Privacy policy
This page describes what each app actually does with your data. Where an app sends something off your device, it is named below.
Wy5
Browsing topics and community results needs no account. Adding a list of your own, using groups and sending a report need Sign in with Apple. Wy5 asks Apple for no name and no email address, so what Apple passes on is the identifier it issues to Wyome for you. That identifier is the same every time you sign in, so it is not anonymous, though it means nothing outside Wyome. The Wyome server keeps it with your account, along with when the account was created, when it was last used and last saved a list, and whether Wyome has suspended it. When the server is set up to do so, it also keeps a token from Apple whose only use is to disconnect Wy5 from your Apple Account when you delete the account. Each sign-in starts a session that lasts 90 days, of which the server stores only a scrambled form. Signing out removes the session from the phone at once and asks the server to end it too; if the phone is offline at that moment, the session simply runs out.
On your iPhone, the session and Apple’s identifier are kept in the iOS Keychain, readable only on that phone and never carried to another one by a backup. iOS can keep Keychain items after an app is deleted, so reinstalling Wy5 may find you still signed in. Everything else Wy5 keeps on the phone is in one file: your lists, every earlier version of each, the topics you have opened, details of the entries you have picked or opened, and the display name you last used in a group. Like any app’s data, that file is included in your iPhone’s own backups. Signing out leaves it in place; Reset this phone, in Settings, clears it. Your submitted lists are also kept with your account, so signing in on a new phone brings them back.
A list is five entries in order, each matched to a Wikidata record. When you submit one, the server stores which five, in what order, against your account, with when it was first sent and last changed. It is counted into the topic’s community results with no name attached, and a topic’s results are withheld until at least three lists have been counted, because with fewer they would be somebody’s own list. Remove my list takes one list off the server and the results are recalculated without it; it comes off the phone too, though its earlier versions stay in List history there.
Groups are joined with an eight-character code, under a display name of up to 40 characters that you choose for each group. Everyone in a group sees its name, its code, who hosts it and every member’s display name, and can see any member’s full list, under their display name, for any topic that member has made one for — not only the topics made in that group. Behind the names, what the app receives also carries a number for each member’s account, the same in every group, so someone in two groups with you can tell it is you in both. Topics are made only inside groups. A topic made in a group is visible only to that group’s members and to anyone who already has a list on it: it is not listed in Browse and cannot be found by searching. A name that matches a topic already open to everyone takes you to that topic instead, and it stays open to everyone. The host can rename the group, change its code and remove members. A removed member’s display name is kept, visible only to the host, so that they can be let back in. When the last member leaves, the group is deleted.
You can block a member of a group you share. Wyome stores who blocked whom, when, and the display name they had at the time, so that Settings can list them. It is one-way and they are not told: in your groups you stop seeing their name, their lists and the topics they started, unless you have a list on one, and nothing changes for them. A block is removed when you unblock them in Settings, or when either account is deleted.
You can report a group, a member, a topic or an entry. A report stores who sent it, who or what it is about, the group or topic it concerns, copies of the names and titles involved as they were at that moment, and anything you write, up to 500 characters. It is emailed to Wyome, with the account numbers involved and the group’s code, and kept for Wyome to review. It is not shown to the group or to the person reported, and nothing is removed automatically. A report is kept even after the group it is about has been deleted. Separately, an automatic filter refuses certain words in topic titles, group names and display names. It does not apply to what you write in a report, so a report can quote what it is reporting.
Like every web server, the one at wy5.app records each request in a log: the time, your IP address, the app and iOS version your phone reports, and the address requested, which can include a search you typed or a group’s join code. These logs are kept for about a week. The site statistics page publishes totals drawn from them — requests per site per day, and per country, worked out as described under This website below — along with a short list of the most requested addresses across Wyome’s sites. Wy5’s addresses are never among them: because they can hold a search or a group’s join code, a request to wy5.app is counted, but everything in its address after the first part is dropped before anything is kept. To limit abuse, the server also keeps counts of recent requests for about a day, filed under a scrambled form of your account number, of Apple’s identifier while you sign in, or of your IP address and the date.
Looking things up happens between your phone and Wikimedia directly. As you type an entry, the phone sends what you have typed to Wikipedia and Wikidata to find matches, fetches summaries from Wikipedia, and loads pictures from Wikimedia Commons; only freely licensed pictures from Commons are shown, and results pages load them too, signed in or not. Wikimedia sees your IP address and what you searched for, and nothing about your account. When a list names something Wy5 has not seen before, Wyome’s server checks it with Wikidata, sending only the record’s identifier.
Find on Amazon opens an Amazon search for the entry’s title and creator, carrying Wyome’s affiliate tag, which is the same for everyone, so Amazon can attribute a resulting purchase to Wyome and we may earn a commission. The app adds nothing about you to the link. Once you are on Amazon, Amazon sees you as it sees any visitor, including whether you are signed in to it. This is the only commercial tracking in any of the five.
Wy5 is free and has no in-app purchases. It contains no advertising, no analytics and no third-party code, and nothing in it tracks you across other apps or websites.
WyHome
Everything you record is kept on your device. When you are signed in to iCloud it is also synced between your iPhone, iPad and Mac through your own private iCloud database, including the files you attach; Apple holds that copy under your Apple Account, and Wyome cannot see it. There is no Wyome account and no analytics. One thing leaves your devices otherwise: finding a contractor. Searching sends the search text to Apple Maps, along with your location if you allow it, so results can be local. If you search near your home address, that address is sent to Apple to turn it into a map location. Reminders and calendar entries are written into your own Apple apps.
WyAuto
Your vehicles, entries, service schedules and receipt photos are kept on your iPhone. When you are signed in to iCloud they are also synced through your own private iCloud database, so your other devices see them; Apple holds that copy under your Apple Account, and Wyome cannot see it. There is no Wyome account and no analytics.
Two things leave the phone, and only when you ask. Looking up a vehicle’s rated figures sends its VIN to the US National Highway Traffic Safety Administration’s vehicle decoder, and its year, make and model to the US Department of Energy’s fueleconomy.gov, to fetch the official mpg and the fuel grade it requires. Nothing else goes with them, though like any web request those services see your IP address. Scanning an odometer, pump or receipt reads the text on the iPhone itself; the photo goes nowhere except, kept as a receipt, into your own iCloud sync.
WyHealth
This is the only one of the five that handles health data, and it is the only one where what you store leaves your device by design rather than as an aggregate. The iPhone app reads the Apple Health categories you approve and uploads them; the website reads them back. You choose the categories when the app first asks and can change or withdraw them at any time in iOS Settings under Privacy & Security → Health. Signing in uses your Apple account and no password is ever stored. Unlike Wy5, WyHealth does ask Apple for your email address, so your account can be found again and so you can be written to about it; if Apple gives a relay address in its place, that is what is kept.
What the app uploads is the readings themselves, not a summary of them: each figure with its unit, when it was recorded, and the name of the app or device that recorded it. Steps, distance, energy, exercise and stand minutes and flights climbed are added up into one figure per day on the phone before they leave it; heart rate, blood oxygen and respiratory rate go up as a day’s low, average and high; weight and other body measurements, blood pressure, blood glucose, resting and walking heart rate, heart-rate variability, cardio fitness and recovery go up reading by reading. Workouts are sent as the activity, its start and end, duration, energy and distance — never the route, and never your location. Sleep goes up as its stages, and the heart notices Apple Watch raises — high, low and irregular rhythm — as the time each was raised. The app also sends an identifier for your phone, so a later sync can pick up where the last one stopped. On the website you can fill in a profile — a display name, date of birth, sex, a goal and your preferred units — and that is held with the rest.
The website also works out a weekly training plan for you from your readings — chiefly your resting and maximum heart rate, how much you have trained in the last four weeks, and the goal on your profile. It does this with fixed rules written into WyHealth itself; nothing is sent to an outside service or an AI model to do it. Each plan is stored with the rest of your data, together with a copy of the figures read when it was made, so the site can tell when your readings have moved enough to need a new plan and can show you why it changed. A new plan is added alongside the earlier ones rather than replacing them, and they are deleted along with the rest of your data.
All of it is held by Supabase, a database service Wyome rents, on Amazon’s servers in Oregon, in the United States. Both the app and the website talk to it directly. The website’s own server, the one that draws your dashboard, reads your data from there each time you open a page and keeps no copy; it is the same DreamHost server as the other four sites, and what it logs is what every web server logs — the page requested, the time, your IP address and browser — never a reading. Every row is scoped to your account in the database itself rather than by the code that queries it, so one account cannot read another’s data even if a query is wrong. Readings and values are never written to server logs — the logs record how many samples arrived and of what type, never what they said. wyhealth.app loads nothing from anyone else: no analytics, no third-party script, and its fonts are served from the site rather than from a font network.
Your health data is not sold, not shared with advertisers, not used to train anything, and not shown to anyone else. There is no public element to WyHealth as there is to Athenana: nothing you store appears on a public page, and there is no community aggregate.
Lab results and a paid subscription are planned and not yet built, and nothing is sent to anyone for either today. When they arrive, the text of an uploaded lab PDF would be sent to Anthropic’s Claude API to read the marker names and values off it, and payment would be handled by Stripe so that Wyome never sees your card details. This page will be updated before either is switched on.
Athenana
Items you save are stored on a server operated by Wyome. When you save a link, the server fetches that page and, once, the pictures inside its article; it keeps a copy of the page, the pictures and a screenshot, and uses them for tagging and search. Reading the item afterwards shows those pictures from the copy stored on Athenana’s own server and never from the site, so the site is not told when you come back to it; a picture the server could not fetch is simply not shown. You can sign in with your Apple account, with a passkey, or with an email address and a password; a password is stored only as a one-way hash, never in a form that can be read back, and a passkey stores only a public key. You can also create an account with an email address alone: Athenana emails that address a link, and no account exists until the link is followed. Until then the server keeps only the address, and deletes it within a day of the link expiring an hour after it was sent. The password is chosen after following the link. If the address already has an account, the email says so instead of offering a second one, and the sign-up page answers identically either way. Pages and images you save are sent to Anthropic’s Claude API so they can be tagged automatically, so text in images can be read, and so a saved page can be summarised; that content is not used to train models. Searching with a question or several words also sends your search words, and the titles, tags and opening lines of the matching items in your library, so the results can be ranked by meaning. Pressing Answer from these also sends up to 2,000 characters from each of the five best-matching items (your note, its summary or its saved text), so an answer can be written from them. The same API ranks the morning reading list described below.
You can also save into Athenana from a browser extension for Chrome, Edge, Firefox and Safari, and from a Shortcut on iPhone and iPad. The extension sends the page’s address and title, its description and preview image where the page publishes them, and a screenshot of the visible tab — plus the text you had selected, if you saved a highlight. It sends these only when you click its button or use its right-click item: it does not read the pages you visit otherwise, and it does not run in the background, except to re-check the pages you have asked it to watch. For a watched page it asks for that one site, fetches the page in your browser about once an hour — with your own sign-in in Chrome; Safari sends none — and sends only the titles, links and a short excerpt of each item it finds to your server. It also fetches the page once when you pick it, to check the items are there. A page you stop watching is never fetched again. Because that screenshot is taken by your own browser rather than by the server, it shows the page as you were seeing it, including anything you were signed in to. The Shortcut saves a link or a photo you share to it, using a token that can do nothing but add items.
Nothing you save is published unless you turn publishing on for your account, which is off until you do. Each item then sits at one of three levels. Private is yours alone. Blog gives it an address of its own and puts it on your blog. Public is all of that and adds it to Athenana’s shared front page. A single account setting still hides everything whatever the individual items say, and it is deliberately a separate switch from the one that turns publishing on — an emergency stop and an opt-in are different things.
The shared front page carries saved links and nothing else. A note, a highlight and an image can reach the blog level but never that page, because it is an anonymous stream and those are your own words, your own choice of passage and your own picture. Screenshots are never published anywhere, because a screenshot photographs the tab as it was and can catch a page you were signed in to. The pictures on the front page come from the publisher of the page being linked to, and a private comment you attach to a link is not published with it.
That front page is also a feed, at feed.xml and feed.json, and a blog carries its own. Worth knowing before you publish something: a page can be edited or taken down, but a feed entry has already been fetched and stored by whatever was subscribed to it.
A blog exists only if you choose a handle for it on your account page, and it lives at athenana.com/blog/ plus that handle. No handle means no blog and no address; the handle is the whole opt-in. It is independent of the shared front page, so you can have either without the other.
A blog is also an account on the fediverse, which anyone on Mastodon can follow; a follow is accepted on its own, since everything on a blog is already public. When a note, a photo or a highlight is set to Blog or Public it is delivered, signed, to every follower’s server, which keeps its own copy and shows it there — a link you kept is never delivered, only what you wrote, and a delivered post cannot be taken back from servers that already hold it. Athenana stores each follower’s account and inbox address, their name, their server and whether it still answers; the follower count is public and the list is not. Replies, likes and boosts sent back are discarded, not stored. Clearing the handle takes the blog and the account down.
If you connect a Bluesky account, each new blog post is also posted there — its text, its link and any pictures already on the blog — and deleted there if the post leaves the blog. Nothing private, and no summary, is ever sent.
When a link is on the front page, the site it points to is told so, using the web’s standard Webmention notification; it carries the front page’s address and the link itself and nothing else, and sites that accept these often display them publicly. When a link leaves the front page, a follow-up notice is sent so the site can drop the mention.
An article can also be narrated and delivered to you as a podcast. When you ask for that, the article’s text is sent to Google’s speech service and comes back as audio, which is kept on Wyome’s server and played from there — your podcast app never contacts Google. It happens only for articles you flag. The feed has no password, because podcast apps cannot sign in: it is protected by a long unguessable address you can replace at any time, so treat it as a key rather than a link. Narrated audio is deleted after 90 days; the article itself stays.
An article can also be sent to your Kindle. When you ask for that, the article’s text, its picture, its address and any passages you kept from it are emailed to the Kindle address you saved, and Amazon converts the file and keeps it as a personal document in your Amazon account. Nothing else about the item goes with it — not its tags, and your note only if you tick the box for it — and nothing is sent unless you press the button, one article at a time. The address is accepted only if it is at kindle.com, so it cannot be used to mail your reading anywhere else. Wyome’s server sends it; your browser never contacts Amazon.
When you save a Google Maps or Apple Maps link, Wyome’s server sends Apple the pin’s coordinates, once, to draw a picture of the map that is kept on the server; nothing else about you or the item goes with it, and the link is saved private, so the map appears on your front page or blog only when you choose to publish it. Pressing Explore map loads Apple’s interactive map from Apple’s servers into your browser, which then talks to Apple directly; nothing contacts Apple until you press it.
A saved link can also be shared to eleven services: Bluesky, Mastodon, RecLeague and Hacker News, which your browser goes to, and micro.blog, Are.na, Pinboard, Raindrop, Readwise Reader, Karakeep and MyMind, which Athenana’s server posts to with a token you gave it. What leaves is the same for all eleven — the original address you saved and the item’s title, and for two of them the item’s picture, never the archived copy, the screenshot, the summary, the tags, your note or the passages you kept — and nothing in the Share block is shared unless you press the button, one item at a time. The four your browser goes to open that service’s own composer with the address and title filled in, so the service sees your IP address and those two things whether or not you then post; Mastodon’s is the server you named, and RecLeague is one of two places a picture goes with the link, only for an item you have published. The seven the server posts to receive the title and the address as a post on your micro.blog; the address and the item’s picture — the one you chose or the page’s own, never a screenshot — as a block in the Are.na channel you chose, served from Athenana at an address that stops working after a day, whether or not the item is published; or the address and title as a bookmark in your own Pinboard, Raindrop, Readwise Reader, Karakeep or MyMind account, MyMind through the same key you gave for importing from it. No button exists until its token does.
Athenana counts how often each link on that front page is followed. A link there that points to another site goes through Athenana’s server before it takes you on, so that the following can be counted; what is stored is which link, and when. Not who: no address, no cookie and no identifier of any kind is kept beside it, so nothing there can tell one reader from another, or the same reader twice. Requests that say they are robots, and the invisible fetches a browser makes to speed up pages you have not clicked, are not counted — a robot that does not say so is. Visits to the front page and to each link’s page are counted too — a number per page per day, with no address, cookie or identifier kept beside it, and each page shows its own total for the last ninety days. The counts are deleted after ninety days, and nothing on a blog, in a feed, or behind a sign-in is counted at all.
Those counts choose the five links in a weekly email you can ask for. If you ask, your address is stored so the email can be sent to you, and you are sent a link to confirm you meant it: nothing goes to an address that has not followed that link. It is used for nothing else and given to nobody, every issue carries a link that removes you, and the email carries no tracking — its links go straight where they point, and there is no hidden image that reports whether you opened it. The pictures and typeface in it come from Athenana’s own server and are the same for every reader, so nothing in the message can tell anyone that you opened it.
A blog on Athenana can offer its posts by email, if its writer turns that on. It works the same way: you confirm before anything is sent, your address is used for that blog’s posts and given to nobody, the writer sees only how many people subscribe, and every email carries a link that removes you and nothing that reports whether you opened it.
Some sites refuse to serve a page to a server while serving it normally to a browser. When that happens Athenana asks the Internet Archive for an older copy, which means that page’s address is sent to archive.org. Only the address, and only for a page that could not be fetched directly.
Athenana also builds a reading list for you each morning, and building it means searching outside your own collection. It takes fifteen topics — the ten tags you use most, plus five drawn at random from the rest of the tags you use — and searches three places for each: Google News, Hacker News (through Algolia, which runs its search) and Flipboard. Only the topic word is sent, by the server rather than by your browser, with no identifier, no cookie and nothing about what you have saved; what comes back is a list of links to follow if you want them, and nothing on that page loads from any of the three. What those three do see, over time, is what this account reads about — the topics come from your own tags rather than from a generic list. A topic you drop on the Discover screen is never searched for again. The same morning, Athenana writes a short briefing of what your feeds carried, from their headlines and one-line summaries, and — for feeds you tick for it — five short facts per story, fetching the article itself when the feed’s summary is only a line. Both are written by Anthropic’s Claude API from text the server sends; your browser contacts nobody, and nothing from your own library goes into the briefing. The feeds you have told Athenana you read elsewhere are fetched each night by the server too, listed under Regular feeds, and given a briefing of their own the same way. Opening a story from that list works the same way: the server fetches the article and its pictures once, keeps them for a day, and shows them to you from Athenana, so the publisher sees the server rather than you. A page the server cannot read offers you a link to the publisher instead, and that visit is your browser’s. When you open one of the Google News stories, the server first asks Google where it lives, sending only that story’s link, so Google learns the story was opened by the server rather than by you.
A saved song gets a play button, and so does a saved YouTube link. Along with a saved podcast episode, described next, it is one of the few things here that can reach somebody reading a page rather than the person whose account it is. When you save an Apple Music link, Athenana recognises it as a song from the address itself — it makes no extra request to Apple to do that — and asks YouTube whether it has the same recording, sending the artist and title and nothing about you. When you save a YouTube link there is nothing to ask: the address already names the video, so the server only fetches its cover picture. Either way, that cover is copied onto Athenana’s own server. That copy is the point: a page with a video on it loads nothing from Google and tells Google nothing, not even that the page was opened. Pressing play is the moment that changes. The player then loads from youtube-nocookie.com, and Google sees your IP address and which video you played. Play buttons can appear on the shared front page and on blogs, not only on your own pages, so this is worth knowing whether or not you have an account.
A saved podcast episode gets a play button too. Athenana reads the episode’s details — the show, the title, how long it runs — from the page it saved, and copies the artwork onto its own server, so a page carrying an episode loads nothing from anyone until you press play. What plays is not stored here: the audio comes from wherever the show is hosted, which differs from show to show and is often a redirect that exists to count downloads. That host sees your IP address and that the episode was played. Like the video player, the button carries no address for the audio until you press it, and like the video player it can appear on public pages as well as your own. Athenana’s Discover lists, its feed search and its Feeds page carry both buttons as well — for a podcast or video file one of your feeds publishes, and for a YouTube story — and there too nothing is loaded from the show’s host or from Google until you press play. An episode you save from one of your feeds keeps its play button for you, and appears on your public pages without it. Where you stopped listening is remembered in your own browser, not on Athenana’s server.
You can also bring a library in as a file — the bookmarks file any browser exports, a Pinboard, Reeder, Instapaper or Pocket export, or Safari’s Reading List, from its export or a copy of its Bookmarks.plist. It is read on Athenana’s server and nothing in it is sent anywhere: no page in it is fetched during the import, and the file itself is deleted as soon as its rows are in. A bookmark marked private there is saved as private. Everything from a Reading List is saved as private. Reading the pages afterwards, a few hundred a day, is the same process described above for anything you save.
You can also forward an email into Athenana, to an address of your own that only you know. The message is saved as a page: its text, and the pictures it carries. A PDF attached to it is saved as a document of its own, private like the message; nothing else attached is kept. A picture the sender attached inside the message needs no request — it arrived with it. A picture the message merely links to somewhere else is fetched once, by the server, when the message is saved, and a copy is stored. Newsletters carry invisible images whose only purpose is to tell the sender that a message was opened, and by whom, so anything declaring itself 1×1 or 0×0 — the shape of a tracking pixel — is never requested. A tracker that declares an ordinary size is requested like any other picture, so this reduces that tracking rather than preventing it. Reading a saved message shows you its words and its pictures, each one from the copy the server stored and never from the sender — a picture it could not fetch is simply not shown. The stored pictures also become the item’s thumbnail and the photograph Athenana takes of the message itself, which is where you see how it arrived. Every one of them is served from Athenana’s own server, so reading a saved message contacts nobody.
If you ask for a second, separate address, newsletters can be sent there to be kept rather than saved. Each issue is kept whole, with who sent it, for 90 days from the day it arrives, or until you tell Athenana to forget that sender; this is the only place Athenana keeps a sender’s address or a raw message, and the forwarding address above is unchanged. If you tell Athenana to forget a sender, that sender’s address is kept on a short list so later mail from it can be refused, until you undo that. Nothing in an issue is fetched from anywhere until you save it: reading one shows its words without pictures, so the sender is not told you opened it. Saving it works exactly like forwarding the same message, with its pictures fetched once and stored, and the sender and raw message not kept. Until you save one, issues are never sent to the AI briefing or any model.
You can also post into Athenana from any third-party app that speaks Micropub, the web’s standard for that. Such an app works only with a token: one you issue to it from your own account, or one it asks for over IndieAuth, where you are shown the app’s address and what it wants to do and, if you approve, it receives a token for that and your blog’s address as your identity. Either can be revoked from your account.
You can connect an AI assistant, such as Claude, to read your Athenana library. It sees what you see there, private items included but never screenshots, and it can change nothing. What it reads goes to that assistant’s company under its own terms, so connect only one you trust. Each connection has its own token, listed under Settings → Tokens, and revoking it stops it at once.
A paid Athenana plan is bought in one of two ways. On the web it is taken through Stripe: you type your card on Stripe’s own page, Stripe is given your name and email address, and Wyome never sees your card number; Athenana keeps Stripe’s identifiers, your plan, whether it is paid, and the card’s type and last four digits. In the iPhone app it is an App Store subscription sold by Apple, which takes the payment and keeps your card. The app hands Apple a random token Athenana made for your account, so the purchase can be matched to it; Apple then reports the subscription to Wyome’s server — its transaction number, plan, price, dates and whether it will renew — and sends no card, name or address. Refunds of App Store purchases are Apple’s to give, and you cancel one in your Apple account settings. With your own Anthropic key, Athenana’s calls to Anthropic go to your account and nothing about them is recorded. On the Hosted plan they use Wyome’s key, and the server records how many tokens each call used, never what was in it; those counts outlive the account, with nothing left to say whose they were. When you subscribe on the web, the time and the renewal wording you were shown are kept, and in the app, which version of its subscription screen you bought from; and each sign-in records its time and method — never an IP address — for a year, so that if a payment is disputed, those records and counts of how the account was used can be sent to Stripe for your bank. Billing emails come from Wyome’s server: what you signed up for, a reminder before a yearly plan renews and once a year on a monthly one, and a note when a card is declined or Apple cannot renew, a plan lapses or free access ends. A web payment a card network flags as likely fraud is refunded in full and the subscription cancelled at once. A payment Apple took is Apple’s to settle if it is disputed, and nothing about the account is sent to Apple for it.
Common to all
None of the five carries advertising, and none carries a third-party tracker that runs on its own. The one thing counted anywhere is Athenana’s tally of how often each front-page link is followed, described above — a number against a link, kept on Wyome’s own server, with nothing beside it that identifies anyone. There are two third-party players anywhere — the YouTube player behind a saved song’s play button in Athenana, and the audio player behind a saved podcast episode’s — and each loads only when you press play, as described above. The one commercial identifier anywhere is Wy5’s Amazon affiliate tag, described above. No data from any of them is sold, and none is shared with anyone except where this page says otherwise — Apple Maps for WyHome’s contractor search, iCloud for WyHome’s sync between your own devices, Wikipedia and Wikidata for Wy5’s look-ups, Amazon for Wy5’s shopping links, the Claude API for Athenana’s tagging, summaries and reading-list ranking, Stripe when you pay for an Athenana plan on the web, Apple when you pay for one in its iPhone app, Amazon when you send a saved article to your Kindle, Apple when Athenana draws a saved map link’s picture, MyMind when you import your library from it with a key you gave, Bluesky, Mastodon, RecLeague, Hacker News, micro.blog, Are.na, Pinboard, Raindrop, Readwise Reader, Karakeep or MyMind when you share a saved link to one of them, the Internet Archive when a site refuses to serve Athenana a page, Google when Athenana looks for a saved song’s video or you press play on one, a podcast’s own host when you press play on a saved episode, Google News, Hacker News and Flipboard when Athenana searches for your morning reading list, an AI assistant’s company when you connect one, and the Webmention notices Athenana’s front page sends to the sites it links. All five sites run on servers at DreamHost, which hosts them and can see whatever is on them; WyHealth’s health data is the exception: it is held by Supabase in the United States, and the DreamHost server only reads it to draw your page.
This website
wyome.com itself sets no cookies, runs no JavaScript and contains no analytics. Its fonts are served from this site rather than from a font network, so loading a page here contacts nobody but this server.
Like nearly every web server, this one records each request it serves in a log: the address of the page requested, the time, the requesting computer’s IP address, the browser it identified itself as, and — where the browser sends one — the address of the page you followed a link from. These logs are kept for about a week and then deleted automatically. They are not analytics: nothing is added to a page to produce them, no identifier is set, and no visitor is followed from one page to the next.
The site statistics page publishes two narrow slices of those logs. The first is referrals: the addresses of pages that linked here, how many times each was followed on a given day, and which page it led to. The second is counts — how many requests each site received per day, and how many came from each country.
The country map is worked out from IP addresses, and that deserves saying plainly. The address is compared against the public register of which blocks are allocated to which country, on this server, using a copy of that register held here; no address is sent anywhere to do it. Only the resulting per-country totals are kept. The addresses themselves are never stored beyond the ordinary log described above, never published, and cannot be recovered from the page: a country total is the only thing that survives. Nothing on the page carries an IP address, a browser identifier, a time of day, or anything else describing an individual visit, and there is no way to work back from it to a person.
Data retention & deletion
Wy5 keeps your account and your submitted lists for as long as the account exists, so the lists stay counted in the community results. You can delete the account from within the app, in Settings. Apple asks you to confirm it is you once more, so that Wy5 can also be disconnected from your Apple Account, and the receipt afterwards says whether Apple confirmed that. Deleting removes the account, every list, every session, your memberships and display names in every group, including any record of being removed from one, and any block you made or that names you; the community results your lists counted in are recalculated without them. A group you were the only member of is deleted with you; in one that has other members, hosting passes to whoever has been in it longest. On the phone, your lists, their history, the topics and entries it had stored and the display name it remembered are cleared, and you are signed out. Some things outlive the account with nothing left to say they were yours: the titles of topics you started, the names of groups you started that others are still in, and reports — the ones you sent and the ones about you — with the names, titles and text copied into them, as well as the emails Wyome received about them. Server logs age out after about a week, and the abuse counts after about a day. If you would rather ask, support@wyome.com will do it.
WyHome stores nothing on a Wyome server, so there is nothing of yours for us to delete. On an iPhone or iPad, deleting the app removes that device’s copy; on a Mac, the app’s data folder stays until you remove it yourself (it is in ~/Library/Containers/com.wyome.WyHome). If iCloud is on, the records are also in your own iCloud account, and removing the app does not remove that copy; you can delete it from your device’s iCloud storage settings. Deleting a job, property or file in the app deletes it on every device signed in to the same account.
WyAuto stores nothing on a Wyome server either, but if iCloud is on, its records and receipt photos are also in your own iCloud account, and removing the app does not remove that copy. Deleting a vehicle in the app deletes it, with its entries, receipts and schedules, on every device signed in to the same account.
Athenana keeps what you save until you remove it. You can delete your account from within it, under Settings: it removes the account and everything saved in it, including anything in the trash. If you would rather ask, support@wyome.com will do it.
A lapsed Athenana plan keeps sign-in, reading, export and deletion, and hides your blog rather than deleting it. After 180 days lapsed, or in dispute with your bank, the account is deleted with everything in it, with an email 30, 7 and 1 days before. Whatever happens to a subscription, you can always export everything you saved, as text; the pictures and PDFs themselves stay on the server. Deleting the account cancels a web subscription first, and Stripe keeps its own record of your payments; it does not cancel an App Store subscription, which Apple goes on charging for until you cancel it in your Apple account settings.
A new Athenana account opens when it subscribes, and before then it can already save from the iPhone app. Those saves are only stored on Wyome’s server — nothing is fetched from the sites they point to and nothing is sent to an AI service until the account opens — and it holds up to 200 saves. An account not opened within 30 days is deleted with everything saved to it, unless it currently has free access from us.
WyHealth keeps your readings until you ask for them to be removed. There is not yet a delete button in the app or on the site; write to support@wyome.com and the account, its readings and its profile are deleted together. Withdrawing Health access in iOS Settings stops new readings leaving the phone but does not remove what was already uploaded.
Questions: support@wyome.com.