Privacy Policy / 隱私權政策

English 繁體中文

English

Effective date: November 7, 2026

This English version is a translation of the Traditional Chinese original. If the two versions conflict, the Traditional Chinese version prevails.

This document describes what the Podling app (the “Service”) collects, where it sends it, how long it keeps it, and how it is deleted. Podling is in beta. If any statement here does not match what the Service actually does, that is a defect; please let us know.

Summary

Who collects your data

The Service is developed and operated by Coyotech, which collects and controls the data described below. Contact: karta09104@gmail.com.

What we collect, and why

Anonymous device identity

On first launch the app asks the server for a token and stores it on your device (in the app’s local preferences). Every later request carries it, so the server knows which data is yours. The server stores only a SHA-256 hash of the token, when it was created, and when it was last used (updated at most once a day). If the server ever does not recognize the token, the app gets a new one and keeps the few most recent old ones on the device, so it does not lose your data over a mistaken rejection. Uninstalling the app deletes the token from your device but not the data on the server.

On the server, the token points to a user, and your data belongs to that user. An anonymous user has one phone; a signed-in account can have several phones’ tokens pointing to it, which is how your data follows you between phones.

Account (optional)

When you sign in with Apple or Google, we store which provider it was, the user identifier that provider gives us for you, and the email address in its sign-in token (Apple may give a private relay address; we keep the newest one each time you sign in). We do not store your name, and we do not store any password, token or other credential of the provider: the sign-in token is checked once against the provider’s public keys and discarded.

If you sign in on a phone where you were not signed in yet, that phone’s data joins the account. If the account already exists (you signed in on another phone first), this phone’s data is merged into it: the same word or phrase becomes one entry with the examples from both, and its status and meaning are the ones changed last; follows, listen-later, playback progress, lookup history and daily activity move into the account. This phone’s anonymous user is then deleted. A merge cannot be undone.

Signing out deletes this phone’s device record on the server; the phone gets a new, empty anonymous identity the next time it connects. The account, its data and its other phones stay.

Subscriptions and purchases

Plus and Pro are subscriptions bought through the App Store or Google Play, and a purchase requires an account. Apple or Google handles the payment; we never receive your card or payment details.

RevenueCat, Inc. (United States) is the service provider that keeps the subscription status for us. The app gives RevenueCat’s SDK a random user identifier of ours (not your token), and RevenueCat receives from the app and from the stores: your purchases (which plan, when it was bought and when it renews or ends, the store’s transaction identifiers), and the technical details its SDK sends with every request: platform, operating system and app version, language settings and the store’s country, and on iPhone Apple’s identifier for vendor (an identifier shared only by one developer’s apps on that phone). Like any server, it also sees your IP address. Our server stores, for a signed-in account, only what it reads back from RevenueCat: the plan, the store, when it ends, and whether it is a trial, will renew, or has a billing problem.

The app talks to RevenueCat only when you open the plans page, or at start when you are signed in. If you never open the plans page and never sign in, nothing is sent to RevenueCat. If you opened it without signing in, RevenueCat keeps that random identifier with no purchases on it; once you delete your data, or that anonymous user is merged into an account, the identifier no longer matches anyone on our side.

Your library

For your token we store: the shows you follow (and when you last looked at each), the episodes on your listen-later list, and for each episode you played, the position you reached, whether you finished it, and when.

Lookups and vocabulary

When you tap a word or phrase in a transcript, we store that you looked it up: which transcript sentence, which words in it, the language of the meaning, and when. When you save a word, phrase or pattern, we store its text, its meaning, its status (new, learning, mastered) and the example sentences you saved it from. The meanings the model produces are also cached on the server by word and language, shared across all users and not linked to you.

Daily activity

The app reports how many seconds you listened or practiced shadowing (listen-and-repeat). For each report we store its kind, the seconds, your phone’s local date and a random event identifier, to show your daily goals and streak. The app does not record your voice; shadowing replays the original audio and pauses for you to speak aloud on your own.

Shows you add

If you paste a podcast RSS address, the server reads that feed and adds the show to a catalog that every user shares; the address is stored there, not under your token (the show is linked to you only if you follow it). Do not add a private feed whose address contains a personal key: it would be stored in the shared catalog.

Searches

The search term you type is sent through our server to Apple’s search service (see below). The server does not store it. It keeps the results in memory for up to 6 hours so an identical search does not call Apple again, and counts how often your token searches in order to rate-limit it.

Transcripts, translations and study sheets

When you ask for an episode’s transcript, the server downloads the episode’s audio from the publisher, transcribes it, and stores the transcript, the study sheet made from it, and a Chinese translation if you ask for one. These belong to the episode and are shared by everyone who plays it; they are not part of your data and contain nothing about you. The server’s temporary copy of the audio is deleted once the episode is processed, and in any case by a 7-day rule. Google Cloud Storage keeps a deleted object recoverable for a further 7 days, so a copy can exist for up to 14 days after it was made.

Usage analytics

To see how Podling is used and where it can improve, the app sends usage events to Google Analytics for Firebase, the same Google Analytics property that the Podling web app and trypodling.com use. The events are: which screen you open; searching (the search term, lower-cased, with email addresses and long numbers removed); adding a show; starting an episode and reaching 25, 50, 75 and 100 percent of it; asking for a transcript; looking up and saving a word or phrase; starting and finishing shadowing; turning bilingual subtitles on or off; opening a study sheet; opening, starting, completing or failing a sign-in (with Apple, Google or email); signing out; deleting your data; opening the plans page; starting, completing, failing or restoring a purchase, and opening subscription management. They carry the episode and show numbers, the plan and price involved, and a short error code when something fails, never your email address, your name, the words you look up, or any other text you type.

With them go: the app instance ID, a random identifier the Firebase SDK creates for this installation; while you are signed in, your account’s random user identifier (the same one given to RevenueCat, not your token or email), so that the app and the web app count one person once; whether you are signed in, and your plan. The SDK also sends your device model, operating system and app version, and language, and Google derives an approximate location (country, region, city) from your IP address. The app does not use the advertising identifier (IDFA, or Android’s advertising ID), Apple’s identifier for vendor or Android ID for analytics; ad storage, ad personalisation and Google signals are off.

Whether it is on depends on where you are, as on the web app. While you have not chosen, the app asks our web server (app.trypodling.com), which tells it from your IP address’s country whether consent comes first; that answer is not stored. In the EEA, the UK and Switzerland, or when the answer does not come, nothing is sent until you choose “Allow” in the app’s one-time prompt. Elsewhere it is on, and this policy is the notice. Either way you can turn it off or on at any time: Learn tab, settings icon, Privacy, “Share usage analytics.” Turning it off stops it at once. Your choice is kept on your phone.

Google keeps the events for the period set in our Google Analytics data retention setting. Deleting your data in the app does not delete events already sent; email us if you want them deleted.

On your device only

The app writes a local file of timing measurements (for example how fast the lookup card appeared). It is never uploaded; it leaves your phone only if you open the app’s measurement page and share the file yourself. That page is hidden in the released app. Podcast covers are downloaded and cached by your device from the hosts that serve them.

System logs

When a request fails, the app server’s system log records the error message. The web server is not configured to write an access log; its own system log may record your IP address when a connection fails. These are used to fix faults and kept for the server’s log rotation period. We do not deliberately log the content of your library.

What we do not collect

Where your data goes

Apple and Google (sign-in)

If you choose to sign in, the sign-in screen is Apple’s or Google’s, and they process what you enter there under their own privacy policies. Our server fetches their public keys to check the sign-in token. On Android, signing in with Apple runs in Apple’s web page and returns through our server, which passes the result back to the app without storing it. When you delete an account that signed in with Apple, our server sends Apple the authorization code from your confirming sign-in, to revoke Podling’s access.

RevenueCat (subscriptions)

See “Subscriptions and purchases” above. Our server asks RevenueCat for a signed-in account’s plan after a purchase or restore, when RevenueCat notifies it of a change, and when a renewal is due; RevenueCat’s data is processed under its terms as our service provider.

Google Analytics for Firebase (usage analytics)

See “Usage analytics” above. Google processes the events as our service provider under the Google Analytics terms; it receives them directly from your phone, so it sees your IP address.

Apple (podcast search)

The search term is sent from our server to Apple’s iTunes Search API, together with a country code (Taiwan). Apple sees our server’s IP address, not yours, and not your token.

Google Gemini API (meanings, study sheets, translations)

For a lookup, the server sends the Gemini API the tapped words, the transcript sentence they are in with the sentence before and after, and the language of the meaning. To make a study sheet or a translation, it sends the episode’s transcript text. It does not send your token. The text is processed under Google’s Gemini API terms.

OpenAI (speech-to-text)

To make a transcript, the server converts the publisher’s episode audio and sends it, in pieces of at most 5 minutes, to OpenAI’s whisper-1 transcription API. It is the publisher’s audio, not anything you recorded, and your token is not sent. The audio is processed under OpenAI’s API terms.

Podcast publishers and their hosts

Our server fetches the show’s RSS feed, the episode’s audio and, where the feed links one, the official transcript from the publisher’s hosts; they see our server’s IP address. Your phone streams the audio you play directly from the publisher’s host, and loads cover images from the hosts that serve them, so those hosts see your IP address and what you play, as they would in any podcast app. We do not control them; their own policies apply.

Your device’s speech engine

The pronunciation of a word is spoken by your phone’s built-in text-to-speech engine, and we do not send the word to anyone for it. Which engine that is, and whether it works offline, is up to your phone’s settings.

Google Cloud (hosting)

Our server, database and file storage run on Google Cloud. All communication between the app and the server uses HTTPS.

Where data is stored

Our server and database are in Google Cloud’s us-west1 region (Oregon, United States), and so is the file storage. RevenueCat processes data in the United States. Apple, Google (Google Analytics included) and OpenAI may process data in the United States or other countries. By using the Service, you acknowledge that your data will be transferred outside the country or region where you are.

Retention and deletion

Data tied to your token has no automatic deletion period: not after inactivity, and not when you uninstall the app. It is kept until you delete it in the app; an account and its data likewise stay until the account is deleted. Purchase records are also kept by Apple or Google under their own policies. Usage analytics is kept by Google for our Google Analytics data retention setting. The only other expiry periods are those stated above: the audio copy (up to 14 days, counting the 7 days Google Cloud Storage keeps a deleted object), the search cache (up to 6 hours), and backups (below).

Deleting your data

In the app, open the Learn tab, tap the settings icon, choose “Delete my data” and confirm. The server deletes, in a single step: your device record, follows, listen-later list, playback progress, lookup history, saved vocabulary and examples, and daily activity. Deletion takes effect immediately and cannot be undone. Before asking, the app tries to send any listening time it had not yet sent, so that it is deleted with the rest. After the server confirms, the app forgets your token (and the few old tokens it kept) and drops any such time it could not send, so the next time it talks to the server it registers as a new, empty installation. If the server cannot be reached, nothing is deleted and the app says so; try again later.

When you are signed in, the same place reads “Delete account.” It deletes the account and all of the data above, and every phone signed in to it is signed out. If you signed in with Apple, you are asked to sign in with Apple once more to confirm, and Podling’s Apple authorization is revoked; if that confirmation cannot be done, you can still delete the account, and then you remove Podling yourself in your Apple ID settings (Sign in with Apple). The account’s RevenueCat record is deleted too. Deleting the account does not cancel a subscription: cancel it in the App Store or Google Play subscription settings. If you do not, the store’s next renewal makes RevenueCat create a record for that identifier again, which no longer matches anyone on our side.

If you uninstalled the app, or cannot use it, email karta09104@gmail.com. If you signed in, write from the account’s email address and we can find the account by it. Otherwise we hold no email address or name for you, so we cannot match a person to an installation: we can find and delete your data only if you can tell us your anonymous token, which the app does not display. If you cannot, we will tell you so rather than delete the wrong person’s data. We aim to reply within 30 days.

What remains after a deletion:

Security

Your rights

Under Article 3 of Taiwan’s Personal Data Protection Act, you may request to inquire into or review your personal data, to obtain copies of it, to supplement or correct it, to have its collection, processing, or use stopped, and to have it deleted. Email karta09104@gmail.com; for requests by email, the identification limit described under “Deleting your data” applies; deleting in the app needs no identification.

Children

The Service is not designed for anyone under 13, and we do not knowingly collect their data. A parent who finds that their child is using the Service may email us to request deletion.

This web page

This is a static web page: no cookies, no analytics, and no third-party resources are loaded.

Changes to this policy

When we change this policy, we will update the effective date above. A change that affects how we handle your data is also announced in the release notes of the store listing.

繁體中文

生效日：2026 年 11 月 7 日

本文件以繁體中文版為原文，英文版為翻譯；兩者如有歧異，以繁體中文版為準。English version

這份文件描述 Podling 這個 app（以下稱「本服務」）蒐集什麼、送到哪裡、存多久、怎麼刪。Podling 目前是 beta。如果哪一句和實際行為不符，那是缺陷，請告訴我們。

重點

誰在蒐集

本服務由 Coyotech 開發與營運，是下列資料的蒐集者與保有者。聯絡方式：karta09104@gmail.com。

我們蒐集什麼，為什麼

匿名裝置身分

第一次啟動時，app 向伺服器要一組 token，存在你的裝置上（app 的本機偏好設定）。之後每個請求都帶著它，伺服器才知道哪些資料是你的。伺服器只保存 token 的 SHA-256 雜湊、建立時間與最後使用時間（一天最多更新一次）。如果伺服器不認得這組 token，app 會換一組新的，並在裝置上保留最近幾組舊的，避免一次誤判就弄丟你的資料。解除安裝 app 會刪掉裝置上的 token，但不會刪掉伺服器上的資料。

在伺服器上，token 指向一個使用者，你的資料屬於這個使用者。匿名使用者只有一支手機；登入的帳號可以有好幾支手機的 token 指向它，你的資料就是這樣跨手機同步。

帳號（選用）

你以 Apple 或 Google 登入時，我們保存：用哪個服務登入、那個服務給你的使用者識別碼，以及它的登入憑證裡的電子郵件（Apple 可能給轉寄用的電子郵件；每次登入都換成最新的那個）。我們不保存你的姓名，也不保存那個服務的任何密碼、權杖或其他憑證：登入憑證只用那個服務的公開金鑰驗證一次，之後就丟掉。

在還沒登入過的手機上登入時，這支手機的資料直接成為帳號的資料。如果帳號已經存在（你先在另一支手機登入過），這支手機的資料會併進帳號：同一個字或片語合併成一筆、例句累加，狀態與意思取最後修改的那一份；追蹤、稍後收聽、播放進度、查詢紀錄與每日活動移進帳號。之後這支手機原本的匿名使用者會被刪除。合併無法復原。

登出會刪除伺服器上這支手機的裝置紀錄；下次連線時，這支手機會拿到一個新的、空白的匿名身分。帳號、帳號裡的資料與其他手機都不受影響。

訂閱與購買

Plus 與 Pro 是經由 App Store 或 Google Play 購買的訂閱，購買需要登入。付款由 Apple 或 Google 處理，我們不會收到你的卡號或付款資料。

RevenueCat, Inc.（美國）是替我們保存訂閱狀態的服務提供者。app 給 RevenueCat 的 SDK 一組我們產生的隨機使用者識別碼（不是你的 token）；RevenueCat 從 app 與商店收到：你的購買紀錄（哪個方案、購買時間、續訂或結束的時間、商店的交易識別碼），以及它的 SDK 每個請求都會送的技術資訊：平台、作業系統與 app 版本、語言設定、商店的國家，在 iPhone 上還有 Apple 的 identifier for vendor（同一個開發者的 app 在那支手機上共用的識別碼）。和任何伺服器一樣，它也看得到你的 IP 位址。我們的伺服器只為登入的帳號保存從 RevenueCat 讀回來的：方案、商店、結束時間，以及是否為試用、是否續訂、是否有付款問題。

app 只在你打開方案頁，或你已登入時在啟動時，才和 RevenueCat 連線。從沒打開方案頁、也沒登入的話，不會有任何東西送到 RevenueCat。如果你沒登入就打開過方案頁，RevenueCat 會留著那組沒有購買紀錄的隨機識別碼；你刪除資料、或那個匿名使用者併進帳號之後，那組識別碼在我們這邊對不到任何人。

你的資料庫

在你的 token 底下，我們保存：你追蹤的節目（與你最後一次看它的時間）、稍後收聽清單上的單集，以及你播放過的每一集：聽到的位置、是否聽完、時間。

查詢與單字本

你點逐字稿裡的字或片語時，我們記下這次查詢：哪一句逐字稿、句中哪幾個字、釋義的語言、時間。你收藏字、片語或句型時，我們保存它的文字、釋義、狀態（新、學習中、已熟悉）與你收藏時的例句。模型產生的釋義也會依字詞與語言快取在伺服器上，所有使用者共用，與你無關聯。

每日活動

app 會回報你聽了或跟讀（聽一段、停頓、自己開口跟著說）多少秒。每一筆我們保存類型、秒數、你手機的當地日期與一組隨機的事件識別碼，用來顯示每日目標與連續天數。app 不錄你的聲音；跟讀只是重播原音並停頓，讓你自己出聲。

你加入的節目

你貼上 podcast 的 RSS 網址時，伺服器會讀取那個 feed，把節目加進所有使用者共用的目錄；網址存在目錄裡，不存在你的 token 底下（只有你追蹤它，節目才和你有關聯）。請不要加入網址內含個人金鑰的私人 feed，它會被存進共用目錄。

搜尋

你輸入的搜尋字，會經由我們的伺服器送到 Apple 的搜尋服務（見下）。伺服器不保存它；結果在記憶體裡最多放 6 小時，讓相同的搜尋不必再問 Apple，並計算你的 token 搜尋的頻率以限速。

逐字稿、翻譯與預習表

你要一集的逐字稿時，伺服器會從發行者那裡下載該集音檔、轉成文字，並保存逐字稿、由它產生的預習表，以及你要求時的中文翻譯。這些屬於那一集，由所有播放它的人共用，不是你的資料，也不含任何與你有關的內容。伺服器上暫存的音檔在該集處理完就刪除，最晚也依 7 天的規則刪除；Google Cloud Storage 會再讓被刪的物件可還原 7 天，所以一份音檔從產生起最久可能存在 14 天。

使用分析

為了了解 Podling 怎麼被使用、哪裡可以改進，app 會把使用事件送到 Google Analytics for Firebase，和 Podling 網頁版、trypodling.com 是同一個 Google Analytics 資源。事件包括：你打開哪個畫面；搜尋（搜尋字轉成小寫，並去掉電子郵件地址與長串數字）；加入節目；開始播放一集，以及聽到它的 25%、50%、75%、100%；要求逐字稿；查詢與收藏字詞；開始與完成跟讀；開關雙語字幕；打開預習表；打開登入、開始登入、登入成功或失敗（Apple、Google 或電子郵件）；登出；刪除資料；打開方案頁；開始、完成、失敗或恢復購買，以及打開訂閱管理。事件帶著單集與節目的編號、相關的方案與價格，以及失敗時的簡短錯誤代碼，絕不包含你的電子郵件、姓名、你查的字或你輸入的其他文字。

一起送出的有：app instance ID，Firebase SDK 為這個安裝產生的隨機識別碼；你登入時，你帳號的隨機使用者識別碼（和給 RevenueCat 的是同一組，不是你的 token 或電子郵件），讓 app 與網頁版把同一個人算成一個人；是否登入，以及你的方案。SDK 也會送出裝置型號、作業系統與 app 版本、語言，Google 並從你的 IP 位址推得大略位置（國家、地區、城市）。app 不把廣告識別碼（IDFA、Android 廣告 ID）、Apple 的 identifier for vendor 或 Android ID 用於分析；廣告儲存、廣告個人化與 Google 信號都關閉。

會不會送，和網頁版一樣看你所在的地區。在你還沒選擇之前，app 會問我們的網頁伺服器（app.trypodling.com），它依你 IP 位址所在的國家回答是否要先同意；這個回答不會被保存。在歐洲經濟區、英國與瑞士，或問不到答案時，要等你在 app 只出現一次的提示裡選「允許」才會送；其他地區預設開啟，這份政策就是告知。不論哪裡，你都可以隨時關閉或開啟：學習分頁、設定圖示、隱私、「分享使用分析」。關閉後立即停止。你的選擇存在你的手機上。

Google 依我們在 Google Analytics 設定的資料保留期限保存這些事件。在 app 裡刪除資料不會刪掉已經送出的事件；如果你希望刪除，請寄信給我們。

只在你的裝置上

app 會寫一個本機的計時量測檔（例如查字卡多快出現）。它不會上傳；只有你打開 app 裡的量測頁並自己分享，它才會離開手機。正式版 app 隱藏這個頁面。節目封面由你的裝置從提供它的主機下載並快取。

系統紀錄

請求失敗時，app 伺服器的系統紀錄會記下錯誤訊息。網頁伺服器沒有設定寫存取紀錄；它自己的系統紀錄在連線出錯時可能記下你的 IP 位址。這些紀錄用於排除故障，依伺服器的系統紀錄輪替期間保存。我們不刻意記錄你資料庫裡的內容。

我們沒有蒐集的

資料會送到哪裡

Apple 與 Google（登入）

如果你選擇登入，登入畫面是 Apple 或 Google 的，你在那裡輸入的資料依它們自己的隱私權政策處理。我們的伺服器會向它們取得公開金鑰來驗證登入憑證。在 Android 上以 Apple 登入時，流程在 Apple 的網頁裡進行，經過我們的伺服器回到 app，伺服器只轉交結果、不保存。刪除以 Apple 登入的帳號時，我們的伺服器會把你確認時再次登入取得的授權碼送給 Apple，撤銷 Podling 的授權。

RevenueCat（訂閱）

見上面的「訂閱與購買」。我們的伺服器在購買或恢復購買之後、RevenueCat 通知有變更時，以及該續訂的時候，向 RevenueCat 讀取登入帳號的方案；RevenueCat 以我們的服務提供者身分，依它的條款處理資料。

Google Analytics for Firebase（使用分析）

見上面的「使用分析」。Google 以我們的服務提供者身分，依 Google Analytics 條款處理這些事件；事件由你的手機直接送給它，所以它看得到你的 IP 位址。

Apple（podcast 搜尋）

搜尋字由我們的伺服器連同國家代碼（台灣）送到 Apple 的 iTunes Search API。Apple 看到的是我們伺服器的 IP 位址，不是你的，也看不到你的 token。

Google Gemini API（釋義、預習表、翻譯）

查字時，伺服器送給 Gemini API 的是你點的字、它所在的逐字稿句子與前後各一句，以及釋義的語言；產生預習表或翻譯時，送的是該集的逐字稿文字。不送你的 token。文字依 Google 的 Gemini API 條款處理。

OpenAI（語音轉文字）

為了產生逐字稿，伺服器把發行者的單集音檔轉檔後，每段最長 5 分鐘，送到 OpenAI 的 whisper-1 轉錄 API。那是發行者的音檔，不是你錄的任何東西，也不送你的 token。音檔依 OpenAI 的 API 條款處理。

Podcast 發行者與其主機

我們的伺服器會向發行者的主機抓節目的 RSS feed、單集音檔，以及 feed 有連結時的官方逐字稿；它們看到的是我們伺服器的 IP 位址。你的手機則直接向發行者的主機串流你播放的音檔，並從提供封面的主機載入封面圖片，所以那些主機會看到你的 IP 位址與你播放的內容，和任何 podcast app 一樣。我們無法控制它們，適用它們自己的政策。

你裝置的語音引擎

單字的發音由你手機內建的文字轉語音引擎念出，我們不為此把字送給任何人。那是哪一個引擎、能不能離線運作，取決於你手機的設定。

Google Cloud（主機）

我們的伺服器、資料庫與檔案儲存都在 Google Cloud 上。app 與伺服器之間的通訊一律使用 HTTPS。

資料存放在哪裡

我們的伺服器與資料庫位於 Google Cloud 的 us-west1 區域（美國奧勒岡州），檔案儲存也在那裡。RevenueCat 在美國處理資料。Apple、Google（含 Google Analytics）與 OpenAI 可能在美國或其他國家處理資料。使用本服務，即表示你知悉你的資料會被傳送到你所在的國家或地區以外。

保存與刪除

綁在你 token 上的資料沒有自動刪除期限：閒置不會刪，解除安裝 app 也不會刪，要等到你在 app 裡刪除；帳號與帳號裡的資料也一樣保存到帳號被刪除為止。購買紀錄另外由 Apple 或 Google 依它們的政策保存。使用分析由 Google 依我們在 Google Analytics 設定的資料保留期限保存。其他有期限的只有上面寫的：音檔暫存（最久 14 天，含 Google Cloud Storage 讓被刪物件可還原的 7 天）、搜尋快取（最多 6 小時），以及備份（見下）。

刪除你的資料

在 app 裡開啟學習分頁，點設定圖示，選「刪除我的資料」並確認。伺服器會一次刪除：裝置紀錄、追蹤、稍後收聽清單、播放進度、查詢紀錄、單字本與例句，以及每日活動。刪除立即生效，無法復原。刪除前，app 會先嘗試送出尚未送出的收聽時間，讓它跟其他資料一起被刪除。伺服器確認後，app 會忘記你的 token（以及它留著的幾組舊 token），並丟棄送不出去的收聽時間，下次連線時就以全新、空白的安裝重新註冊。如果連不上伺服器，就不會刪除任何東西，app 會告訴你；請稍後再試。

登入時，同一個位置是「刪除帳號」。它會刪除帳號與上述所有資料，所有登入這個帳號的手機都會登出。以 Apple 登入的，會請你再以 Apple 登入一次來確認，並撤銷 Podling 在 Apple 的授權；如果確認不了，仍然可以刪除帳號，這時請自己到 Apple ID 設定（使用 Apple 登入）移除 Podling。帳號在 RevenueCat 的紀錄也會一併刪除。刪除帳號不會取消訂閱：請到 App Store 或 Google Play 的訂閱設定取消。沒有取消的話，商店下一次續訂時，RevenueCat 會以那組識別碼重新建立紀錄，而它在我們這邊已經對不到任何人。

如果你已解除安裝 app，或無法使用它，請寄信到 karta09104@gmail.com。如果你登入過，請用帳號的電子郵件寄信，我們可以用它找到帳號。否則我們沒有你的電子郵件或姓名，無法把某個人對應到某個安裝：只有在你能告訴我們你的匿名 token 時，我們才找得到並刪除你的資料，而 app 不會顯示它。如果你拿不出來，我們會直接告訴你，不會冒險刪到別人的資料。我們的目標是 30 天內回覆。

刪除之後仍然留著的：

安全

你的權利

依台灣《個人資料保護法》第 3 條，你可以請求查詢或閱覽你的個人資料、請求製給複製本、請求補充或更正、請求停止蒐集、處理或利用，以及請求刪除。請寄信到 karta09104@gmail.com；以電子郵件提出的請求適用「刪除你的資料」一節說明的身分確認限制；在 app 內刪除不需要身分確認。

兒童

本服務不是為 13 歲以下的人設計，我們也不會刻意蒐集他們的資料。家長發現孩子在使用本服務，可以寄信要求刪除。

這個網頁

這是靜態網頁：沒有 cookie、沒有分析、沒有載入任何第三方資源。

政策變更

修改時會更新上方的生效日。變更若影響我們如何處理你的資料，也會寫在商店頁面的更新說明裡。