







No wonder iTerm2 is so bloated. They just added a web browser inside the terminal. Who asked for this?Empty iTerm2 window now uses 400MB of RAM :( pic.twitter.com/JhkHp4WUjM— Sebastian Aaltonen (@SebAaltonen) September 22, 2025
Suhail on Twitter / X
He who owns the browser, shall own the memory. https://t.co/ENA04zwo82— Suhail (@Suhail) June 24, 2025
The Performance of Open Source SoftwareMemShrink
Firefox has long had a reputation for using too much memory. The accuracy of that reputation has varied over the years but it has stuck with the browser. Every Firefox release in the past several years has been met by skeptical users with the question "Did they fix the memory leak yet?" We shipped Firefox 4 in March 2011 after a lengthy beta cycle and several missed ship dates–and it was met by the same questions. While Firefox 4 was a significant step forward for the web in areas such as open video, JavaScript performance, and accelerated graphics, it was unfortunately a significant step backwards in memory usage.
Chrome silently installs a 4 GB local LLM on your computer
You did remember to opt out of AI, didn't you?

ellen livia ᯅ on Twitter / X
here's how Claude Code actually handles memory : all 8 phases 🧵Our team at @mem0ai use @claudeai a lot, we deeply care about memory. here is a summary of how it works 👇User Input -> Context Assembly -> History System -> API / Query -> Response -> SummaryPhase 1: session… pic.twitter.com/hcZbJzbUxB— ellen livia ᯅ (@ellen_in_sf) March 31, 2026
Software Sucks Now
WebAssembly browser preview · V8
WebAssembly or Wasm is a new runtime and compilation target for the web, now available behind a flag in Chrome Canary!
Reasons not to use Facebook | Hacker News
Granted, he's an incredible programmer who contributed immensely to the development of our modern operating systems and the tools we use, and people here love to bring that up ("what - you DON'T KNOW WHO STALLMAN IS? SHAME? HOW DARE YOU CRITICISE HIM?")
vincent on Twitter / X
Session management is working and now it will keep the model in memory instead of constantly loading. Now to fix the hardest part, gibberish issue.ALMOST THERE ~80% DONE 🔨🔨🔨Demo: @UnslothAI Llama 3.2 1B on CPU3.6 tokens/sec to about 4.1 tokens/sec (gibberish) pic.twitter.com/1tp8dXQrLo— vincent (@t0kenl1mit) August 5, 2025
⚙️ The State of WASM
This is another research about WASM aka WebAssembly from Avareum Team on 2022–03 to explore the new frontier of full stacks development.

ChatGPT will now use its 'memory' to personalize web searches | TechCrunch
ChatGPT will now use its 'memory' to personalize web searches, thanks to a new feature called Memory with Search.

Shlok Khemani on Twitter / X
Super interesting that eve doesn't ship with any long-term or cross-session memory (yet). The main context-management lever out of the box is conversation compaction. This is unlike OpenClaw and Hermes, each of which comes with a default opinionated memory implementation. https://t.co/DDIyc4rFtW— Shlok Khemani (@shloked) June 17, 2026
Introducing the OpenMemory Chrome Extension
AI memory Chrome extension for LLM memory and retrieval augmented generation. OpenMemory enables persistent AI agent memory across web browsing sessions.

I really don’t like ChatGPT’s new memory dossier
Last month ChatGPT got a major upgrade. As far as I can tell the closest to an official announcement was this tweet from @OpenAI: Starting today [April 10th 2025], memory …

Web bloat isn’t a knowledge problem | Christian Heilmann
There is a consensus among all browser makers and lovers of the web: the current state of the web is a mess. The average web page is 2.4 megabyte big and has over 200 requests.
Anemll on Twitter / X
Up to 3.5x faster LLM inference on Apple Neural Engine:ANE is a Tensor Processing Unit, unlike GPU, it requires fixed-shape tensors. The KV cache size is set at compile time. A fixed 4096 context always runs at the slowest speed, even for short replies.Variable Context: start… pic.twitter.com/a1mLlmMYq7— Anemll (@anemll) February 16, 2026
interesting thread about E2EE DMs when deeply embedded in a social app, versus not: content embed visibility, interaction leakage, message history, importance of defaults. (use germ and/or signal!)
Jon Millican
I've seen a lot of chatter today about Instagram ending support for optional End-to-End Encrypted DMs. I helped to lead on cryptography and privacy for E2EE in Messenger and Instagram for many years, until I left around 10 months ago, so it's disappointing to see it being removed from Instagram.