4 ms·

Show a button to check for updates on any version timestamped older than X months where the button hasn't been clicked in a while. Show the button on the crash

Show a button to check for updates on any version timestamped older than X months where the button hasn't been clicked in a while. Show the button on the crash dialog. Don't open a single socket without approval. Let the user decide. This is a fundamental principle of human interaction -- ask first before touching someone's stuff.

BugWatch 5y ago
Having written my rant(s) elsewhere in this thread, I'm just gonna telepathically project the "(slams the palm on the desk) THANK YOU!" gif.
CJefferson 5y ago
The problem nowadays, is with everything connected to everything else, some bugs need fixing fast. Imagine if there is a buffer overflow in Audacity, meaning someone can take over your machine with a malformed sound file. I'd want to know about that ASAP -- not after a month. On average, I believe most users benefit from an automated check for updates. It should be possible to disable it, but we should first worry about keeping users safe.
- nitrogen 5y ago
  On average, I believe most users benefit from an automated check for updates. It should be possible to disable it, but we should first worry about keeping users safe. They should still ask first, at first run. I actually enable update checks for a lot of things, if I'm given the option up front. I'll submit crash reports and even periodic telemetry if I know that the crash reporting system doesn't download and run arbitrary code, shows me everything that will be submitted, and allows me to omit needlessly identifying details (e.g. custom kernel strings). Some projects (KDE) make it really hard to submit crash reports, by requiring an account, etc. It must be optional, it must be easy, and it must be transparent. Then users will volunteer crash reports and stats. Respect breeds respect.

4 ms·

Show a button to check for updates on any version timestamped older than X months where the button hasn't been clicked in a while. Show the button on the crash

by nitrogen 5y ago

BugWatch 5y ago
Having written my rant(s) elsewhere in this thread, I'm just gonna telepathically project the "(slams the palm on the desk) THANK YOU!" gif.
CJefferson 5y ago
The problem nowadays, is with everything connected to everything else, some bugs need fixing fast. Imagine if there is a buffer overflow in Audacity, meaning someone can take over your machine with a malformed sound file. I'd want to know about that ASAP -- not after a month. On average, I believe most users benefit from an automated check for updates. It should be possible to disable it, but we should first worry about keeping users safe.
- nitrogen 5y ago
  On average, I believe most users benefit from an automated check for updates. It should be possible to disable it, but we should first worry about keeping users safe. They should still ask first, at first run. I actually enable update checks for a lot of things, if I'm given the option up front. I'll submit crash reports and even periodic telemetry if I know that the crash reporting system doesn't download and run arbitrary code, shows me everything that will be submitted, and allows me to omit needlessly identifying details (e.g. custom kernel strings). Some projects (KDE) make it really hard to submit crash reports, by requiring an account, etc. It must be optional, it must be easy, and it must be transparent. Then users will volunteer crash reports and stats. Respect breeds respect.