Thanks for your work uncovering this. Imposing this broad and undisclosed censorship upon the user is unacceptable, so we have stopped hosting Tesseract on our instance.
As I understand it, LS would fall under the 'microenterprise' classification in the DSA and therefore be exempt from the age assurance measures contained therein, so I think we can remain in the EU for the time being.
Implementing this shouldn't have any downsides, even if it doesn't prove to be very effective, so I have added the header to all requests. Thanks for the suggestion.
Yes to all three. Regression of the political climate in the US makes me uneasy about operating services from there (even if it has been via an EU-based company up to this point). EU law also provides strong privacy protections against US overreach via apparatus such as the CLOUD Act and FISA 702, which can potentially allow US federal agencies to compel cloud service providers there to hand over any and all data they request - without any disclosure to us or you. We have been able to secure a more favourable deal on a root server in Germany, so there is some economic motivation.
We're operating out of Australia. Hosting our data in Europe will provide additional privacy protections to users given that US-based companies can be compelled to provide hosted data to government agencies via a number of mass surveillance measures (CLOUD Act, FISA 702, etc.) without any disclosure.
That said, Lemmy is inherently an open platform and users should operate on the assumption that their activity is public.
Won't switching from an unsigned 64-bit integer to a signed one of the same size effectively halve how far into the future they can handle dates, exhausting it in 2038?
Yeah, same story - the few I looked at looked like they were from consumer ISPs. We had ~7m requests from Brazil, ~900k from Argentina, and a little less from Ecuador, Colombia and Russia.
I already use Navidrome, but I discovered Explo through your post, so thanks! It seems to work well in that it brings in the tracks that it should, but I don't think I can keep using it because it pollutes my 'Recently Added' list in Navidrome with 50 new albums, each with a single track. If I could somehow prevent that, I think I'd keep using it. I tried using an .ndignore file but that didn't work - it stops them showing up in Recents, but also prevents the tracks from working in the playlist that Explo generates.
The connectivity issue with LW is resolved and we are catching back up on their sent activities - currently about 400k activities behind, but we should be up to date in maybe 6 hours or so. You can check under Lagging Instances here to see our current status.
We would consider it if it has the blocklist function removed. Whether the fork gets ongoing maintenance is a concern though.
Thanks for your work uncovering this. Imposing this broad and undisclosed censorship upon the user is unacceptable, so we have stopped hosting Tesseract on our instance.
Okay, that was intense.
So sad. Loved him in The Hunt for Red October. Going to watch Possession now as I've never seen it, but it sounds good.
As I understand it, LS would fall under the 'microenterprise' classification in the DSA and therefore be exempt from the age assurance measures contained therein, so I think we can remain in the EU for the time being.
Not that I know of. I can make a larger version of the logo available for anyone who wanted to produce these themselves.
Implementing this shouldn't have any downsides, even if it doesn't prove to be very effective, so I have added the header to all requests. Thanks for the suggestion.
Yes to all three. Regression of the political climate in the US makes me uneasy about operating services from there (even if it has been via an EU-based company up to this point). EU law also provides strong privacy protections against US overreach via apparatus such as the CLOUD Act and FISA 702, which can potentially allow US federal agencies to compel cloud service providers there to hand over any and all data they request - without any disclosure to us or you. We have been able to secure a more favourable deal on a root server in Germany, so there is some economic motivation.
We're operating out of Australia. Hosting our data in Europe will provide additional privacy protections to users given that US-based companies can be compelled to provide hosted data to government agencies via a number of mass surveillance measures (CLOUD Act, FISA 702, etc.) without any disclosure.
That said, Lemmy is inherently an open platform and users should operate on the assumption that their activity is public.
Won't somebody think of the Eloi?
I only learned this recently too watching the Rocky Horror cast reunion.
You're right, I was thinking about 32-bit timestamps. Definitely not an issue for 64-bits.
Won't switching from an unsigned 64-bit integer to a signed one of the same size effectively halve how far into the future they can handle dates, exhausting it in 2038?
The hyprland issues I'm aware of, but what's the issue with suckless?
Yeah, same story - the few I looked at looked like they were from consumer ISPs. We had ~7m requests from Brazil, ~900k from Argentina, and a little less from Ecuador, Colombia and Russia.
We were getting DDoS'd by a bunch of South American IPs (mostly Brazilian) from maybe 18 hours ago.
We had this issue on and off for a few weeks at least, causing massive postgres CPU spikes. I ended up blocking large page params with an nginx regex.
I already use Navidrome, but I discovered Explo through your post, so thanks! It seems to work well in that it brings in the tracks that it should, but I don't think I can keep using it because it pollutes my 'Recently Added' list in Navidrome with 50 new albums, each with a single track. If I could somehow prevent that, I think I'd keep using it. I tried using an
.ndignorefile but that didn't work - it stops them showing up in Recents, but also prevents the tracks from working in the playlist that Explo generates.The connectivity issue with LW is resolved and we are catching back up on their sent activities - currently about 400k activities behind, but we should be up to date in maybe 6 hours or so. You can check under Lagging Instances here to see our current status.