Tootfinder

I wouldn't have thought that quantitative analyses of retrospective national bibliographies would be that painful: data access via SRU, OAI, and REST API; another resource has a JSON dump, another one again consists of various ttl for which you have to set up your own sparql endpoint. And I've not even arrived at formats, metadata standards and cataloguing peculiarities 🤯 so everything's #FAIR

The curse of The vOICe is that it is so mature, safe, globally available and affordable that it is of little interest to investors (big money) or mass media (hype). Better stick proprietary electrodes into the brain: it may cause seizures, hemorrhages, fail beyond repair, etc.
ChatGPT: "It's a curse for The vOICe — but it's a curse that points to its strengths."

ChatGPT - Rozdíl TTL vs CMOS
Shared via ChatGPT

i don't personally do retrocomputing but a lot of the design for #GlasgowInterfaceExplorer is made with an explicit purpose of making retrocomputing nicer (for example, the support for true 5 V TTL that we went to great lengths to preserve, vs. only supporting 3.3 V at most)
i'm quite happy to add support for more retro interfaces to it too

I've read this story a couple of times, though it's been many years. It's maybe not the first King story I'd have thought of making a movie from. And Mark Hamill as the Major? interesting choice.
https://www.imdb.com/title/tt10374610/?ref_=fn_all_ttl_1

The Long Walk (2025) | Horror
The Long Walk: Directed by Francis Lawrence. With Garrett Wareing, Mark Hamill, Judy Greer, Charlie Plummer. A group of teenage boys compete in an annual contest known as "The Long Walk," where they must maintain a certain walking speed or get shot.

My main bugbear with OVH is that they *still* don't offer "failover IP" (where you can fairly quickly move an IP between different dedicated servers) for IPv4, not IPV6.
And, no, they don't offer "Bring Your Own IP" for IPv6 either, only for IPv4.
So when I recently migrated us to a new server I had to do the DNS TTL dance with the IPv6 side of things.
Oh, and it's still only a /64 per server. Not that anything stops you from just using <…

Helix Parallelism: Rethinking Sharding Strategies for Interactive Multi-Million-Token LLM Decoding
Nidhi Bhatia, Ankit More, Ritika Borkar, Tiyasa Mitra, Ramon Matas, Ritchie Zhao, Maximilian Golub, Dheevatsa Mudigere, Brian Pharris, Bita Darvish Rouhani
https://arxiv.org/abs/2507.07120 …

Helix Parallelism: Rethinking Sharding Strategies for Interactive Multi-Million-Token LLM Decoding
As LLMs scale to multi-million-token KV histories, real-time autoregressive decoding under tight Token-to-Token Latency (TTL) constraints faces growing pressure. Two core bottlenecks dominate: accessing Feed-Forward Network (FFN) weights and reading long KV caches. While Tensor Parallelism (TP) helps mitigate the cost of FFN weight reads, it does not scale well for attention. When TP width exceeds the number of KV heads, it leads to inefficient KV duplication, limits parallelism, and constrains…