accelerated ignorization
accelerated ignorization
tomorrow at @c-base.org dissent-and-datalove.institute/events/2025-...
When we discuss Open Source AI, the devil is in the (unreleased) data.
memory-alpha.fandom.com/wiki/Spot says the special edition DVD includes a deleted scene which answers that question
Screenshot of DS9 S03E11 showing an electronic sign indicating that today is Friday, the date is "8/30/24", and the temperature is 15C.
29 years ago Deep Space Nine predicted that today it would be 15C in San Francisco (✅), the US would have switched to using Celsius (❌), but would still be writing dates in M/DD/YY format (✅), and Benjamin Sisko will lead a large civil uprising this Sunday (❓) memory-alpha.fandom.com/wiki/Bell_Ri...
screenshot of GPT-4o conversation: 9.11 or 9.9 what's bigger 9.11 is bigger than 9.9. are you sure Yes, I am sure. In numerical terms, 9.11 is greater than 9.9. how much bigger To find out how much bigger 9.11 is than 9.9, you can subtract 9.9 from 9.11: 9.11 - 9.9 = 0.21 So, 9.11 is 0.21 units bigger than 9.9. how did you arrive at the number 0.21 I arrived at the number 0.21 by performing a simple subtraction: Start with the larger number: 9.11 Subtract the smaller number: 9.9 So, the calculation is: 9.11 - 9.9 = 0.21 This shows that 9.11 is 0.21 units greater than 9.9.
screenshot of GPT-4o saying: Let's go through the calculation step by step again: Identify the numbers: We have 9.11 and 9.9. Align the decimal points for clarity: Code 9.11 -9.90 Subtract: Start from the rightmost digit: In the hundredths place: 1 - 0 = 1 In the tenths place: 1 - 9 cannot be done without borrowing, so we borrow 1 from the 9 in the units place, making it 8. Now we have 11 - 9 = 2. In the units place: 8 - 9 cannot be done, so we have 8 - 0 = 8. Putting it all together, we get: Code 9.11 - 9.90 = 0.21 So, 9.11 is indeed 0.21 units greater than 9.9. If you have a specific aspect of the calculation you'd like me to clarify further, please let me know!
en.wikipedia.org/wiki/United_... 👀
amazing how many different ways there are to be wrong
LFG
oh my god
there is a lot going on but I'm having a hard time deciding whats the stupidest thing happening
the "latest from follows" feed does that bsky.app/profile/did:... (though its short name is "bestoffollows", i think its display name "latest from follows" is actually correct?). see also bsky.app/profile/did:... (though it's currently missing posts from many of my mutuals...) /cc @why.bsky.team
when will activitypub decouple identity from data hosting, aggregation, and moderation? :)
pretty cool, though i'm still far from fully understanding the details about how this bridge works
i finally tried it out; i posted these two replies to @evan.cosocial.ca.ap.brid.gy bsky.app/profile/leif... bsky.app/profile/leif... and i see the first here cosocial.ca/@leif.bsky.s... on the AP side but it looks like the second one didn't make it. maybe because it contains a GIF?
screenshot of evan's crunchbase profile showing his rank as 7,352
fyi, 2 days later:
All the images in this thread are geographic features from the US National Map (apps.nationalmap.gov/viewer) whose names are officially recognized by the US Board on Geographic Names: www.usgs.gov/us-board-on-...
hi @snarfed.org i just read your status update ( snarfed.org/2024-04-11_b... ) and am still curious about these questions i asked a couple months ago
Moral consistency requires that if you say thanks to a chatbot you must say thanks to every object you use
if the AP to ATP side were opt-in (which i know is not your current intent...) this would ensure that multiple bridges don't result in duplicate ATP identities for a single AP user. afaict there is no way to prevent duplicates in the opposite direction (short of ATP making changes to accommodate AP)
what about the other direction? will you make it possible for AP instance operators who aren't ready (yet) to run their own bridge just do handle verification for now so user@example.com can be @user.example.com? and will you let AP users eventually take full control of the DIDs you create for them?
Text that says the city has canceled its contracted with ShotSpotter. The text reads “the city of Chicago will not renew its contract with sounthinking that expires February 16, 2024 and will decommission the use of ShotSpotter technology on September 22,2024”
In a rebuke of surveillance technology, the city of Chicago is decommissioning its use #ShotSpotter and canceling its contract.
The thing about Taylor Swift and Travis Kelce is that last week the NSA confirmed to me that the agency is buying up Americans' internet browsing data from shady data brokers without a warrant.
Alas, we
Who wished to lay the foundations of kindness
Could not ourselves be kind.
But you, when at last it comes to pass
That man can help his fellow man,
Do no judge us
Too harshly.
— To Posterity, Bertold Brecht
screenshot of: APPENDIX D PRACTICAL CONSIDERATIONS A. Interacting with legacy providers a) Selecting web providers: The providers usable in CanDID are those that allow the desired attributes Attr to be accessed online. It is critical, however, that users not be able to alter these attribute values. Nationally assigned unique identifiers (e.g., SSNs), as used in our unique-identifier policy, are typically not user modifiable. Attributes used in a string- matching policy, e.g., name and address, may be user-alterable, however. So care is required in choosing providers that prohibit or limit such alteration.
thanks, i hate it
yep! see you there :)
Psyched to be doing a keynote at #37C3 on Dec 27th... I'm gonna talk about Psyops, Hypermemes, UFOs and generative AI. Also launching a CCC-specific instance of CYCLOPS, the CTF/ARG game we've been developing in the studio paglen.studio/37C3/