Writing

MAY 4, 2026

The One-Way Doors That Remain

Almost every practice we have for building software was designed around an assumption that’s quietly stopped being true. Test coverage budgets, code review queues, refactoring debt, junior hiring pipelines, the ratio of product managers to engineers, the number of services we maintain, the cost of saying no to a feature. Each of these is a knob set somewhere on the dial, and the position of the dial was determined by a single underlying parameter: writing code was slow and expensive.

MAR 16, 2026

Don't Juggle Other People's Balls

A few months ago I stumbled across Murray Bowen’s work on overfunctioning and underfunctioning in family systems. It’s a framework from therapy, not management, but the pattern it describes hit uncomfortably close to home. It gave me a name for something I’d been watching (and doing) for years across every team I’ve been part of.

MAR 13, 2026

Abolish Golf

Golf courses occupy over 2.2 million acres of American land. That’s more than 3,500 square miles, larger than Delaware and Rhode Island combined. Globally, countries dedicate more land to golf courses than to solar and wind energy installations. We are in the middle of overlapping housing, climate, and public health crises, and we’ve decided that maintaining private lawns for a sport played primarily by wealthy people is a reasonable use of some of the most valuable land on the planet. It isn’t.

FEB 26, 2026

The New Open Source Is Specs, Not Code

This week, a Cloudflare engineering manager “rebuilt” 94% of the Next.js API surface in under a week. He used Claude across 800 AI coding sessions, spending roughly $1,100 in API tokens. The result is something resembling Next.js that benchmarks at 4x faster build times and 57% smaller bundles than Next.js itself. Almost every line of code was AI-generated. The human contribution was architectural direction: deciding what to build, steering the AI away from dead ends, making design decisions.

AUG 8, 2019

Innovation Outsiders

The vast majority of big pillar innovation work being done in corporate america is relegated to innovation outsiders – groups external to the core product delivery teams (Labs, Innovation Teams, etc.). They are expected to find a business partner or a sponsoring executive that can fund their work to incrementally transform existing products. This approach has been historically fraught, both at more traditional as well as companies renowned for their technical innovation. The difficulty is driven by myriad issues including difficulty aligning roadmaps, funding, awareness and accommodating technical risk. This post lays out a hypothesis that you must reverse the flow of innovation, have it driven by transformative products and let the technical innovation derive from that.

MAY 22, 2012

Course Complete!

So… it has been a while since my last update. This spring’s semester wrapped up well for both me, as well as the students in my Python for social scientists course.

JAN 22, 2012

Computation Frameworks for the Social Sciences (aka I’m teaching a class)

This spring I am teaching my first course! It is a pretty small seminar, 8-10 graduate students, but they all seem excited about the material. It is a “Programming for Political Scientists” course that will use Python to both teach people how to write good software as well as show them how people are using software in the discipline currently. I hope to spend the first half of the course covering basic software engineering and computer science concepts before moving on to some specific applications. Hopefully, by the end of the course all of the students will have built something that they can use to further their research agendas (e.g. a web scraper to supplement a data set).

DEC 30, 2011

Rook

As I said in my last post, I am working on setting up a statistical web service using R. I decided to use the Rook library to do so and wanted to give a brief overview of how Rook works for others who might be interested.

DEC 19, 2011

Rook and R Webservices

Recently I have been working on setting up a webservice that does some non trivial statistical work. Normally, my go to when building web services is Ruby/Rails due to ease of use, then I offload anything computationally intensive to something more optimized (e.g. a C or Java application on the same box). In this case however, partly because of my co-author’s skill, partly discipline norms, and a whole lot of R being awesome for this sort of thing, the statistical work is going to be done in R.

OCT 17, 2011

Exponential Growth and Resource Consumption

In August, while on the road moving to Duke, I heard a great interview on NPR with David Suzuki. During this interview he talked about the relationship between anything that grows exponentially and the resources that it consumes. In particular he focused on how we are likely to misperceive the remaining level of resources because humans are bad at understanding exponential growth (see anything by Kurzweil).

OCT 4, 2011

Getting into Git

This monday, I was asked by [Mike Ward(https://web.duke.edu/methods/)] to give a talk to his ICEWS group at Duke about the wonders of source control and how to get started using git. I was excited to do it because I am a big fan of git and use it on almost everything that I create myself (yes I use it on my CV…). While the opportunity to introduce it to others was great, I went into it a little bit of trepidation because in the past when I have talked about git or other DVC systems, my audience is at least sold on the idea of version control… because they are software developers. That was not the case for this audience, but I think I may have converted them nonetheless.

NOV 3, 2009

Bullet Plugin for Rails

Recently I have been working on improving the performance in one of my clients Rails applications, trying to get page load times down. Looking at the query log it was obvious that there was way too much activity going on for what was being rendered on the page–there were N+1 queries all over the place. This was nothing a few joins (or eager loads as the rails folks like to say) wouldn’t fix. So after changing the queries to make sure that all the necessary data was gathered in just one pass, I wondered how many other places this was cropping up in the application. That is when I stumbled upon flyerhzm‘s Bullet Gem.

APR 14, 2009

LAMP Development for Windows

I have been doing development in PHP for a long time and have yet to find a really satisfactory solution on Windows. When I recently flattened my home PC and had to restart I thought I would try a new setup and here is what I’m using. Its pretty good to me so far, and I think I’m going to run with it for the time being:

SEP 12, 2008

Building Grammars in .NET

As I’ve shown before (here, here, and here), the Microsoft .NET Speech API allows you to quickly and easily build applications that take advantage of the good folks at Microsoft Research‘s work on speech recognition. The general process is that you can construct a grammar that the engine will recognize for you, and then event handlers for those recognized events will be triggered. This process then, will only be as effective as the grammars that you can construct for the the SpeechRecognitionEngine. In this post I’ll show you some of the things that you can do to construct Grammars in C#.

JUL 6, 2008

Getting your C# app to run in the System Tray

While making some additions to the Speech Framework, I needed to figure out how to get my WPF application to run in the system tray. For the speech framework, the sys-tray is a natural place for the application to run, as we ideally don’t want to see any UI and the application should be always on. I opted out of a service, because I did want the rich UI of an application and for it to show up in task manager, etc. It was very straightforward, but I thought I’d write up the steps for future reference.

JUN 20, 2008

Javascript, Flash and cross-browser clipboard fun

Recently I have been wrestling with writing some code that will manipulate text on a user’s clipboard in a web browser. I thought I would share my findings here for others who might be interested in the same topic.

JUN 15, 2008

Miles per Gallon vs. Gallons per Mile

A new study by Richard Larrick of Duke’s Fuqua School of Business, recently published in Science magazine shows that most people have the wrong intuition about how to save gas. This is because they misunderstand what the “miles per gallon” statistic actually represents. As a result of this they may not be doing the optimal things to decrease their gas consumption.

JUN 15, 2008

Speech Plugins: Weather

In this entry we’ll look at building a basic plugin for a speech engine (Microsoft Speech III), that will respond to simple queries about the weather. For this tutorial we will be using the Yahoo Weather API. Note that if you are just interested in how to use their API, it should be easy enough to extract what you’re interested without doing the previous tutorials.

JUN 13, 2008

Microsoft Speech III: Setting up a pluggable infrastructure

This post will build on the work done in Microsoft Speech II, where we saw how to interact with our computer via voice. In this tutorial we will build a plugin infrastructure which we will use to add functionality to our Speech Project in the future. While we won’t be adding any new functionality in this tutorial, we will be refactoring what we had into a “Personality” plugin. Please note that these ideas aren’t uniquely mine and I borrow heavily from Matthew Cochran’s excellent tutorial on the subject of plugins – I just adapted them for the speech project’s purposes.

JUN 11, 2008

Microsoft Speech II: Getting your computer to listen

In this post I’ll show you how to take our simple project from the first example, Microsoft Speech I, and get it to understand input spoken into a microphone. Note that everything I’m showing can be found in the MSDN pages for the Speech API. Before we get started, make sure you have a microphone and it’s hooked up to your PC and all that other obvious stuff….

JUN 5, 2008

Microsoft Speech I: Getting your computer to talk

This is the first of what will hopefully be a series of posts that detail my explorations with the Microsoft Speech recognition and synthesis libraries built into Vista. I am working on automating certain parts of my home and (after seeing Iron Man), realized that I needed to see if I could get my house to talk with me. Now I’m not trying to write an app that I can chat with in such a way that it would pass a the Turing test, but I would love something that could respond to simple voice commands. I imagine sitting in my living room reading, discovering a word that I’m not sure the definition of, and yelling at my house to look it up for me. We’ll see how far I get.