@fanf@mendeddrum.org2026-05-17 11:42:02
from my link log —
Faster zlib/DEFLATE decompression on arm64 and x86.
https://dougallj.wordpress.com/2022/08/20/faster-zlib-deflate-decompression-on-the-apple-m1-and-x86/
saved 2022-08-20
@fanf@mendeddrum.org2026-05-17 11:42:02
@datascience@genomic.social2026-05-16 10:00:01
The {esquisse} package makes it easy to plot your data in different ways with a drag and drop interface: #rstats
@bthalpin@mastodon.social2026-05-15 10:29:35
For "reasons" I've been increasingly using ffplay to play music from the command line (e.g., formats that make mplayer & vlc glitch), so I wrote a script to make it play playlists:
while read p; do
ffplay -nodisp -autoexit $p
done < $1
3 LOC, no AI!
@Techmeme@techhub.social2026-06-01 00:10:36
Twitch rolls out Dual Format, a feature that lets creators stream horizontally and vertically simultaneously, and 2K streaming for partners and affiliates (Cheyenne MacDonald/Engadget)
https://www.engadget.com/2184452/twitch-cr
@Mediagazer@mstdn.social2026-07-14 14:25:45
RoseBerry Media launches vertical streamer Epis, featuring originals and versions of BBC, Fremantle, and All3Media shows adapted for vertical viewing (Scott Roxborough/The Hollywood Reporter)
https://www.hollywoodreporter.com/tv/tv-ne
@bourgwick@heads.social2026-05-11 04:02:22
another album of lost #eno songs, along with the pre-"another green world" demos. https://www.moredarkthanshark.org/eno_int_blitz-nov83.html
@Mediagazer@mstdn.social2026-05-20 17:05:53
Google says it is testing new ad formats in search results and AI Mode, including Conversational Discovery ads, Highlighted Answers, and AI-powered Shopping ads (Anu Adegbola/Search Engine Land)
https://searchengineland.com/google-tests-ne…
@nfdi4culture@nfdi.social2026-07-03 14:22:08
Hier kommen noch ein paar Lesetipps aus unserem Culture Information Portal 🌞
📙Erfolgreicher Auftakt des neuen NFDI4Culture-Jour-fixe-Formats mit dem FID-Netzwerk: https://nfdi4culture.de/id/E6854
📗Workshop zum Performing Arts Thesaurus (PAT):
@azonenberg@ioc.exchange2026-05-29 20:27:06
Can anybody get me some waveform captures (in native ngscopeclient .scopesession format, or any of the supported binary/CSV waveform formats we can import) for CAN-FD?
There's a six year old ticket against libscopehal for adding FD support to the CAN protocol decoder but I don't have anything around that speaks it which makes developing a decode tricky.
@lil5@social.linux.pizza2026-05-03 10:00:52
@jake4480@c.im2026-06-28 20:52:20
New blog post on the importance of owning physical (and digital) media, when possible: https://spacetimetech.wordpress.com/2026/06/28/the-importance-of-owning-physical-and-digital-media-when-possible
@Techmeme@techhub.social2026-05-20 17:56:09
Google says it is testing new ad formats in search results and AI Mode, including Conversational Discovery ads, Highlighted Answers, and AI-powered Shopping ads (Anu Adegbola/Search Engine Land)
https://searchengineland.com/google-tests-ne…
@mxp@mastodon.acm.org2026-04-27 16:54:28
@… @… Yes, the idea is much simpler, and the data formats they use are also limiting the complexity. So, it’s treated as a problem akin to, say, file conversion: Markdown − DOCX − LaTeX − troff − Markdown should id…
@mxp@mastodon.acm.org2026-04-27 16:54:28
@… @… Yes, the idea is much simpler, and the data formats they use are also limiting the complexity. So, it’s treated as a problem akin to, say, file conversion: Markdown − DOCX − LaTeX − troff − Markdown should id…
@vosje62@mastodon.nl2026-05-22 01:14:56
/2
New ad formats built with Gemini coming to Google Search - #Google
https://blog.google/products/ads-commerce/google-marketing-live-search-ads/
@matthiasott@mastodon.social2026-05-22 16:09:24
@… @… Yes, in their screenshots on the blog post about new ad formats, they are labelled as “sponsored.“ Although I don’t know how they might do that inline within a written answer, I suppose they would need to – or perhaps mark …
@arXiv_csPF_bot@mastoxiv.page2026-06-11 07:42:10
The Brain That Goes Quiet: Serving a Large Model's Knowledge at 131 Tokens per Second on an 8 GB Laptop by Removing the Large Model from the Runtime Path
Myeong Jun Jo
https://arxiv.org/abs/2606.12154 https://arxiv.org/pdf/2606.12154 https://arxiv.org/html/2606.12154
arXiv:2606.12154v1 Announce Type: new
Abstract: In earlier work I showed that a 35B-class Mixture-of-Experts model can be loaded and executed on a consumer laptop with 8 GB of GPU memory. That result solved a placement problem and immediately exposed a different one: even correctly placed, the large model needed roughly four seconds to answer, because it was still being invoked at every query. This paper documents what happened when I stopped invoking it. During an offline phase, the large model reads source documents and writes verified answer entries into a structured knowledge store; at runtime, only a lightweight router, a deterministic renderer, and a 1B-class model are active. On the same 8 GB laptop, end-to-end response time fell from approximately 4,465 ms to 518 ms, effective end-to-end throughput rose from 15.7 to 131 tokens per second, and the small model's streaming decode rate held at 226-237 tokens per second with a time-to-first-token of 29-62 ms. The bottleneck is structural: three different large models (Qwen, Gemma, and GLM class) all showed the same multi-second runtime cost, and all three produced usable knowledge stores offline. On a 563-entry store built from seventeen real documents, keyword routing collapsed to 1.5% top-1 accuracy while BM25-based routing reached 92.8% (99.4% top-3), and a confidence gate raised effective top-1 to 98.0% by escalating 12.3% of queries. Exact-match fidelity of the small model ranged from 9/9 to 0/9 across envelope formats carrying identical content. A 16-case verification gate blocked all ten corrupted entries while admitting all six supported ones.
toXiv_bot_toot
@marcel@waldvogel.family2026-05-20 16:15:23
Who would expect anything else?
#Enshittification #Google #AI
@chrysn@chaos.social2026-04-23 22:01:36
@UP8@mastodon.social2026-04-21 22:28:33
🩲 Ubuntu Looks To Strip GRUB To The Bare Minimum For Better Security
#linux
@Techmeme@techhub.social2026-07-01 12:20:53
Sony says all new PlayStation games from both first- and third-party developers will be sold in digital formats from January 2028, ending physical game discs (Stephen Totilo/Game File)
https://www.gamefile.news/p/sony-drops-playstation-discs-2028-ps3-vit…
@Techmeme@techhub.social2026-04-29 19:36:27
Google says the Gemini app can now generate files; supported formats include PDF, Microsoft Word and Excel, Google Docs and Sheets, plain text, and Markdown (Abner Li/9to5Google)
https://9to5google.com/2026/04/29/gemini-app-generate-files/
@Mediagazer@mstdn.social2026-04-28 04:30:35
Regal Cinemas charged $50 for opening-night tickets to Dune: Part Three in premium formats, as the industry adopts airline-style audience segmentation (Ben Fritz/Wall Street Journal)
https://www.wsj.com/business/media/the-50-
@Techmeme@techhub.social2026-06-22 14:25:40
Instagram is testing horizontal video on Instagram for TV, plans to experiment with longer-form storytelling and episodic series, and launches on Samsung TV (Katie Kilkenny/The Hollywood Reporter)
https://www.hollywoodreporter.com/tv/tv-ne
@Mediagazer@mstdn.social2026-06-22 16:10:33
Instagram is testing horizontal video on Instagram for TV, plans to experiment with longer-form storytelling and episodic series, and launches on Samsung TV (Katie Kilkenny/The Hollywood Reporter)
https://www.hollywoodreporter.com/tv/tv-ne
