{"canonical_url":"https://engineer.company/privacy/","content_markdown":"# Privacy\n\nThis site sets **no cookies**, loads **nothing from a third party**, and runs\n**zero JavaScript**. Its pages, menus and links use HTML, CSS and the browser's\nown controls. The structured data for search engines is data, not executable\ncode.\n\nThere is no analytics tag, tracking pixel or embedded third-party script or\nfont. Everything needed to read the page arrives as HTML, styles, local fonts\nand images. What is and is not stored in your browser has a page of its own:\n[cookies](/cookies/), with the short answer and the way to check it.\n\nThat is not a policy we promise to keep; it is how the site is made. You can\ncheck it: view the source, or open your browser's network panel and watch which\nservers it talks to. The answer is one — ours.\n\n## What the server records\n\nOur web server keeps an access log, the way essentially every web server does.\nIt is the only way we can tell which pages people read, and whether anyone finds\ntheir way to the contact page. One piece of configuration writes every log on\nthat server, so each address that records anything records it the same way, and\nthat covers more addresses than this one: fifteen today — this site, the `www.`\nversion of it, our `git.` subdomain, the seven focused addresses the [credits\npage](/credits/) lists (one of which answers to a second name as well), the\nshort address that redirects to this page, `engineer.center` and its `www.`\nversion, which redirect to this site, and the host that serves nothing but our\nmail policy. For each request it stores:\n\n- **The page requested**, as your browser asked for it, and the date and time.\n  A link can carry more after a `?`, and the log keeps that as it was sent; the\n  summaries below keep the page alone, plus one word if the link said where it\n  was published (`utm_source` — ChatGPT adds `utm_source=chatgpt.com` to the\n  links in its answers).\n- **Whether it worked** — the response status, so we can find broken links.\n- **Your browser's self-description** (the \"user agent\"), which is how we\n  separate people from search-engine crawlers. A crawler that gives a name is\n  checked against the networks its owner publishes, using the blurred address\n  below.\n- **The site you came from**, if you followed a link — see below.\n- **A blurred network address** — see below.\n\n## Two things worth being precise about\n\n**Your IP address is never stored in full.** It is shortened at the moment the\nline is written, not cleaned up afterwards: the last part is replaced with zeroes\nbefore anything touches the disk. What remains identifies a rough network, not a\nperson or a household — an entire office and an entire street can look identical.\nThe server records the address under two names and masks both, and it drops the\nheaders that can carry an address of their own — the ones a company proxy or a\ncontent delivery network puts in front of you — before the line is written. What\nreaches the disk is that rough network, which is not much to lose, hand over, or\nbe asked for.\n\n**We do keep the site you came from, and that is a real disclosure.** If you\narrive by clicking a link, your browser tells us which page it was, and we keep\nthe site's name — `example.com`, not the full address of the page, and never\nanything after the `?`. We keep it because it is the only way to know whether\nanyone finds us at all, and from where. We would rather write that plainly here\nthan collect it quietly.\n\n## How long, and who sees it\n\nA line lives about **thirty days**, on every address, busy or quiet. Our server\nsoftware closes each log file once a day, and sooner if it reaches about\n**10 MB**, and deletes a closed file twenty-eight days after closing it. Add the\nday a line can wait for its file to close and the day a closed file can wait to\nbe deleted, and that comes to thirty. Every night a separate check reads every\nfile back and fails if any line in it is older than that.\n\nThis has gone wrong twice, and we would rather tell you about both than fix them\nquietly. In September 2026 we found four quiet addresses that filled 10 MB so\nslowly that their log had never been closed at all and still held every line\nsince we switched logging on. The daily job we added then to close those logs\ndid not hold. On 26 October 2026, seven of our thirteen addresses were still\nwriting into a file the job had already marked as closed and would not delete\nwhile it kept growing, and one address still kept a closed file forty-seven days\nold. The server software now closes and deletes the files itself, and the job is\ngone.\n\nSummaries outlive the log, and we would rather say so than let you find out. A\nprogram on the server reads the log and prints counts; the counts travel to our\nown computer, where the summaries are written and kept in a private repository\nof ours. Each address has four of them — the last thirty days for people, the\nsame for everything automated, a row per day and a ranking per month — plus a\npage for each month and each year. They are written afresh on every run and the\nrepository keeps its history, so every earlier edition of them stays too.\n\n**A daily tally**, kept indefinitely: so many pages read, so many of them by\ncrawlers, so many reaching the contact page. Dates and counts, nothing else.\n\n**A monthly ranking**, also kept indefinitely: which pages were read most that\nmonth, which sites people arrived from and which source a tagged link named,\nwhich crawlers and automated scanners called, and which blurred networks the\nautomated probes came from — each as a name and a number of times. It is how we\ncan still answer \"what did people read last March\" once March's log is gone.\n\nWhat that ranking is not: it is never a record of what any one visitor did. A\npage has a count beside it and nothing else — not who asked for it, not when\nbeyond the month, not what else they read. The lines are never joined back\ntogether, because the pieces needed to join them are not kept.\n\nWe do not send the log or private reports to an analytics company. The private\nsummaries stay on our own computer. The public page totals below are the limited\npart shown to any reader.\n\n**Public page counts.** A page's counter estimates successful HTML GET requests\nfrom the existing masked web log, separately for each site and page. Known bots,\nprobes, assets and failed requests are excluded. Reloading can count again;\nthese are estimated page views, not unique people or evidence of a conversation\nor an order.\n\nRecorded views come from available log history. The collector can import older\nretained masked logs, preserving its original initialization record and the\nobserved log period. Older history can have gaps; known observations still count,\nwith that limitation disclosed. This is not a complete lifetime total, and old\ndaily or monthly report estimates are not added. The server updates the number\nroughly every thirteen minutes, without browser JavaScript, cookies, extra\ntracking requests or browser storage.\n\nThe counter keeps aggregate totals, observed time bounds and internal log-reading\npositions and file bookkeeping, not visitor identifiers or individual request\nhistories. Totals can outlive the raw log. A dash means the counter is not\ninitialized, is stale, or has lost reliable coverage of newly collected logs.\nKnown gaps in older imported history do not erase recorded observations. A valid\nzero means no qualifying request was observed in the available history. Existing\nraw-log retention and the daily and monthly report archives remain unchanged.\n\n**The other ways in.** This site is also served over Gemini and Gopher, two small\nprotocols with no access log at all, and over Tor. Since September 2026 we count\n**how many requests** the Gemini and Gopher copies answered on each day, and\nnothing else: no address, no page, no visitor. **We count nothing at all over Tor.** Visitor counters and analytics stay disabled for our onion service. Tor is designed to hide who accesses it.\n\nCounting them showed us something we had not known, so here it is. The Gopher\ncopy is started afresh for each visitor by the operating system, which names that\nmoment after **the address it came from** and writes the name into its own log.\nOur server software never sees it and no setting of ours could have stopped it.\nWe found it the day we first counted, and that log now expires after thirty days\nlike everything else here. It was keeping things for longer.\n\n**Where our own server is.** One machine, in Toronto, Canada. The log is written\nthere and stays there; the summaries are written from it on our own computer.\nThe [credits page](/credits/) names the server, along with everything else this\nsite is built on and served by.\n\nWe do not build profiles, and we cannot follow you across other websites. For as\nlong as a line is in the log it carries no name, no email address, no cookie and\nno complete address. It does carry your browser's full self-description, a\nblurred network, the language your browser asks for and a timestamp to a\nfraction of a second, all on one line. A rare browser on a small network narrows\nthat down further than we would like, so we write it here rather than tell you\nthat identifying you is impossible.\n\n## Getting in touch\n\nMail sent to [welcome@engineer.company](mailto:welcome@engineer.company) is read by us and kept in our own mailbox, hosted with\nApple iCloud. We use what you write to answer you. We do not add you to a mailing\nlist, and there is no mailing list to add you to.\n\nIf you want to know what we hold about you, or want it removed, write to the same\naddress and we will answer honestly — which, given the above, will usually be\n\"nothing that identifies you\".\n\nQuestions this page does not answer go to the same place —\n[ask us directly](/contact/) *(read by a person, not a pipeline)*.\n\n\u003chttps://engineer.company/privacy/\u003e\n","date_modified":"2026-11-12T05:34:23Z","description":"How this site handles data: no cookies, no tracking scripts, and a server log that never holds a full IP address.","formats":{"gemtext":"https://platform.engineer.company/privacy/index.gmi","gopher":"https://platform.engineer.company/privacy/gophermap","html":"https://platform.engineer.company/privacy/","json":"https://platform.engineer.company/privacy/index.json","markdown":"https://platform.engineer.company/privacy/index.md","pdf":"https://platform.engineer.company/privacy/index.pdf"},"kind":"page","language":"en","title":"Privacy","url":"https://platform.engineer.company/privacy/","version":1}