list can be bookmarked or shared, which means the term appears in your browser history and in any address you pass on.

Cookies and analytics

The website’s own code does not set cookies or include advertising pixels. Its fonts, images and application scripts are served from this domain.

No analytics script is delivered. Cloudflare Web Analytics is switched off, so I receive no page-view reports. GitHub Pages and Cloudflare still process requests to host and protect the site, as the Hosting section below describes.

Email Address Obfuscation is disabled so email links work without scripts. The address is public in the HTML pages, their plain-text alternates and the machine-readable index. The contact page also spells it out for anyone whose browser does not open an email app.

Cloudflare also asks browsers to report network errors to Cloudflare for seven days and to prefetch pages you are likely to open next (the nel, report-to and speculation-rules response headers). Neither carries the figures you type into the calculators. A prefetched or shared page address can carry question-search terms, because the search places them in the address; do not put confidential information in a search you may share or leave in your history.

On a sample of page loads, Cloudflare adds a report-only security policy header, which asks your browser to tell Cloudflare which scripts the page loads and which connections it makes. It blocks nothing. A report includes the page address, which can hold question-search terms, but not the figures you type into the calculators. If Cloudflare treats a request as suspicious, it can answer with its own verification page, which runs Cloudflare’s code and may set a security cookie; see Cloudflare’s cookie documentation. Cloudflare makes these additions itself; they are not part of this website’s source. The calculators work out their figures in your browser without sending them to a server.

When you select Human or Machine view, your browser saves that display preference on this device. The preference is not sent to a server. You can change it with the view switch or remove it by clearing this site’s browser storage.

Hosting

GitHub Pages hosts this website’s files, and Cloudflare delivers them to your browser. These providers process technical information about visits, which may include IP addresses, request details and security or performance data. See the GitHub Privacy Statement and Cloudflare Privacy Policy for their handling of that information.

Feedback and external links

If you email me, I receive your email address and the content you send. Use fictional examples when reporting a problem. Do not include taxpayer information, client files or other confidential records. An email stays in my mailbox until I delete it, GitHub keeps issues public in its own systems under its terms and I cannot delete copies other people hold.

GitHub issues are public. Links to GitHub and other websites take you to services with their own privacy policies. Software you download or install has its own documented data handling; this notice covers the website.

For questions about this notice or information you have sent me, email ryan@duguid.com.au. To ask what personal information I hold, have it corrected or complain about how I handled it, email that address, or take a complaint about the handling of personal information to the Office of the Australian Information Commissioner. Other project links are on the Contact page.

AI crawlers and citation

The site’s robots.txt names each crawler and its treatment. Search indexing and the fetches an assistant makes to answer or cite for a user are allowed for the agents it names, and Google-Extended is allowed so Gemini grounding can cite the site. Training crawlers from Common Crawl (CCBot), Bytespider, OpenAI (GPTBot), Anthropic (ClaudeBot) and Apple (Applebot-Extended) are allowed, while the other training crawlers it names, Amazonbot among them, are disallowed. The machine-readable index and the full page text are published for citation, and robots.txt is the current policy if this summary and the file ever differ.

If an AI answer misquotes this site, email ryan@duguid.com.au with the page address and the quoted text. Use fictional examples only; do not send client records or taxpayer information.

Using this site

Read the advice and review boundary and the site boundary before relying on a tool or example. Check each page’s assumptions, source links and review date, and have a qualified professional review outputs before acting on them.

The website's own text, data and source code are available under its Apache License 2.0, which includes its warranty terms; each page's structured data names that licence. Linked projects and third-party assets carry their own licences. Check those terms before reusing or distributing them.

Accessibility

I aim for this site to meet WCAG 2.2 level AA. Automated checks run on every change: axe scans each page at phone and desktop widths, and other tests cover the skip link, keyboard focus, one main heading per page and pages that fit those widths without scrolling sideways.

Automated checks find only some barriers. I have not tested the site with screen readers or on physical phones and tablets. If something stops you using a page, tell me through the contact page, naming the page and what happened.