Add Telemetry page (#1020)

* Add new Telemetry page in feature references

* Incorporate feedback

Add section 'Why collect telemetry data', change amplitude metrics to bulleted list, add instructions for disabling telemetry on a cloud service

* Add telemetry FAQ answer

Also incorporate feedback on telemetry page, and move it to /learn/what_is_meilisearch

Co-authored-by: Tommy <[email protected]>

Author: dichotommy

.vuepress/config.js

@@ -49,6 +49,7 @@ module.exports = {
             '/learn/what_is_meilisearch/philosophy',
             '/learn/what_is_meilisearch/sdks',
             '/learn/what_is_meilisearch/comparison_to_alternatives',
+            '/learn/what_is_meilisearch/telemetry',
             '/learn/what_is_meilisearch/contact',
           ],
         },

learn/what_is_meilisearch/telemetry.md

@@ -0,0 +1,84 @@
+---
+sidebarDepth: 2
+---
+
+# Telemetry
+
+MeiliSearch collects anonymized data from users in order to improve our product. This can be [deactivated at any time](#how-to-disable-data-collection).
+
+Note that while MeiliSearch is running in [`development` mode](/reference/features/configuration.md#environment), no data is collected.
+
+## What kind of data do we collect?
+
+We use two data-collection tools: [Amplitude](https://amplitude.com/amplitude-analytics) and [Sentry](https://sentry.io/welcome/). Both can be disabled by [following the below procedure](#how-to-disable-data-collection).
+
+**Amplitude** collects the following datapoints:
+
+- Amount of time the MeiliSearch instance has been running
+- MeiliSearch version
+- Technical specs of the device running MeiliSearch
+- Number of documents per index
+
+**Sentry** collects information about errors that occur while using MeiliSearch—in other words, bug reports and diagnostics.
+
+**We will absolutely never identify users, track users, or collect private information**, including IP address, email, or website URLs. All of our data collection is **GDPR-compliant** and **secure**.
+
+::: tip
+
+In the future, we may collect additional metrics relating to new features. This page will be kept up-to-date, but for the most precise information be sure to check the release notes after each new version. <Badge text="soon" type="warn"/>
+
+:::
+
+## Why collect telemetry data?
+
+We collect anonymized data for only two reasons: so that we can improve our product, and so that we can continue working on this project full-time.
+
+In order to better address our users' needs and chart a clear path forward for MeiliSearch, we need reliable quantitative information. The information we collect helps us fix bugs, evaluate the success of certain features, and identify our strengths and weaknesses as a product.
+
+We also need to prove that people are actually using MeiliSearch! Usage metrics help us justify our existence to investors so that we can keep this project alive.
+
+## How to disable data collection
+
+Data collection can be disabled at any time by setting a command-line option or environment variable, then restarting the MeiliSearch instance.
+
+:::: tabs
+
+::: tab Command-line option
+
+```bash
+meilisearch --no-analytics=true
+```
+
+:::
+
+::: tab Environment variable
+
+```bash
+export MEILI_NO_ANALYTICS=true
+meilisearch
+```
+
+:::
+
+::: tab Cloud service
+
+```bash
+# The following procedure should work for all cloud providers,
+# including DigitalOcean, Google Cloud Platform, and Amazon Web Services.
+# First, open /etc/systemd/system/meilisearch.service with a text editor:
+
+nano /etc/systemd/system/meilisearch.service
+
+# Then add --no-analytics=true at the end of the command in ExecStart
+# Don't forget to save and quit!
+# Finally, run the following two commands:
+
+systemctl daemon-reload
+systemctl restart meilisearch
+```
+
+:::
+
+::::
+
+For more information about configuring MeiliSearch, read our [configuration reference](/reference/features/configuration.md).

resources/faq.md

@@ -235,7 +235,13 @@ In general, we recommend the former. However, if you need to reduce the size of
 
 - **More relevancy rules => a larger database**
   - The proximity [ranking rule](/learn/core_concepts/relevancy.md#ranking-rules) alone can be responsible for almost 80% of database size
-- [Faceting](/reference/features/faceted_search.md) also consumes a large amount of disk space
+- Adding many attributes to [filterableAttributes](/reference/features/settings.md#attributes-for-faceting) also consumes a large amount of disk space
 - Multi-lingual datasets are costly, so split your dataset—one language per index
 - [Stop words](/reference/features/stop_words.md) are essential to reducing database size
 - Not all attributes need to be [searchable](/reference/features/field_properties.md#searchable-fields). Avoid indexing unique IDs.
+
+## Why does MeiliSearch send data to Amplitude? Does MeiliSearch track its users?
+
+**MeiliSearch will never track or identify individual users**. That being said, we do use Amplitude and Sentry to collect anonymous data about user trends and bug reports.
+
+You can read more about what metrics we collect, why we collect them, and how to disable it on our [telemetry page](/learn/what_is_meilisearch/telemetry.md). Transparency is very important to us, so if you feel we are lacking in this area please [open an issue](https://github.com/meilisearch/documentation/issues/new/choose) and let us know! ❤️