My point obviously isn’t to attack you, but any Linux server would have the full history of connections, so at least in theory, you could still compare your normal days to the two Canvas days and get a list of IP addresses that could be attributed to those of Canvas users. Even if there is a lot of junk, it’s still closer to being personally identifiable than just four views counts with nothing else we have access to via mystats.wtf. Moreover, a domain name registered to a single person is more likely to expire or be used for something malicious than one from an established organization. Grant should accept that there is a risk and not try to micromanage just some links.
An error occurred while parsing this post.
If this isn’t an Atlas submission, you can ignore this comment.
Exception details: Expecting value: line 80 column 5 (char 997)
@grant@toast.ooo As with 3minutes, your editorializing seems biased. Why is there a Wikipedia link before the actual link? Pawper from Cassette Beasts didn’t get the same treatment. Serious Sam, Subnautica 2, KDE, Factorio and Inugami Korone didn’t get the same treatment. And I’m sure the list isn’t exhaustive.
@grant@toast.ooo It’s not quite fair to attempt to scare people off by prefacing to say it’s a tracking link. Though views and voluntary post-watch replies are tracked, there’s zero personally identifiable information collected, which makes it drastically more ethical than the conventional tech used for analytics, like Google Analytics and Cloudflare, present on countless websites like the linked Serioussam.com, pawb.social, Reddit or YouTube and others, without any warning on the Atlas. There are even low-traffic and personal websites (e.g. Redjard.org) linked without a warning, allowing their creators to fully monitor the requests within the set time frame and gather everything up to the IP addresses.
RetroFed
Thanks for your submission! I’ve created an MR for you: https://sc07.dev/fediverse.events/canvas-atlas/-/merge_requests/529
Thanks for your submission! I’ve created an MR for you: https://sc07.dev/fediverse.events/canvas-atlas/-/merge_requests/528
Thanks for your submission! I’ve created an MR for you: https://sc07.dev/fediverse.events/canvas-atlas/-/merge_requests/527
Thanks for your submission! I’ve created an MR for you: https://sc07.dev/fediverse.events/canvas-atlas/-/merge_requests/526
Thanks for your submission! I’ve created an MR for you: https://sc07.dev/fediverse.events/canvas-atlas/-/merge_requests/525
Thanks for your submission! I’ve created an MR for you: https://sc07.dev/fediverse.events/canvas-atlas/-/merge_requests/524
Thanks for your submission! I’ve created an MR for you: https://sc07.dev/fediverse.events/canvas-atlas/-/merge_requests/523
Thanks for your submission! I’ve created an MR for you: https://sc07.dev/fediverse.events/canvas-atlas/-/merge_requests/522
Thanks for your submission! I’ve created an MR for you: https://sc07.dev/fediverse.events/canvas-atlas/-/merge_requests/521
Thanks for your submission! I’ve created an MR for you: https://sc07.dev/fediverse.events/canvas-atlas/-/merge_requests/520
Thanks for your submission! I’ve created an MR for you: https://sc07.dev/fediverse.events/canvas-atlas/-/merge_requests/519
My point obviously isn’t to attack you, but any Linux server would have the full history of connections, so at least in theory, you could still compare your normal days to the two Canvas days and get a list of IP addresses that could be attributed to those of Canvas users. Even if there is a lot of junk, it’s still closer to being personally identifiable than just four views counts with nothing else we have access to via mystats.wtf. Moreover, a domain name registered to a single person is more likely to expire or be used for something malicious than one from an established organization. Grant should accept that there is a risk and not try to micromanage just some links.
To be fair I can’t see if someone came from canvas or elsewhere. It’s not like I added a /canvas that redirects to / and makes it unique.
I get ~5-30k external page loads a day (realistically probably 95% ai crawlers). There’s no way I can filter out a few people from canvas in that.
Thank you ♥️
Thanks for your submission! I’ve created an MR for you: https://sc07.dev/fediverse.events/canvas-atlas/-/merge_requests/518
@grant@toast.ooo did I do something wrong?
An error occurred while parsing this post. If this isn’t an Atlas submission, you can ignore this comment. Exception details: Expecting value: line 80 column 5 (char 997)
@grant@toast.ooo As with 3minutes, your editorializing seems biased. Why is there a Wikipedia link before the actual link? Pawper from Cassette Beasts didn’t get the same treatment. Serious Sam, Subnautica 2, KDE, Factorio and Inugami Korone didn’t get the same treatment. And I’m sure the list isn’t exhaustive.
@grant@toast.ooo It’s not quite fair to attempt to scare people off by prefacing to say it’s a tracking link. Though views and voluntary post-watch replies are tracked, there’s zero personally identifiable information collected, which makes it drastically more ethical than the conventional tech used for analytics, like Google Analytics and Cloudflare, present on countless websites like the linked Serioussam.com, pawb.social, Reddit or YouTube and others, without any warning on the Atlas. There are even low-traffic and personal websites (e.g. Redjard.org) linked without a warning, allowing their creators to fully monitor the requests within the set time frame and gather everything up to the IP addresses.
Thanks for your submission! I’ve created an MR for you: https://sc07.dev/fediverse.events/canvas-atlas/-/merge_requests/517