Kode Vicious
In Search of Quietude

Unless you're writing only for yourself, whatever software you build will need to work with what others are building. So, the meta question is: "How much communication is required?"

queue.acm.org/detail.cfm?i...

Security Is Part of Every Critical User Journey

How else would you make sure that product security decisions serve your customers?

Raise your hand and propose that security outcomes and risks be defined at each step along critical user journeys.

queue.acm.org/detail.cfm?i...

Unleashing the Power of End-User Programmable AI

The design of an AI-first, program-synthesis framework built around a new language, Universalis, designed for workers to read, the computer to execute, and ready to be analyzed and manipulated

queue.acm.org/detail.cfm?i...

Concurrency in #WebAssembly

Experiments in the web and beyond

We discuss how concurrent programs are compiled to #Wasm today given the limitations of multi-core concurrency support, and also discuss further expanding Wasm's concurrency capabilities

queue.acm.org/detail.cfm?i...

When Is WebAssembly Going to Get DOM Support?
Or, how I learned to stop worrying and love glue code

Can I build my C#/Go/Python library/app into my website so it runs with good performance?

queue.acm.org/detail.cfm?i...

WebAssembly: How Low Can a Bytecode Go?:
New performance and capabilities

Wasm is still growing with new features to address performance gaps. With many languages and many targets, Wasm could one day become the universal execution format for compiled applications.

queue.acm.org/detail.cfm?i...

#WebAssembly: Yes, but for What?

WebAssembly has found a niche but not yet filled its habitable space. We look at #Wasm wins and losses, identify patterns, and extract commonalities. From those, we predict the future.

queue.acm.org/detail.cfm?i...

@wingolog.org

Research for Practice:
The Point is Addressing

Even something as innocent as addressing comes from a rich design space filled with tradeoffs between important considerations such as scaling, transparency, overhead, and programmer control.

queue.acm.org/detail.cfm?i...

AI: It's All About Inference Now

Model inference has become the critical driver for model performance.

Test-time compute scaling offers a new, exciting avenue to increase model performance beyond what can be achieved with training

queue.acm.org/detail.cfm?i...

Develop, Deploy, Operate

A holistic model for understanding the costs and value of software development:
- Tensions between factors
- 4 distinct forms of impact
- Warning against measuring against unknown counterfactuals
- Consensus mechanism for estimating DDR costs.

queue.acm.org/detail.cfm?i...

Generative AI at the Edge:
Challenges and Opportunities

We will soon see embodied, federated, and cooperative models become commonplace, enhancing our lives in the background. However, these models won't just compute; they will communicate, create, and adapt.

queue.acm.org/detail.cfm?i...

The Soft Side of Software:
Peer Mentoring

Stop waiting for a senior mentor to appear. Your peers are some of the most valuable mentors you'll ever find. Leverage those relationships, share insights, and bring value to every conversation.

queue.acm.org/detail.cfm?i...

Kode Vicious:
Can't We Have Nice Things?
Careful crafting and the longevity of code

We build apparatus to show some effect. This is how empiricists operate. They have a hunch, build an apparatus, run an experiment, refine the hunch, and then wash, rinse, and repeat.

queue.acm.org/detail.cfm?i...

Drill Bits
Sandboxing: Foolproof Boundaries vs. Unbounded Foolishness

Sandboxing mitigates risks of software so large, it's likely to harbor security vulnerabilities

To safely harness useful yet opaque libraries, a simple mechanism provides confinement—or does it?

queue.acm.org/detail.cfm?i...

Operations and Life
A Clean Approach to Process Optimization

My soap-loading technique isn't revolutionary, but it demonstrates process design: You can eliminate delays in starting a process by front-loading tasks whenever possible.

queue.acm.org/detail.cfm?i...
@yesthattom.bsky.social

Kode Vicious
Analyzing Krazy Kode

Code communicates our intentions to other humans. Code, like prose, has style and color and subtlety. When you open a function and start reading, try to think, "How do I feel about reading this."

queue.acm.org/detail.cfm?i...

From Function Frustrations to Framework Flexibility

Fixing tool calls with indirection

The principle of indirection can be applied to introduce a paradigm shift: replacing direct value manipulation with symbolic reasoning using named variables.

queue.acm.org/detail.cfm?i...

How to Evaluate AI that's Smarter than Us

Evaluating AI models that surpass human expertise in the task at hand presents unique challenges.

Exploring three strategies: functional correctness, AI-as-a-judge, and comparative evaluation

queue.acm.org/detail.cfm?i...
@chiphuyen.bsky.social

String Matching at Scale: A call for interdisciplinary collaboration and better-directed resources

Software engineering is an open-world problem with multiple decentralized players. Abstraction can be a treacherous pathway.

queue.acm.org/detail.cfm?i...

Fifty Years of Open Source Software Supply Chain Security

We are all struggling with a massive shift that has happened in the past 10 or 20 years in the software industry. For decades, software reuse was only a lofty goal. Now it's very real.

queue.acm.org/detail.cfm?i...
@swtch.com

The Surprise of Multiple Dependency Graphs

It should be easy to avoid installing vulnerable OS software, but dependency graphs are complex. Developers and users of a package may end up with different graphs, leading to vulnerabilities.

queue.acm.org/detail.cfm?i...

The Soft Side of Software

My Career-limiting Communication

Use punchy visuals to make content easier to digest with your most important points clearly highlighted. Be thoughtful about your content. You've got a lot riding on it.

queue.acm.org/detail.cfm?i...

@katematsuda.bsky.social

Systems Correctness Practices at AWS:

Leveraging Formal and Semi-formal Methods

Building reliable and secure software requires a range of approaches to reason about systems correctness.

queue.acm.org/detail.cfm?i...

Intermediate Representations for the Datacenter Computer: Lowering the Burden of Robust and Performant Distributed Systems

Raised level of abstraction and support for universal data formats means new avenues for leveraging heterogeneous computing targets

queue.acm.org/detail.cfm?i...

Simulation:
An Underutilized Tool in Distributed Systems

Simulation has a huge role to play in AI systems: We need an efficient, fast, and cost-effective way to train AI agents to operate in our infrastructure, and simulation absolutely provides that capability.
queue.acm.org/detail.cfm?i...

Operations and Life
Give Engineers Problems, Not Solutions

This is about providing the "why" instead of the "how." Instead of dictating specific solutions, present the problem and desired outcome, and let your team figure out how to solve it.
queue.acm.org/detail.cfm?i...
@yesthattom.bsky.social

#KodeVicious
The Drunken Plagiarists
Working with Co-pilots

The trick of an LLM is to use a little randomness and a lot of text to Gauss the next word in a sentence. Seems trivial and not a measure of intelligence. But it's clever and does have some applications

queue.acm.org/detail.cfm?i...

#DrillBits
Retrofitting: Principles and Practice

Saddling time-tested workhorse software with unexpected new burdens is often the best way to seize an opportunity or trounce a perennial annoyance, but adding radically new functionality to production code isn't easy.

queue.acm.org/detail.cfm?i...

You Don't Know Jack About AI...
And ChatGPT probably doesn't either

For a long time, it was hard to pin down what exactly AI was. Fast-forward to 2024, and we all now know exactly what AI is.

AI = ChatGPT.

Or not.

queue.acm.org/detail.cfm?i...

Kode Vicious
Building on Shaky Ground

queue.acm.org/detail.cfm?i...

The CrowdStrike catastrophe happened because of architectural issues in hardware and in systems software. We should be building systems that make writing a virus difficult, not child's play.