Pages like https://respublicae.eu/@CopernicusEU/media?min_id=108069272592688414 will appear rather broken, but they're rarely accessed. (And unfortunately they were quite broken already, for example with links to a now squatted Nitter domain...)
To reduce costs (and possibly save energy), we're experimenting with moving older files to cold storage.
For now the simplistic approach is a #Scaleway object storage lifecycle rule to move all attachments older than 180 days to the Glacier class. It's not possible for such rules to target only originals, or mp4 files, or files above a certain size.
In a few months we can consider whether to refine this approach. Feedback appreciated.
"#Nitter is dead", proclaimed the maintainer after yet more #TwitterAPI changes.
https://github.com/zedeus/nitter/issues/1155#issuecomment-1913361757
Some instances managed to survive a bit longer but now we're suspending updates of #ResPublicae accounts. We'll reassess in a little while.
Since 2023-08-10, our mirrors almost always failed.
We're now trying the #Nitter instance by the excellent @nolog as source. (Trying to avoid excess load.)
Please support them and the Nitter developers if you can:
https://nolog.cz/en/support/
https://liberapay.com/zedeus
Fermeture de notre service Nitter
https://nitter.lacontrevoie.fr/
Maintenir une instance Nitter est devenu de plus en plus complexe et chronophage ces derniers mois.
Nous pensons que notre temps et notre énergie seront mieux investi·es ailleurs, ce pourquoi nous avons décidé de fermer notre instance Nitter.
Si vous souhaitez toujours utiliser Nitter, retrouvez la liste des instances ici :
https://github.com/zedeus/nitter/wiki/Instances
(1/3)
I was introducing someone I know to #Mastodon a few years back, starting with:
"So let's find a server that suits your niche interests. What are you looking for?"
...thinking that I knew all the fedi communities pretty well.
They responded cheekily "Is there a european-foreign-policy server?"
Of course I came up short, and they ended up making an account on a generalist server.
But fedi has grown a lot since!
So happy that eupolicy.social, respublicae.eu and social.network.europa.eu exist 🔥
#ResPublicae accounts now have over 70,000 followers (not unique).
Under the hood, Mastodon keeps track of many more users who for some reason interacted with our instance. Over time that becomes rather wasteful, so we've finally activated regular data purging. Such #MastoAdmin hygiene saves us 200 GB in storage.
Even better, we no longer have to worry about holding potentially personal data about thousands of users for no reason. Our #privacy policy now looks better.
https://respublicae.eu/terms
Our source is broken again, we'll reconsider the alternatives. (Such as self-hosting or switching to a Nitter source.)
And we're now on v3.5.10.
More information on the importance of the v3.5.9 release:
https://www.bleepingcomputer.com/news/security/critical-tootroot-bug-lets-attackers-hijack-mastodon-servers/
#ResPublicae has been upgraded to Mastodon v3.5.9, a security release.
https://github.com/mastodon/mastodon/releases?q=v3.5&expanded=true
The upgrade was painless. Thank you #MastoDev and especially @Claire[@]sitedethib.com for offering backports to pre-v4 releases and for the clear instructions.
While the details of CVE-2023-36459 have not been published yet, the title suggests the vulnerability may be exploitable even on an instance like ours where no third party can directly post anything.
https://github.com/mastodon/mastodon/releases?q=v3.5&expanded=true
To reduce truncation errors, we caved and increased the character limit for posts to 2000 characters.
#ResPublicae mirrors are catching up on the recent posts, using RSS input for the ever-wonderful #PleromaBot.
We're only able to get the 20 most recent posts from #RSSHub, so any account which posted more than that in the past ~2 weeks will have some posts missing. Some posts also get a bit messed up in the data conversion, so please be patient. (Most issues are already reported upstream.)
Please send patches or donations at https://pleromabot.robertoszek.xyz/pleroma-bot/!
So it finally happened. The #TwitterAPI access used to update #ResPublicae was cut off, all requests get an HTTP 401 response. Some discussion at
https://github.com/robertoszek/pleroma-bot/issues/120#issuecomment-1593688412
Venez nous retrouver aux @geekfaeries ! Notre stand se trouve au Village du Libre avec @Framasoft, @Mozilla et l' @aprilorg !
Et merci beaucoup @Pouhiou pour cette grue géante en crochet !!
Due to the growth of the #fediverse, which may have passed 10 million registered users depending how you count, we had to update our #DigitalServicesAct (#DSA) disclosures.
https://respublicae.eu/about/more
Other possibilities exist. The principle of least astonishment could be a guiding factor:
https://en.wikipedia.org/wiki/Principle_of_least_astonishment
We could mark posts from certain accounts unlisted if they regularly post to hashtags in ways which don't belong.
We could also forcefully add CWs to posts, but so far we don't. It's hard to come up with a CW which would be applicable for all accounts, or for entire classes of accounts equally. A catch-all CW like "From birdsite 🐦" is broadly correct but also potentially unhelpful.
Admin and host of respublicae.eu. Controlled by @nemobis.
The profile and banner photos are CC-0 by W.carter and GrandCelinien.