Tootfinder

Opt-in global Mastodon full text search. Join the index!

No exact results. Similar results found.
@Techmeme@techhub.social
2025-11-18 16:30:55

Google says Gemini 3 Pro scores 1,501 on LMArena, above 2.5 Pro, and demonstrates PhD-level reasoning with top scores on Humanity's Last Exam and GPQA Diamond (Abner Li/9to5Google)
9to5google.com/2025/11/18/gemi

Google launches Gemini 3 with state-of-the-art reasoning, ‘generative UI’ for responses, more
Google today announced Gemini 3 with the goal of bringing “any idea to life.” The first model available in this family is Gemini 3 Pro...

@tiotasram@kolektiva.social
2025-10-20 11:16:23

Day 26: Emily Short
If you know who Short is, you know exactly why she's on this list. If you don't, you're probably in the majority. She's an absolutely legendary author within the interactive fiction (IF) community, which gets somewhat pigeonholed by stuff like Zork when there's actually a huge range of stuff in the medium some of which isn't even puzzle-focused, and Short has been writing & coding on the bleeding edge of things for decades.
I was lucky enough to be introduced to Short's work in graduate school, where we played "Galatea" as part of an interactive fiction class. Short uses a lot of clever parser tricks to make your conversation with a statue feel very fluid and conversational, giving to contemporary audiences a great example of how vibrant interaction with a well-designed agent can be in contrast to an LLM, if you're willing to put in some work on bespoke parsing & responses (although the user does need to know basic IF conventions). While I didn't explore the full range of Galatea's many possible outcomes, it left a strong impression on me as a vision for what IF could be besides dorky puzzles, and I think that "visionary" is a great term to describe Short.
If you'd like you get a feel for her (very early) work, you can play Galatea here: #30AuthorsNoMen

@kexpmusicbot@mastodonapp.uk
2026-01-18 18:53:23

🇺🇦 #NowPlaying on KEXP's #Roadhouse
Richie Havens:
🎵 Follow
#RichieHavens
billwhite.bandcamp.com/track/f
open.spotify.com/track/7MUB5N5

@BBC3MusicBot@mastodonapp.uk
2026-01-20 00:30:28

🔊 #NowPlaying on #BBCRadio3:
#ThroughTheNight
- María Dueñas plays Bruch with the NDR Elbphilharmonie
Violinist María Dueñas, recipient of the 2025 Gramophone Instrumental and Young Artist of the Year awards, plays Bruch's Violin Concerto no 1.
Relisten now 👇
bbc.co.uk/programmes/m002q16k

@tiotasram@kolektiva.social
2026-01-18 23:17:10
Content warning: ICE & resistance

In case anyone was wondering about the relevance of #LandBack in the current moment, via CrimeThinc an article on the Minneapolis resistance states:
"""
The Whipple, a federal building in Fort Snelling on the outskirts of the Minneapolis and St. Paul, has long been a regional headquarters for ICE, having previously housed other federal agencies. The complex is located across the street from a National Guard base, down the road from a military base, and next to the preserved fort itself. The fort sits on the sacred site of the convergence of two rivers. It was one of the earliest sites of colonization in the area; at one time, it was a concentration camp holding native Dakota people.
"""
If at any point in the past you ever felt that maybe Native soverignty was a niche issue, or so far from being realized that other causes were more important or relevant, things like this are a good reminder that that cause: overturning the colonial order, is the *same* cause as any meaningful change from the fascist status quo. Things like a "return to democracy" aren't necessarily bad, but the rot runs to the root of this nation, and any intervention that doesn't go that deep is going to leave us right back in this situation again later on.
The fact that ICE is detaining Native Americans is not at all a mistake given their white supremacist aims.
Article link: #ICE #LandBack

@kexpmusicbot@mastodonapp.uk
2026-01-20 01:14:33

🇺🇦 #NowPlaying on KEXP's #DriveTime
OutKast:
🎵 Phobia
#OutKast
kemetic7ntr.bandcamp.com/track
open.spotify.com/track/0rSSvnF

@Techmeme@techhub.social
2025-11-20 10:40:44

Sources: Salesforce's Slack tells customers in China, Hong Kong, Macau, and Taiwan that they must migrate their accounts to partner Alibaba before February (The Information)
theinformation.com/briefings/s

Salesforce’s Slack to Stop Direct Service in China, Farm Out to Alibaba
Salesforce’s Slack has told customers in mainland China, Hong Kong, Macau and Taiwan that they must migrate their accounts to partner Alibaba Group before February, if they want to continue with instant-messaging service, according to a Slack customer in mainland China and an Alibaba Cloud employee. The move is in response to a privacy and data security law that came into effect this year in

@Techmeme@techhub.social
2025-11-19 16:46:55

A new bipartisan Senate bill amends Section 230 to make commercial social media platforms responsible if their recommendation systems cause foreseeable harms (Lauren Feiner/The Verge)
theverge.com/policy/824054/alg

Lawmakers want to let users sue over harmful social media algorithms
Sens. John Curtis (R-UT) and Mark Kelly (D-AZ) introduced the Algorithm Accountability Act, which would reform tech’s liability shield Section 230.

@Techmeme@techhub.social
2025-11-18 19:25:59

Windows 11 Copilot AI hands-on: despite Microsoft advertising the feature as able to "understand you", it delivers inconsistent and often incorrect responses (Antonio G. Di Benedetto/The Verge)
theverge.com/report/822443/mic

Talking to Windows’ Copilot AI makes a computer feel incompetent
Microsoft is advertising its Windows Copilot AI as “the computer you can talk to.” How does that hold up to testing, and how does it track with CEO Satya Nadella’s ambitions?

@Techmeme@techhub.social
2025-10-20 02:35:34

Alibaba Cloud details a GPU pooling system that it claims reduced the number of Nvidia H20 required by 82% when serving dozens of LLMs of up to 72B parameters (Vincent Chow/South China Morning Post)
scmp.com/business/article/3329

Alibaba Cloud claims to slash Nvidia GPU use by 82% with new pooling system
The new Aegaeon system can serve dozens of large language models using a fraction of the GPUs previously required, potentially reshaping AI workloads.