Updating the filters works; just got the comment archive back online. It's just changes to how filters are generated, so all API filters previously in use are now invalid. Not sure if this was intentional, but I somewhat doubt they'll bother rolling it back
I can't talk about the full details yet, but I can say they've rolled out bot detection (this can be verified by anyone by clicking F12 and looking for CloudFlare-related network requests), and it's so aggressive that it's quick to kill boson. The block method then forces it to stay down for 30-60 (or more? not sure) minutes per block. None of the alleged exemptions (that I can't talk about) work, so the workaround of running boson under my account isn't reliable either
@gparyani Sorry for the delay - an internal announcement a day later distracted me, because it determined whether there was a point in doing anything or not. The answer, for now, is that the on-site comment archive has been axed and moved to Discord. Not sure how it deals with alt text yet, and I haven't checked.
@Spevacus They're dropping the requirement some time next week, so trying to get anyone whitelisted now is likely going to take more time than they'll spend removing the requirement
Ugh, so yes. That's the single dumbest thing I've ever heard. "We have a spam problem. We're redirecting all questions via the SG. I know, let's restrict flagging access in the SG! That totally won't affect our spam problem negatively at all"
Anyway, wouldn't worry too much about it. As far as changes go, it isn't too negatively impactful just by it happening (unlike the CF migration. RIP Boson). The data still lives on
In terms of the google-ness of the site, it's not really any different. Google can poke at more user data if they really want to, but Google is already built into the site in several places. Google login, Google ads, and Google analytics are already in use. Just moving from their own hardware to Google-owned hardware functionally doesn't do anything worse than switching to Cloudflare (which is bad because a more centralised internet is fundamentally bad and not what the internet was designed
@SPArcheon-onstrike You rang? :p On a more serious note, this isn't really that big of a problem. They're moving to GC in particular because they have a partnership with Google. They probably negotiated good GC terms when they partnered (though we'll never find out the real details there). As a side-note, the data dump itself is hosted on Cloudflare infrastructure (cloudflare storage), so the only difference now is that google gets to see all traffic hitting SO
