惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

博客园 - 【当耐特】
云风的 BLOG
云风的 BLOG
罗磊的独立博客
C
Check Point Blog
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
Blog — PlanetScale
Blog — PlanetScale
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
月光博客
月光博客
大猫的无限游戏
大猫的无限游戏
Google DeepMind News
Google DeepMind News
Engineering at Meta
Engineering at Meta
N
Netflix TechBlog - Medium
宝玉的分享
宝玉的分享
Recent Announcements
Recent Announcements
酷 壳 – CoolShell
酷 壳 – CoolShell
博客园_首页
J
Java Code Geeks
Apple Machine Learning Research
Apple Machine Learning Research
人人都是产品经理
人人都是产品经理
爱范儿
爱范儿
I
InfoQ
Hugging Face - Blog
Hugging Face - Blog
T
Tailwind CSS Blog
B
Blog RSS Feed

Hacker News

GitHub - SeanFDZ/macmind: Single-layer transformer in HyperTalk for the classic Macintosh Show HN: Agent-cache – Multi-tier LLM/tool/session caching for Valkey and Redis Bonsai 1-bit WebGPU - a Hugging Face Space by webml-community Moving a large-scale metrics pipeline from StatsD to OpenTelemetry / Prometheus GitHub - Nightmare-Eclipse/RedSun: The Red Sun vulnerability repository GitHub - SethPyle376/hiraeth: Local AWS emulator focused on fast integration testing, with SQS support, SQLite-backed state, and a debug-friendly web UI. GitHub - macOS26/Agent: Any AI, replaces Claude Code, Cursor, OpenClaw. Over 18 LLM providers (Claude, OpenAI, Gemini, Ollama, Zai, HF, Qwen) wired into a native Mac app that writes code, builds Xcode projects, bumps versions, manages git, automates Safari, use AppleScript, JS or Accessibility, extend Agent! w/ MCP Servers, run tasks from your iPhone via Messages. YouTube now lets you turn off Shorts I Made a Terminal Pager Burgers | マクドナルド公式 Commands — HackerNews CLI documentation ChatGPT for Excel PiCore - Raspberry Pi Port of Tiny Core Linux Live Nation illegally monopolized ticketing market, jury finds Google Broke Its Promise to Me. Now ICE Has My Data. Founding Engineer at Adaptional | Y Combinator CRISPR takes important step toward silencing Down syndrome’s extra chromosome GitHub - saffron-health/libretto: The AI toolkit for building reliable browser automations US v. Heppner (S.D.N.Y. 2026) no attorney-client privilege for AI chats [pdf] Retrofitting JIT Compilers into C Interpreters IPv6 – Google The Accursèd Alphabetical Clock Cybersecurity Looks Like Proof of Work Now Fragments: April 14 Cal.com Goes Closed Source: Why AI Security Is Forcing Our Decision | Cal.com - Scheduling Software for Online Bookings Laravel raised money and now injects ads directly into your agent When moving fast, talking is the first thing to break Too much Discussion of the XOR swap trick – Heather Cafe Introduction to Spherical Harmonics for Graphics Programmers The Grand Line
The Doctor Who Treats Patients With a Gaming Mouse
Josh Centers · 2026-06-16 · via Hacker News

Dr. James Ries sees patients in 37 states. He never touches a stethoscope, never walks a hospital floor, and almost never types a full sentence into a chart.

His main clinical instrument is a Razer Naga V2 Pro. It’s the same mouse World of Warcraft raiders use to fire off 12 abilities without lifting their fingers from the buttons.

That sounds like a joke. It isn’t.

Dr. Ries is the founder of Twenty Mile Medical, a telehealth practice handling urgent care, mental health, and weight management visits across most of the country. He runs his entire clinical workflow through a setup that pairs an MMO gaming mouse with TextExpander Snippets, and he says it’s the single most important tool in his practice.

We had Dr. Ries on a recent webinar to explain the setup. What we expected to be a conversation about clinical documentation turned into something stranger and more useful: One of the most interesting productivity workflows we’ve seen anyone build on top of TextExpander, and a set of principles that work just as well for a customer support team or a law firm as they do for a telehealth practice.

The problem he was trying to solve

Telehealth has a quality-control problem that nobody talks about much. You hire good clinicians. You give them logins and patient queues. You assume the training they brought from prior practices is enough to deliver consistent care.

It isn’t.

“Every provider had their own version,” Dr.Ries told us. “Some are really meticulous, some are really fast, but they missed out on things. The patients themselves experience completely different interactions depending on who they saw.”

The drift wasn’t about clinical judgment, but about everything that surrounds the judgment: Documentation, patient communication, follow-up instructions, the empathetic phrasing that turns a transactional encounter into something resembling care. Discharge instructions going out at 8 am from a fresh provider look nothing like the ones going out at 4 pm from the same provider after eight hours of visits.

That’s a workflow problem, not a training problem.

What a monster Snippet actually is

Most TextExpander users build dozens of small Snippets, each tied to a specific phrase. You type ;sig and your signature appears. Type ;addr and your address fills in. Predictable, useful, and learnable in an afternoon.

Dr. Ries went the other direction.

He built what he calls monster Snippets. Each one is a single Snippet containing an entire clinical scenario with branching options inside it. One Snippet for sinus infections that covers the introduction, the assessment, prescription options, patient instructions, and follow-up plan. Another for psych refills. Another for general patient communication. Each is essentially a small interactive form that the provider walks through in seconds.

When he expands one, he gets a fill-in dialog with checkboxes, dropdown menus, and optional sections. He picks what applies, removes what doesn’t, and the Snippet generates a complete, clinically appropriate response. The whole thing takes about as long as a normal conversation, except every required element is there, every time.

One scenario. One Snippet. Everything inside it.

Watch Dr. Ries run a psych refill, l and you see why this matters. A returning patient, minimal improvement, looking to increase the dose. Some sexual side effects. Depression is controlled, but anxiety is still running high. He clicks through the built-in options, bumps the medication to 100 mg, and the entire note is ready: The assessment, the medication change, the check-in plan, the internal follow-up, the handouts, and the suicide resources.

Every required element is there because the Snippet carries them. “You’re not worn out,” he says, “because it remembers to send them everything.”

Not every Snippet is a monster. Dr. Ries keeps single-purpose buttons alongside the big ones: One for his signature, one for his intro. The branching Snippets handle whole clinical scenarios. The small ones handle the things he types constantly. Both live on the same mouse, a thumb-press away.

Why the mouse matters

You can run monster Snippets from any keyboard. Dr. Ries could type abbreviations like everyone else. He doesn’t, and his reasoning is worth understanding even if you never buy a gaming mouse.

The Razer Naga V2 Pro has 12 programmable buttons on the side panel. Dr. Ries maps each button to a single character, typically one of the symbols above the number keys he doesn’t use in normal typing. Each symbol is the abbreviation for one of his monster Snippets.

Dr. Ries Razer Naga Pro gaming mouse

A press of his thumb fires the Snippet. That’s it.

“If you design a system where you’re not searching and you’re not typing,” Dr. Ries said, “you eliminate the cognitive load pretty significantly.”

The science backs him up. Research published in BMJ Open found that physicians make an average of 13.4 clinically relevant decisions per patient encounter, most of them tied to defining the problem, evaluating test results, and weighing drug therapies. Decision fatigue is well documented in clinical settings: Providers make worse calls later in the day, not because they’re bad clinicians but because every micro-decision burns a small amount of mental energy and the deficit accumulates over a shift.

Dr. Ries isn’t saving keystrokes so much as removing the small interruptions that come with picking the right Snippet, remembering the abbreviation, and typing it correctly. His thumb knows where the button is, so his mind stays on the patient.

His PAs don’t all use the gaming mouse. Some prefer a small shortcut remote, a wireless keypad with programmable keys, held in the off hand while the dominant hand stays on the regular mouse. Same principle, different hardware.

Hardware Best for Tradeoff
MMO gaming mouse (12+ side buttons) Solo operators, heavy daily use Steep learning curve to memorize the button map
Wireless shortcut remote Providers who want separation from the pointer Adds a device, but with easier adoption
Stream Deck or similar People who want visible labels More desk space, less seamless than handheld

A webinar attendee asked Dr. Ries specifically about the Stream Deck. He’d tested one, but found it too clunky. “You’re taking your hands off the mouse to do it. The mouse is very seamless, and I forget I’m even using it.”

How to set this up yourself

Configuration is simpler than you’d think.

  1. Pick a gaming mouse with at least 12 side buttons. The Razer Naga line is the obvious choice. Logitech and Corsair both make competitive options.
  2. Install the manufacturer’s software. Every gaming mouse ships with a configurator that lets you map individual buttons to keystrokes.
  3. In TextExpander, create Snippets with single-character abbreviations. Symbols like ~, |, \, or backtick that almost never show up in normal writing could be good candidates. Skip the obvious ones like @ and # since you’ll use those for email and hashtags.
  4. In the mouse software, map each side button to one of those symbols.
  5. Press the button. The symbol fires, TextExpander expands the Snippet, and the response writes itself.

One thing worth knowing if you go this route: Avoid setting up macros that send long keystroke sequences from the mouse software. Map a single character and let TextExpander handle the rest. This keeps your Snippets editable from inside TextExpander where they belong, and the whole system stays portable if you ever switch mice.

Standardization as patient safety

Here’s where the conversation got interesting.

This is Dr. Ries on what actually drives mistakes in telehealth: “Most of the time it’s not the wrong diagnosis, it’s an omission error. The provider didn’t document a key piece of information, forgot to send the patient education, or didn’t include some critical follow-up instructions or discharge instructions.”

The errors don’t come from bad clinical reasoning. They come from a tired provider finishing their fortieth visit and forgetting one of the dozen small things that need to happen at the end of every encounter.

Monster Snippets fix this structurally. The follow-up instructions are in the Snippet. The interaction warning is in the Snippet. The screening question is in the Snippet. The provider can’t forget them because the system doesn’t let them.

Dr. Ries frames this as a clinical leadership responsibility, not an IT one. He builds and maintains the Snippet library himself. He reviews the language. He pushes updates when guidelines change. When a PA finds new empathetic phrasing that works well with patients, Dr. Ries tests it, validates it, and adds it to the shared library so every provider on the team benefits from that single discovery.

The lesson travels outside healthcare almost too easily. Any team with customer-facing communication has the same drift problem. Your morning support reply doesn’t look like your afternoon one. Your best closers’ emails don’t look like your newest hire’s. That variance isn’t a training issue; it’s a tooling issue. Standardize the floor, and the whole team improves without anyone losing their voice.

The empathy angle

We didn’t expect this part.

Dr. Ries kept coming back to empathy. Telehealth, by its nature, strips out a lot of what makes in-person care feel like care. You can’t read body language as well. You can’t lean in. You can’t put a hand on a shoulder. The visit is shorter, more transactional, and more easily reduced to a checklist.

Ries’s argument is that monster Snippets give him time and energy back, which is what he spends on the part of the visit that actually matters. “I actually have done a lot of Snippets that are just based on one keystroke or one press,” he said. “The patient is more like having a conversation, but you’re able to get more information out quickly and empathetically.”

The Snippets also bank good empathetic language over time. When a particular phrasing of a difficult message lands well with a patient, Dr. Ries adds it to the library. Other providers benefit from the phrasing he found. New hires inherit eight years of accumulated bedside manner from day one.

That’s a remarkable thing for a tool to do.

Steal his Snippets

Dr. Ries published his complete clinical Snippet group as a public group on TextExpander. It’s called the Telehealth Starter Pack, and it includes the monster Snippets he showed in the webinar: Patient communication templates, urgent care medication references, psych note structures, the works.

Add the Telehealth Starter Pack to your TextExpander account and start using it today. If you don’t have TextExpander yet, start a free trial, and the Snippet group will be waiting.

A reasonable starting workflow if you’re new to this approach:

  • Add the Telehealth Starter Pack to see what a real monster Snippet looks like in the wild
  • Pick one scenario in your own work where you type the same kind of thing repeatedly
  • Build one monster Snippet for that scenario with fill-in fields for the parts that change
  • Map it to a single symbol abbreviation
  • If it sticks, map that symbol to a button on a programmable mouse or shortcut remote

You don’t need to convert your entire workflow on day one. Dr. Ries built his over eight years, one addition at a time. His rule is simple: When you find yourself typing the same thing repeatedly, add it to the monster Snippet and never type it again. The right first Snippet is the one you’ll use 10 times this week.

Where this goes next

The full webinar with Dr. Ries is available on demand. He walks through specific monster Snippets and takes audience questions about hiring, training, and updating Snippet libraries as clinical guidelines change.

What we keep thinking about, days after the conversation: The gaming mouse is the hook, but it’s the last small optimization on top of a workflow that was already excellent. The real work lives in the Snippets themselves:

  • Deciding what belongs in them. 
  • Maintaining them over the years.
  • Treating workflow design as a clinical leadership problem rather than something to delegate.

Anyone who builds processes for other people to follow should be paying attention.