Apple's Guideline 1.2 requires an app carrying user-generated content to offer a way to report it, a way to block a person, and published contact information users can reach. Reporting and blocking both ship. The published contact did not exist.
/safety is that page, and it doubles as the App Store Support URL, so one page answers both requirements.
The address was tested before it was published
abuse@flamenetmessenger.com is a virtual alias to a real mailbox. A message sent through the production mailserver arrived:
A plausible-looking address nobody reads is worse than none — it consumes the one attempt somebody makes.
It says what we cannot do
Messages are sealed on the sending phone, so we cannot read a conversation to judge a report. The page says that in those words: we are not withholding it, we do not have it.
A safety page that implied content moderation on an end-to-end encrypted service would be a claim this project must not make — and it would mislead the person reporting into thinking evidence they cannot supply is on file. So the page asks for what actually helps: the handle, roughly when, what they'd like done, and anything they can show us.
It also says plainly that this is a small company rather than a moderation department, that emergency services come first if somebody is in danger, and that CSAM is the one thing acted on immediately and reported onward.
The test
Pins the address by name, the Report and Block instructions, the "cannot read" sentence, and the footer link. A tidy-up that drops any of them is a rejected submission, so it should be a red test rather than a review cycle.
Closes the published-contact half of ios#12.
Apple's **Guideline 1.2** requires an app carrying user-generated content to offer a way to report it, a way to block a person, and **published contact information** users can reach. Reporting and blocking both ship. The published contact did not exist.
`/safety` is that page, and it doubles as the **App Store Support URL**, so one page answers both requirements.
## The address was tested before it was published
`abuse@flamenetmessenger.com` is a virtual alias to a real mailbox. A message sent through the production mailserver arrived:
```
to=<patriot@hitthosting.com>, orig_to=<abuse@flamenetmessenger.com>,
dsn=2.0.0, status=sent (250 2.0.0 … Saved)
```
A plausible-looking address nobody reads is worse than none — it consumes the one attempt somebody makes.
## It says what we cannot do
Messages are sealed on the sending phone, so **we cannot read a conversation to judge a report**. The page says that in those words: *we are not withholding it, we do not have it.*
A safety page that implied content moderation on an end-to-end encrypted service would be a claim this project must not make — and it would mislead the person reporting into thinking evidence they cannot supply is on file. So the page asks for what actually helps: the handle, roughly when, what they'd like done, and anything they can show us.
It also says plainly that this is a small company rather than a moderation department, that emergency services come first if somebody is in danger, and that CSAM is the one thing acted on immediately and reported onward.
## The test
Pins the address **by name**, the Report and Block instructions, the "cannot read" sentence, and the footer link. A tidy-up that drops any of them is a rejected submission, so it should be a red test rather than a review cycle.
Closes the published-contact half of ios#12.
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Apple's Guideline 1.2 requires an app carrying user-generated content to offer a way to report it, a way to block a person, and published contact information users can reach. Reporting and blocking both ship. The published contact did not exist.
/safetyis that page, and it doubles as the App Store Support URL, so one page answers both requirements.The address was tested before it was published
abuse@flamenetmessenger.comis a virtual alias to a real mailbox. A message sent through the production mailserver arrived:A plausible-looking address nobody reads is worse than none — it consumes the one attempt somebody makes.
It says what we cannot do
Messages are sealed on the sending phone, so we cannot read a conversation to judge a report. The page says that in those words: we are not withholding it, we do not have it.
A safety page that implied content moderation on an end-to-end encrypted service would be a claim this project must not make — and it would mislead the person reporting into thinking evidence they cannot supply is on file. So the page asks for what actually helps: the handle, roughly when, what they'd like done, and anything they can show us.
It also says plainly that this is a small company rather than a moderation department, that emergency services come first if somebody is in danger, and that CSAM is the one thing acted on immediately and reported onward.
The test
Pins the address by name, the Report and Block instructions, the "cannot read" sentence, and the footer link. A tidy-up that drops any of them is a rejected submission, so it should be a red test rather than a review cycle.
Closes the published-contact half of ios#12.
restored source branch
feat/safety-contact-pagementioned in commit
1066da4245changed the description