标签：#us

The Enhanced Games — a singular sporting competition where a majority of the athletes were on performance enhancing drugs — may herald a new business model that the tech industry is ready to embrace.

All tests run on an 8-year-old MacBook Air. You're transferring a batch of large files over MTP. The first one flies at 45 MB/s. Then the second file starts — and you're at 30 MB/s. The third is slower still. Nothing changed. Same cable, same device, same app. So what's happening? The Cause Is in the Protocol Itself Between every file, MTP requires a full negotiation cycle — SendObjectInfo followed by SendObject . This isn't an implementation detail you can optimize away. It's how MTP works. During that gap, a few things happen in sequence: The Android device's flash controller is still committing the previous file to storage The USB pipe is flushed and re-established for the next object The device's MTP stack is processing metadata before it's ready to receive data again The result is a speed dip at every file boundary. The longer the previous file, the longer the device needs to catch up. What I Tried Building HiyokoMTP, I went through the obvious candidates: Tokio thread pool exhaustion — sync Read/Write calls blocking async threads were a real issue. Fixing it improved overall stability, but didn't eliminate the inter-file dip. Chunk size tuning — adjusting the USB bulk transfer buffer (up to 4 MB per chunk) helped peak throughput, but not the boundary behavior. Intentional cooldown between files — adding a short pause actually helped in some cases, giving the device's flash controller time to breathe before the next transfer starts. Why It Can't Be Fully Fixed The inter-file overhead is structural. MTP was designed as a stateful, command-response protocol — not a streaming pipeline. Every file is a discrete transaction with its own negotiation. There's no mechanism to pre-stage the next file while the current one is still writing. Non-async bulk transfer pipelining (similar to io_uring or Zero Copy USB) could theoretically reduce this, but it would require deep nusb-level changes and device-side support that most Android MTP stacks don't expose. MTP vs ADB: A F

A survival guide for when everything goes wrong in production. Your Redis dashboard looks perfect. Hit ratio: 99.2%. Latency: sub-millisecond. Memory usage: 60% of available. Every metric says healthy. Then at 2:47 PM, your API starts returning 500s. Response times spike to 30 seconds. Users can't log in. The dashboard still shows 99% hit ratio because the cache is working — it's serving cached errors to everyone equally fast. Redis is doing exactly what you told it to do. The problem is what you told it to do. Why Single-Threaded Is Fast (Until It Isn't) Redis processes commands on a single thread. No locks. No context switching. No synchronization overhead. One CPU core, fully utilized, can handle 100K+ operations per second because it never waits for another thread to release a lock. The event loop model (similar to Node.js) multiplexes thousands of client connections on a single thread using non-blocking I/O. Read a request, process it, write the response, move to the next. When your commands are simple — GET, SET, INCR — each one takes microseconds. The trap: slow commands block everything. KEYS * on a million-key database? That's a full keyspace scan on the main thread. While it runs, every other client waits. SORT on a large set? Same. LRANGE on a list with 10 million elements? Same. Redis 6.0 introduced I/O threading ( io-threads config) for reading and writing network data on multiple threads, but command execution is still single-threaded. Redis 7.0 improved this further, but the fundamental model hasn't changed. Long-running commands on the main thread stall everything. Rules: Never use KEYS in production. Use SCAN instead — it's cursor-based and returns results incrementally. Watch out for O(N) commands on large data structures: LRANGE , SMEMBERS , HGETALL on million-element structures. Use SLOWLOG to find commands that are blocking the event loop. Pipelining: The Easiest 10x You'll Ever Get Every Redis command involves a network round trip: send request

The honest decision framework, not another flame war. The SQL vs NoSQL debate has been running for 15 years and it still generates more heat than light. Here's the framework that actually helps you decide. The Real Question It's not "SQL or NoSQL." It's: what does your access pattern look like? If your application is mostly reading and writing related data through well-defined queries — orders with line items, users with addresses, products with categories — relational databases are purpose-built for this. JOINs are not expensive when they're indexed. Transactions are not slow when they're scoped correctly. PostgreSQL handles 50 million rows comfortably on a single node. If your application is reading and writing self-contained documents with predictable access by a primary key, and you rarely need cross-document queries — user profiles, product catalogs, content management — a document database simplifies your code. No ORM mapping hell. No migration files for adding a field. If your application writes massive volumes and reads by partition key with eventual consistency — time-series data, IoT telemetry, activity feeds at scale — wide-column stores like Cassandra were built for this specific workload. The 2026 Reality PostgreSQL has eaten NoSQL's lunch in many areas. JSONB support means you can store and query unstructured data inside PostgreSQL with GIN indexes. You get the document model flexibility without giving up transactions, JOINs, and a 30-year ecosystem. For 80% of startups and mid-size companies, PostgreSQL is the only database you need. MongoDB has gotten more relational. Multi-document ACID transactions (since 4.0), schema validation, aggregation pipelines that look suspiciously like SQL. It's converging toward what PostgreSQL already does, but with a different starting point. DynamoDB dominates serverless. If you're in AWS and your access pattern is simple key-value with known query patterns, DynamoDB's pricing model (pay-per-request) and operational s

A survival guide for when everything goes wrong in production. There's a moment every engineer who works with Kafka experiences. You check the producer. Messages are sending. You check the consumer. Nothing. The consumer group shows zero lag because there's nothing to lag behind — as far as the consumer knows, the topic is empty. But it's not empty. The messages are there. Somewhere. In some partition, at some offset, behind some configuration you set six months ago and forgot about. Kafka doesn't lose messages. But it's very good at hiding them from you. Consumer Lag: The Number Everyone Watches Wrong Consumer lag is the difference between the latest offset in a partition and the offset your consumer group has committed. Simple concept. Dangerous in practice. The mistake: treating lag as a single number. Lag is per-partition. If you have 30 partitions and one consumer is stuck on partition 17 while the others are healthy, the total lag looks manageable. But partition 17's data is hours behind, and whatever downstream system depends on that data is serving stale results. Monitor lag per partition. Tools like Burrow, Kafka Exporter for Prometheus, or even kafka-consumer-groups.sh --describe break it down. If one partition's lag is growing while others are stable, you have a stuck consumer, a hot partition, or a poison message. A poison message is a record your consumer can't process — malformed data, unexpected schema, null where it shouldn't be null. The consumer throws an exception, the offset doesn't commit, and it retries the same message forever. Lag grows. The consumer looks "alive" because it's processing — just not making progress. The fix: dead letter queues. After N retries, move the message to a separate topic, commit the offset, and move on. Alert on the dead letter topic. Investigate later. Don't let one bad record block millions of good ones. Rebalance Storms: The Silent Killer Consumer rebalancing is Kafka's mechanism for redistributing partitions acro

SQL-like Queries in FSRS Plugin for Obsidian Spaced repetition in Obsidian usually works as "show all cards with due earlier than today." That's enough for simple cases, but once you have hundreds of notes, you want to filter, sort, and select. My FSRS plugin now has a query language resembling SQL. It turns a markdown block into a live table that updates with every review. ``` fsrs-table SELECT file as "Note", r as "Retrievability", date_format(due, '%d.%m.%Y') as "Due" WHERE r < 0.7 ORDER BY r ASC LIMIT 20 ``` → the table shows the 20 most "forgotten" cards, sorted by retrieval probability. From Simple Settings to an Embedded DB Initially I planned to offer table settings using standard SQL syntax. But pretty quickly the syntax became a real query language, and the implementation itself — an embedded lightweight DB. High-level test coverage in TypeScript made it easy to iterate on functionality located in the WASM module via an AI agent. When faced with dual-language testing (TypeScript + Rust), the artificial intelligence prefers to do the job properly rather than fake it. After implementing the lexer → parser → AST → evaluator pipeline for numeric values, I extended it to strings, added filtering via WHERE, then functions. Extending the syntax or adding a function came down to a single request to the agent — and a feasibility check. What's Inside fsrs-table Supported Features SELECT — choose fields, rename via AS . WHERE — conditions with = , != , < , > , <= , >= , AND , OR . ORDER BY — sort ascending ( ASC ) or descending ( DESC ). LIMIT — cap the number of rows. date_format() — convert the due date to any text format. Available fields: Field (alias) Type Description file string path to the note due date next review date stability (s) number stability in days difficulty (d) number difficulty retrievability (r) number probability of recall (0…1) reps number total number of reviews state string New, Learning, Review, or Relearning elapsed number days since last r

GitHub热门项目 | A Kiro Client in Rust | Stars: 1,560 | 5 stars today | 语言: Rust

GitHub热门项目 | 💥 Blazing fast terminal file manager written in Rust, based on async I/O. | Stars: 38,839 | 74 stars today | 语言: Rust

Happy to announce HeliosProxy !! Far beyond a pooling tool, HeliosProxy ** is a next-gen programmable Postgres data-plane. **Works with PostgreSQL-compatible databases , not only HeliosDB. It starts as a PgBouncer-compatible wedge, then adds the operational surface teams usually build from multiple tools: connection pooling failover and transaction replay shadow execution anomaly detection edge cache controls admin REST API embedded admin UI signed WASM plugins OCI-style plugin artifacts Kubernetes operator Terraform and Pulumi providers 22 installable Claude/Codex operator skills Install operator skills: heliosdb-proxy install skills PostgreSQL #DevOps #SRE #Database #AIcoding

This is a submission for the GitHub Finish-Up-A-Thon Challenge What I Built pq — jq for Parquet. A 50 MB Rust single binary that wraps DuckDB's query engine in a jq-style expression DSL, optimized for terminal one-liners and unix pipes. $ pq sales.parquet 'group_by .country | sum .revenue | top 3 by sum_revenue' ┌─────────┬─────────────┐ │ country ┆ sum_revenue │ ╞═════════╪═════════════╡ │ US ┆ 19065.00 │ │ FR ┆ 999.99 │ │ DE ┆ 312.00 │ └─────────┴─────────────┘ Where it started. I work in adtech. I look at parquet files dozens of times a day — campaign deliveries, partner exports, audience snapshots. Every existing option was painful: Tool Pain pyarrow / pandas 5-second cold start, 200 MB virtualenv parquet-tools JVM, slow, no query support pqrs Inspector only — can't filter or project duckdb CLI Great engine, but SELECT email FROM 'file.parquet' WHERE country='US' is too verbose to type 50 times a day Spark Are you serious pq is the tool I actually want — single binary, no JVM, no Python, jq-style syntax for piping into the rest of the unix toolbox. It's been my default cat for parquet since v0.5. Demo Repo : github.com/thehwang/parq Latest release : v0.14.0 (this submission) Install : brew install thehwang/parq/pq Tutorial : doc/tutorial.md — 30-minute hands-on walkthrough A taste of what shipped in v0.14: # Streaming JSON output (was the only buffered format until v0.14) $ pq big.parquet '.id, .country' -o json | head -c 200 # returns instantly even on a 40 GB file # Schema-drift gate for CI $ pq diff baseline.parquet candidate.parquet # Schema diff - a: ` baseline.parquet ` - b: ` candidate.parquet ` ## Added (1) | column | type | nullable | |-----------|---------|----------| | ` country ` | VARCHAR | yes | $ echo $? 1 # exits non-zero on drift, slots into CI without scripting And the new TUI Explain panel — press capital E for EXPLAIN ANALYZE , get row-group pruning per scan (this is exactly the panel you see on the cover image at the top of this post): Expla

I haven't seen anything as stupid as the WeWork IPO document in a very long time - that is, until Elon Musk filed to take SpaceX public. WeWork was a joke. SpaceX is a threat. And if Musk and his bankers have their way, you are going to be their bagholder. Lots of the top-line […]

A few weeks ago I posted about building a desktop app that automates GitHub commits because...

The activewear giant has used chemical recycling to make jersey for 16 teams competing in the tournament. But the technique is unlikely to help solve fashion’s waste issue.

On May 29, 2026, OpenAI published its Frontier Governance Framework — and most developers moved on to the next item in their feed. That’s a mistake worth correcting. The document doesn’t announce a new model or lower an API price. It describes how OpenAI measures whether its own systems could enable mass-casualty events, what access controls gate who can reach those capabilities, and how this maps to the regulations — the EU AI Act and California’s Transparency in Frontier AI Act — that are actively shaping compliance requirements for any enterprise deploying frontier AI this year. If you build security tools on OpenAI APIs, the framework’s Trusted Access for Cyber program directly affects what your application can and cannot do. If you operate in a regulated environment, the framework is the vendor-side accountability document your compliance team needs to reference. And if you build on frontier models at all, the risk tier system in this framework governs the capability restrictions you will encounter — and, increasingly, what auditors and procurement teams will ask about when vetting your AI vendor stack. What the Framework Actually Is The Frontier Governance Framework is OpenAI’s published methodology for evaluating the risk profile of frontier models before and after deployment. It covers six functional areas: risk assessment and mitigation, model reporting, security risk management, incident response, external expert input, and framework updates. Each area has defined processes, thresholds, and accountability mechanisms. The core architecture is a tier system applied across four risk domains. Each domain is evaluated independently, with tiers reflecting capability levels that could enable specific categories of harm. A model’s rating in any domain determines what deployment controls apply — what gets blocked at the API layer, who gets elevated access, and what triggers an incident response workflow. The framework was published explicitly to align with two regu

All tests run on an 8-year-old MacBook Air. All results from shipping 7 Mac apps as a solo developer. No sponsored opinion. The most common Tauri v2 frustration: you write a command, invoke it from the frontend, and nothing happens. No error. No crash. Just silence. It's almost always permissions. How Tauri v2 permissions work Tauri v2 introduced a capability system. Every plugin action — reading files, executing shell commands, sending notifications — requires an explicit permission declaration in your config. Without the permission, the plugin call fails silently on the frontend. The Rust code never runs. // src-tauri/capabilities/main.json { "identifier" : "main-capability" , "description" : "Permissions for main window" , "windows" : [ "main" ], "permissions" : [ "core:default" , "fs:read-all" , "fs:write-all" , "shell:allow-execute" , "opener:allow-open" , "global-shortcut:allow-register" , "global-shortcut:allow-unregister" ] } Note: As of Tauri v2.1, shell:allow-open is deprecated. Use tauri-plugin-opener and opener:allow-open instead. The debugging flow When a command does nothing: Open DevTools ( Cmd+Option+I in dev mode) — check the console for a rejected Promise or permission error Check your terminal output — the Rust side logs errors directly in the tauri dev terminal; look for lines like [tauri] permission denied or not allowed Enable verbose logging — set RUST_LOG=tauri=debug before running tauri dev for more detailed backend output Check your capabilities file — missing or misspelled permission identifiers are the #1 cause Permission errors in the console typically look like a rejected Promise with a message such as plugin:shell|execute not allowed . The capabilities file is always the first thing to check. Common permissions you'll need "permissions" : [ "core:default" , "fs:read-all" , // read any file "fs:write-all" , // write any file { "identifier" : "shell:allow-execute" , "allow" : [{ "name" : "my-cmd" , "cmd" : "adb" , "args" : true }] }, "op

What is nbwipers? nbwipers is a CLI tool that strips outputs and metadata from Jupyter notebooks before git commit. Written in Rust - faster than nbstripout Supports git clean filter Works with .ipynb files Why use it? Jupyter notebooks store cell outputs inside the .ipynb file (JSON). This causes problems: Noisy diffs - output changes pollute every commit Repo size - images and large outputs bloat the repo Security - sensitive data can leak in outputs (API keys, query results) The solution: strip outputs automatically on git add via a clean filter. Why not nbstripout? nbstripout is written in Python. It is slow - git status , git diff , and git add all became noticeably slow on this repo because nbstripout was invoked for every .ipynb file. The main cause is Python startup time. With 100+ notebooks, nbstripout can take 40+ seconds where a Rust-based tool takes ~1 second. Faster alternatives: Tool Language Notes nbstripout-fast Rust Up to 200x faster; no git filter install support nbwipers Rust Inspired by nbstripout-fast; adds git filter + pyproject.toml config nbwipers is essentially nbstripout-fast with better git integration. Switching to nbwipers fixed the slowness. Setup 1. Install felixgwilliams/nbwipers is now in the aqua registry as of v4.517.0 . Using aqua , add to aqua.yaml : packages : - name : felixgwilliams/nbwipers@v0.6.2 Then run: aqua install 2. Configure git filter Run once per repo (writes to .git/config ): git config filter.nbwipers.clean "nbwipers clean -" git config filter.nbwipers.smudge cat git config filter.nbwipers.required true Or edit .git/config directly: [filter "nbwipers"] clean = nbwipers clean - smudge = cat required = true required = true makes the commit fail if nbwipers is not installed. This prevents accidentally committing outputs. 3. Add .gitattributes In the repo root, add .gitattributes : *.ipynb filter=nbwipers **/.ipynb_checkpoints/*.ipynb !filter **/.virtual_documents/*.ipynb !filter The !filter lines exclude checkpoint an

The Enhanced Games — a singular sporting competition where a majority of the athletes were on performance enhancing drugs — may herald a new business model that the tech industry is ready to embrace.

The Problem We Were Actually Solving I still remember the day our treasure hunt engine started to show its weaknesses. We had been using a custom-built solution written in Java, and it had served us well until our user base grew exponentially. The engine, which relied heavily on recursive searches and dynamic memory allocation, began to cause performance issues and occasional crashes. Our team was under pressure to find a solution that would allow our server to scale without sacrificing the user experience. After some research, I became convinced that Rust was the answer to our problems. Its focus on memory safety and performance seemed like the perfect fit for our needs. What We Tried First (And Why It Failed) Our first attempt at solving the problem was to simply translate our Java code into Rust. We thought that the language's built-in features would automatically solve our performance and memory issues. However, we quickly realized that this approach was not going to work. The Rust compiler was complaining about lifetime issues and borrow checker errors, which we did not fully understand at the time. We spent weeks trying to fix these issues, but our code was still not stable. I recall one particularly frustrating error message from the Rust compiler: error: cannot borrow self.list as mutable because it is also borrowed as immutable. It was then that I realized we needed to take a step back and rethink our approach. The Architecture Decision We decided to start from scratch and redesign our treasure hunt engine with Rust's strengths in mind. We chose to use a graph-based data structure, which allowed us to take advantage of Rust's ownership model and avoid common pitfalls like null pointer dereferences. We also made use of the crossbeam crate for parallelism and the tokio crate for async I/O. This new design required us to think differently about our problem domain, but it ultimately led to a more efficient and scalable solution. I was impressed by the level of

SpaceX already generated one-fifth of its 2025 revenue from government contracts, the company revealed in its IPO filing.

You stopped being the loop Most people use Opus 4.8 the way they used every model before it: open a chat, type a request, watch the cursor, correct it, repeat. That's a conversation. A dynamic workflow is something else entirely. The shift is this: you stop being the loop. Instead, an orchestrator — plain code you control — spawns subagents you design, fanning out work in parallel, running steps in sequence, judging and merging results, and reporting back when the whole thing is done. Opus 4.8 can drive hundreds of parallel subagents inside a single workflow, with effort control per node so cheap steps stay cheap and hard steps think harder. In this tutorial you'll learn the core patterns by building one concrete thing: a pull-request reviewer that fans out across correctness, security, and performance, then adversarially verifies every finding before it reaches you. // You design the shape. The orchestrator runs it. const found = await parallel ( DIMENSIONS . map ( d => () => agent ( d . prompt , { schema : FINDINGS }))) const deduped = dedupeByFileLine ( found . flatMap ( r => r . findings )) const verified = await parallel ( deduped . map ( f => () => agent ( refutePrompt ( f ), { schema : VERDICT }))) const real = verified . filter ( v => v . refuted === false ) By the end you'll know when to reach for parallel() versus pipeline() , how structured output schemas keep subagents composable, and where to set effort per node. The mental model: it's a graph, not a prompt Stop thinking "I send a prompt, I get a completion." Start thinking: an orchestrator runs a workflow graph, and each node is an agent call. The orchestrator is plain code. It decides what runs, in what order, and what to do with each result. Subagents are the leaf workers — each gets a focused prompt, a structured-output schema, and its own effort setting. The unit of work is no longer the prompt; it's the graph. Two primitives compose every graph, and the difference between them is entirely about ba