The piece traces LISP implementations that fit within extremely small memory budgets, starting with the original 1960 LISP on MIT's IBM 704, which used 32K words of 36-bit core with the LISP system itself taking 12K words, leaving under 20K cons cells. It also cites 1964's PDP-1 LISP, which ran on systems with just 4096 words of 18-bit storage, with the system consuming roughly 2000 words.
t3x.org
· 2026-10-25
Dropbox is considering offering employees cognitive exercises such as memory drills, attention-span training and puzzles to counteract possible mental decline linked to heavy AI use at work. The file-sharing company is worried that as staff increasingly offload tasks to AI tools, their own cognitive sharpness could suffer over time.
fastcompany.com
· 2026-10-25
A newly published language called Goose introduces a memory-safe systems programming model that avoids a heap allocator, garbage collector, and lifetime annotations, instead placing all dynamic values on compiler-managed data stacks. Its creators report that across sixteen benchmarks Goose runs 3.3x faster than idiomatic C++, 1.16x faster than hand-optimized C++, and 1.12x faster than the fastest safe Rust code, while using 1.2x to 1.9x less memory.
aardappel.github.io
· 2026-10-24
A review piece from Android Authority outlines how the author disabled several default Google Photos features—including memory notifications, print promotion alerts, and auto-generated collages—to make the app less intrusive. The adjustments are made through the app's Notifications and Create tab settings.
androidauthority.com
· 2026-10-24
SK Hynix said its subsidiary Solidigm will pause its review of pre-IPO and IPO options, citing recent capital market conditions and share price trends. The chipmaker's shares initially fell 3% before paring losses to trade 0.3% lower on the news. SK Hynix said it will keep exploring other ways to fund Solidigm's planned U.S. semiconductor fabrication plant.
cnbc.com
· 2026-10-23
A technology commentary argues that production AI performance depends less on which foundation model is chosen and more on the serving harness around it—the architecture, memory management, scheduling and optimization that turn model intelligence into a working service. It recommends that organizations define latency, concurrency, cost and traffic requirements first, then select models and infrastructure to match, rather than defaulting to the highest-accuracy model.
technologyreview.com
· 2026-10-22
Counterpoint's Q3 data shows global PC shipments dropped to 57 million units from over 70 million a year earlier, a 21 percent year-over-year decline and a 13.8 percent drop from the prior quarter. The firm forecasts a roughly 27 percent year-over-year decline for Q4 and expects weak conditions to persist into 2027. Lenovo remains the top vendor despite a 23 percent decline, HP fell 31 percent, Dell dropped 25 percent, while Apple slipped only 11 percent, gaining relative market share.
engadget.com
· 2026-10-22
A gaming retrospective piece calls for the return of the distinctive save-file menu designs used by sixth-generation consoles like the Dreamcast, GameCube, and PlayStation 2. It highlights how each system rendered saved games with unique visuals, from the Dreamcast's VMU icons to the GameCube's floating cubes and the PS2's 3D model grid.
scabshack.lol
· 2026-10-22
A developer has published documentation for 'In Rainbows,' a specification that defines standardized memory-mapped input/output devices for the ok virtual machine, an instruction evaluator with basic stack-based computation. The spec, named after the Radiohead album, assigns 'magic' memory addresses so that reading or writing to them triggers interactions with devices like keyboards or displays, rather than plain RAM access. The author notes the project is still actively in development, with several devices implemented and others planned.
hitogata.neocities.org
· 2026-10-22
Jensen Huang told CNBC he doesn't expect a memory oversupply by 2029, saying demand for memory such as HBM and LPDDR will keep rising as data centers expand. His comments came after memory stocks fell sharply following a South Korean opinion column that warned of a looming supply flood and potential AI downturn by 2029.
cnbc.com
· 2026-10-22
Industry analysis suggests that surging demand for memory chips, driven largely by AI companies, is raising component costs for Microsoft and Sony as they develop next-generation gaming consoles. Current Xbox and PlayStation hardware has already seen price increases tied to these chip shortages, and the report warns future console prices could rise far more sharply.
fastcompany.com
· 2026-10-22
TornadoVM, the open-source framework that accelerates Java workloads on CPUs and GPUs, has released version 7.3. The update replaces the Java Native Interface with the Foreign Function & Memory API, removes the experimental JVMCI dependency, and standardizes compiler metadata access across JDK 21, 25, 26 and 27 using reflection, ASM and Unsafe. It also adds a CUDA backend that supports Nvidia's cuTile model via a new TileContext and lets developers call libraries like cuBLAS and CUTLASS directly from TornadoVM TaskGraphs, with equivalent support added for Apple's MXL.
infoq.com
· 2026-10-22