Blog on Justin Abrahms

Scaling at the mediumest company: half-measures save the day

Mon, 09 Mar 2026 00:00:00 +0000

I’ve worked for globocorps like Amazon, eBay, Walmart and Google. I’ve worked for startups you’ve certainly never heard of.. but nothing in between. Thrive Market, an online grocer, is the mediumest company I’ve ever worked for. This company was my very first experience with “medium company problems”. At a startup, efficiency numbers largely don’t matter. At massive companies, all of the low hanging fruit has been picked. Medium company problems are when you’ve got enough scale for it to hurt, but not enough investment for anyone to have looked.

On MCPs vs CLIs

Sat, 07 Mar 2026 00:00:00 +0000

MCPs (Model Context Protocol servers) are services that expose tools for LLMs to use. CLIs (Command Line Interfaces) are shell utilities that programmers use to interact with a tool without using something like a website. Both of these tools are available to many agentic LLMs like Claude Code. Which should you pick?

My take on this is that you should nearly always prefer a CLI.

MCP#

There are three benefits to using an MCP: auth, updates, and preconfiguration.

CI Theater: we deleted our flaky tests and nothing broke

Thu, 26 Feb 2026 00:00:00 +0000

The problem#

At Thrive Market in 2024, we had ~150 flaky tests and a 76% deploy success rate. This meant that one in four deploys failed. Not because the code was wrong, but because the pipeline couldn’t be trusted.

A one line code review took two days to land. Engineers were watching CI fail over and over and retrying ad nauseam. Ephemeral PR environments took 30 minutes to spin up, sometimes requiring 4 retries before they were green. This applied to both the frontend and backend and folks were anguished. When asked “When was the last time these e2e tests caught an actual bug?” 80% of people answered “I know it’s happened, but I don’t remember when”.

Claws don't need to be complicated

Sat, 21 Feb 2026 00:00:00 +0000

Last weekend, I spent time setting up OpenClaw. Within 4 hours, I had hit $20 in usage (thank you, spend caps!). It was clear that if I wanted to experiment with this type of software, my clanker would need to run within the bounds of subscription pricing.

For the purpose of this post, I’ll refer to the software repo I’m using (epiphyte) as my “claw” and the instance of that software with a set of memory/skills/identity as my “clanker” (briar).

How to diff a book against your mind

Wed, 18 Feb 2026 00:00:00 +0000

I was looking into embeddings recently and realized you can embed your personal knowledge base and a book’s sections/chapters/pages, then use cosine similarity to estimate what you’re likely to learn. mindblown.gif

The result is the ability to diff a book against your current knowledge base. This lets you glance at a technical book and see whether you’ll learn something new, deepen your understanding of a topic, or if it’s simply review.

It works like this: The book text is extracted page by page and chunked into 100–800 word segments, then labeled with the nearest table of contents entry so an unread book is still easy to contextualize. Personal notes under 500 words embed as a single chunk while larger ones are split on headings. Embeddings run locally via sentence-transformers, and the tool compares each book chunk to my notes to classify it as novel, a depth gap, or review.

Announcing FRM, a CLI based CRM... 15 years in the making

Tue, 17 Feb 2026 00:00:00 +0000

When I lived in NYC, I frequented hack && tell, a hackers’ meetup where folks showed off side projects and generally were nerds with one another. During that time, I was trying to better understand how to show up for my friends and community. I came up with what some might call a technical solution to a social problem, but, after 15 years of reflection, I’d say it’s a bit closer to a cognitive prosthetic: FRM, the CRM for Friends.

How to generate demo videos with playwright

Thu, 12 Feb 2026 00:00:00 +0000

I came across this post on social media about how a user was using OpenClaw to generate tiktok content for their iOS app automatically. This made me realize that we could be using LLMs to generate product demo videos of the features we work on.

I worked with Claude in order to generate on for a flagd UI I was playing with, slapped some royalty free music on it and here was the result.

Recommended reviewers to build your council

Fri, 06 Feb 2026 00:00:00 +0000

There’s a common pattern of building up a council of reviewers and have them debate a topic. The first ones of these I saw was the Council of HAL. Following that, I’ve seen similar things like with open-code-review, which will bring up a panel of experts and have them debate, then synthesize the feedback and write it into a PR.

In doing some design doc review, I stumbled on a pattern I think is really interesting: asking the robot for a recommendation of other colleagues who might have good insights.

AI Doesn't Work Without Good Inputs (And Neither Do Humans)

Tue, 27 Jan 2026 00:00:00 +0000

AI skeptics complain about code which is verbose, overwrought, unmaintainable, and misguided. In many specific instances, I’ve found that to be correct. The problem to be solved isn’t higher ownership, more diligent human review or the perfect CLAUDE.md file. We need better control over the inputs if we want to tighten the band of acceptable outputs.

The core problem with LLMs going off the rails is ambiguity and complexity. LLMs have better luck solving small, well defined problems when they have adequate context. A lot of working with LLMs today is about scoping down those problems or providing them that corrected context. These are not uniquely LLM problems. Humans also do much better with well defined problems, clear outcomes and good context, and I think that’s a much more useful frame.

agent-chat: a way to monitor multi-agent orchestration

Sat, 24 Jan 2026 00:00:00 +0000

When running multi-agent orchestration like gastown or multiclaude, there’s a lot of moving pieces. It can be noisy to determine what’s really going on, especially as it’s happening across several tmux sessions. To address this, I’ve released agent-chat. It shows communication between the agents as though you were in a chat room with them. One chat-room per repository-equivalent.

In the screenshot, you can see a supervisor coordinating workers to process a PR. When the spec requirement failed, it spawned a new worker to fix the issue - all visible in real-time.

multiclaude, a different take on llm orchestration

Tue, 20 Jan 2026 00:00:00 +0000

Dan Lorenc recently started talking about his LLM orchestration experiment dubbed multiclaude . I’ve used it a bit and thought others who were looking at it might want to know how it’s different.

How it feels#

Multiclaude is similar to Gastown in that it’s a distributed go binary that will help you launch and manage a pile of claude code sessions in tmux. Its vibes are much more “unix tool” and not “I’m a mad mage peering through my orb”.

A Tuesday in 2027: What orchestrated LLMs might look like

Fri, 09 Jan 2026 00:00:00 +0000

What follows is from an LLM session w/ Claude Code, where I was discussing some of the tensions in the current world. This accurately represents where I think things are heading. Is this a perfect analogy? Absolutely not. But it’s directionally interesting and indicative of the levels of change I think we’re looking at in the next 1-3 years.

A Tuesday in 2027

You open your morning dashboard. Not a PR queue - an outcomes board. Overnight, 47 changes shipped across three services. You don’t see diffs; you see deltas in error rates and latency percentiles. Two anomalies flagged. You don’t ask “what changed” - you ask “does it matter?”

Yegge's Developer-Agent Evolution Model

Thu, 08 Jan 2026 00:00:00 +0000

In Welcome to Gas Town , Steve Yegge generated a list of the 8 stages of agentic workflow evolution. I find myself wanting to deep link to it, so I’m replicating it here.

Stage 1: Zero or Near-Zero AI: maybe code completions, sometimes ask Chat questions
Stage 2: Coding agent in IDE, permissions turned on. A narrow coding agent in a sidebar asks your permission to run tools.

Wrapping my head around Gas Town

Mon, 05 Jan 2026 00:00:00 +0000

I read Steve Yegge’s post announcing Gas Town, an LLM orchestrator which allows you to manage dozens of Claude Code instances at once while they make independent progress towards some stated goals. It was a wild ride, but I believe I see the promise.

Context#

Two pieces of context seem useful before diving in. First, I worked with Steve briefly at Google. I remember him working on “stone tools”, which was bringing the features of an IDE to old editors like Emacs and Vim. This was 3 years before LSPs happened. This is to say, I think that Steve has some degree of future sight. I think that’s in play here. His Kool-aid is good. I’ve drunk it before and likely will again.

LLM Session Sharer: Easily share a claude code session without clutter

Sun, 04 Jan 2026 00:00:00 +0000

Trying to learn how to use LLMs is a pain. So many of the day-to-day invocations are hard to pick up. Pairing with people isn’t very scalable. To that end, I saw a slack message by David Edmiston about a tool he wrote:

Trivial, but in OpenCode, just one-shotted (and then tweaked the result a bit) a single page utility that can take a Claude conversation (JSON file exported from the web version via the browser dev tools) via drag-and-drop or file picker dialog and generate a nice HTML view of it, with an option to export the conversation as a markdown file (suitable for github gists) or copy to clipboard as markdown for import into obsidian (the real reason I wanted it, i sometimes rubberduck things with claude and wanted to capture it to chew over later)

Using your (Backstage) service catalog for compliance

Thu, 01 Jan 2026 00:00:00 +0000

At work, I rolled out backstage as a service catalog offering in 2025 across a few hundred git repos. It provided a pathway to get rid of a spreadsheet hand-maintained by a developer about what each repo or service was, who owned it, what it depended on, etc. When our regular compliance reviews came around, it made me think: Can this help us here too? It turns out it could.

Consul leader election issues

Thu, 20 Jul 2023 00:00:00 +0000

Problem: The cluster is in a broken state because consul can’t seem to gather a quorum w/ it’s raft implementation.

In my case, there was a raft peer that was bogus. I accidentally had it advertising it’s IP as 127.0.0.1 , but there was no process who had that node-id at that address.

There are two possible paths out that I know of. You can put a peers.json file in the consul data directory. Or you can manually bring up the consul process with the -bootstrap flag, to allow it to self-elect into a leader. The peers.json file approach worked for me.

A risk-based approach to open source strategy

Wed, 19 Apr 2023 00:00:00 +0000

On it’s face, using open source code is an inherently risky endeavor. We are trusting external developers to write code that we’ll eventually deliver to our users, and we have no recourse if the code is buggy or malicious in some way. In practice, people are generally good and the forces that govern popularity of open source projects help reduce this risk to the point where nearly all folks in the industry are prolific consumers of open source code.

SREcon 2023 trip report

Mon, 27 Mar 2023 00:00:00 +0000

Last week, I attended SREcon 2023. I was invited to attend the “next-gen delivery” working group meetings, due to my involvement within the Continuous Delivery Foundation (CDF). I used the opportunity to attend the rest of the conference as well. While I’ve never been an SRE by title, I’ve toed the line between operations and software development for many years.

Conference makeup#

Surprisingly, the makeup of the conference attendees were different than I expected. It appeared that many of the folks were managerial. Talking with an organizer, they believed this was due to the budget cuts happening across our industry. They believed managers were attending to determine if it was a high-value conference to send their reports to. For similar budgetary reasons, large tech companies were underrepresented. The only participants from many (Google, Dropbox, etc) were those speaking. Ultimately, this made the conference feel less mature than I suspect previous years might have felt. Instead of these companies, financial services and healthcare were well represented.

Feature Flags, Dynamic Config and Experimentation (oh my!)

Tue, 15 Nov 2022 00:00:00 +0000

As part of the OpenFeature project, I’ve been thinking a bunch about feature flags. There’s been ambiguity about how feature flags differ from dynamic config and if that’s the same thing as experimentation (e.g. A * B testing).

At their core, feature flags are really fancy if statements. Those if statements control the behavior we want to manage for our application.

// example OpenFeature java code
if (client.getBoolean("should-show-redesign", /* default */ false)) {
 // ...
} else {
 // ...
}

That may be backed by a file or it may be backed by some giant distributed system. That’s a serialization concern that’s separate from the use of “gimmie this config value”. This brings up the question about how this relates to config in general. Should we use something like to fetch environment variables? system properties?

Understanding Etsy's search service migration

Tue, 15 Nov 2022 00:00:00 +0000

Etsy published a recent article which documented how they migrated from their old search service to one which was based on deep learning. The majority of that article made little sense to me, so I wanted to take a moment to document what the heck that actually meant.

Originally, Etsy’s search service was based on a gradient-boosted decision tree model, which drove personalized search. They had to manually engineer the features of the tree model and their relevancy gains began to plateau. Deep learning models offered several exciting potential improvements, especially because they could embed features directly, incorporate multi-modal data and alter the network architecture itself. The search team ran two passes of ranking and their work was primarily targeted at the second layer.

Git pre-receive hooks for deployment

Wed, 09 Nov 2022 00:00:00 +0000

I host this blog on a git repo that lives on the same box as the webserver. When I push to it, I want to ensure that deploys happen. Previously, it involved SSHing and doing a bit of a dance. Today, I setup a deploy-on-push script through git’s pre-receive hooks. It was surprisingly difficult to find a good example of this pattern, so I wanted to publish my result.

When the pre-receive hook is invoked, it’s given a line on stdin for each ref that was pushed. If the branch is master, then invoke the make publish command in the repo.

re: Zettelkasten concerns

Mon, 12 Sep 2022 00:00:00 +0000

I recently read the article Zettelkasten/Org-roam/Org-brain Is Crap . After sending this response to the author, they asked for a place to link it.

There were a few misconceptions to clear up and a few tips I thought I’d share. For context, I went from a very large eng-log.org file to multiple files (~210) managed through org-roam at the end of 2021. I probably don’t follow zk precisely, but something in that vein. My work is not based in research, but as an architect at a large company.

An attempt at defining an ideal pipeline

Tue, 16 Aug 2022 00:00:00 +0000

The Continuous Delivery Foundation is currently looking to build out a reference architecture, which I think is a fantastic idea. While there are a bunch of social things that need to be figured out to really “get” CI/CD, the [[https://bestpractices.cd.foundation/][Best Practices SIG]] is working to get those well documented. I thought it might be helpful for me to document what my ideal pipeline is.

When a developer submits their pull request, automatic validation begins. We validate:

My thoughts on crypto: 2022

Thu, 27 Jan 2022 00:00:00 +0000

It’s 2022. 1 bitcoin is about $35.5k USD. An ethereum is $2.3kUSD. I hold 0.2ETH, which I bought because I really like Gary Vaynerchuk, and I was interested in his NFT project VeeFriends. Ultimately, those NFTs were too expensive for me to feel comfortable participating, primarily around gas fees.

I first heard about bitcoin in late 2010 or early 2011 at hack&&tell . A finance person was telling me about it and it didn’t make a lot of sense to me. A BTC was worth about $6, but he wouldn’t sell me his.

the function '..' is not known to be defined

Fri, 13 Aug 2021 00:00:00 +0000

I was working on ox-gemini, a gemini exporter for org-mode documents. In the latest versions of emacs, string-replace is a function which replaces a word in a string. It doesn’t exist in emacs 27.2 and earlier, so I added a simple shim.

;; backport for older emacs verisons
(if (not (fboundp 'string-replace))
 (defun string-replace (from to in)
 (replace-regexp-in-string (regexp-quote from) to in nil 'literal)))

The shim says “If you don’t see a string-replace function, use this one). Unfortunately, this left me with this compilation error:

Governance versus Stewardship

Mon, 09 Aug 2021 00:00:00 +0000

In the enterprises I’ve recently been a part of, there’s been lots of discussion about “governance”. Governance is a process where you ask other people for permission to do things. The governing body serves to enforce consistency in things like API design or to ensure adequate testing on deploys to production.

One of the biggest problems with governance is that the beaurecratic process can be quite slow. Relying on an outside authority to manually approve your deploys to production introduces latency which is likely to cause your deploys to accumulate. The subsequent deploys are larger, which means that both their blast radius is bigger and, when things do go wrong, it’s difficult to isolate which change in the batch caused the issue. In this paradigm, introducing this strong-handed governance is actually counter to their own goals.

Federated GraphQL Ops with Apollo Studio

Wed, 28 Jul 2021 00:00:00 +0000

There are several differences between the traditional REST API model and how federated GraphQL operates, which can cause some friction when discussing the tech with your SRE/ops organization. This is a list of issues that I’ve come across in my work and the mitigations we’ve either considered or adopted. This list assumes that you either have access to Apollo Studio or have replicated the relevant functionality to your custom solution.

Learning from Production Incidents

Tue, 15 Dec 2020 00:00:00 +0000

Note: This was originally posted internally at Walmart, and has since been sanitized for public consumption.

The postmortem process is a tool that we use to better understand failures within our systems. There are two ways to view failures within complex systems: “That failure cost us $250,000” or “The company spent $ 250,000 to learn this lesson”. Taking the second approach, this document aims to outline a process which wrings as much value from that lesson as possible.

First and Second order metrics

Sat, 31 Oct 2020 00:00:00 +0000

As part of our efforts of continual improvement, we have been diving deep on the metrics that we want to track at an organizational level. This has been happening at all levels of the organization, which has resulted in multiple contemporary conversations around which metrics we should track. In these discussions with engineers and leaders, we’ve lacked some wording to classify the metrics we were discussing. I believe that thinking about metrics as either first-order metrics or second-order metrics is a useful distinction.

Biogas Digestion system.

Thu, 04 Jun 2020 00:00:00 +0000

A biogas digestion system is a way to convert biological waste (like kitchen scraps, grass clippings, and manure) into usable biproducts of methane and both liquid and solid fertilizers.

Biogas digestion works by creating an anerobic chamber (i.e. a place without oxygen) which breaks down the organic matter you put in it. There are two types of bacteria at play here. One type takes food and converts it into acid. These are called acidogenic bacteria (aka acetogens). Methanogenic bacteria (methanogens) convert the acid from the first system into methane gas.

Is web scraping ethical?

Thu, 04 Jun 2020 00:00:00 +0000

Someone recently asked: “Is web scraping an ethical concept?” I believe that web scraping is absolutely an ethical concept. Web scraping (or screen scraping) is a mechanism to have a computer read a website. There is absolutely no technical difference between an automated computer viewing a website and a human-driven computer viewing a website. Furthermore, if done correctly, scraping can provide many benefits to all involved.

There are a bunch of great uses for web scraping. First, services like Instapaper, which allow saving content for reading on the go, use screen scraping to save a copy of the website to your phone. Second, services like Mint.com, an app which tells you where and how you are spending your money, uses screen scraping to access your bank’s website (all with your permission). This is useful because banks do not provide many ways for programmers to access your financial data, even if you want them to. By getting access to your data, programmers can provide really interesting visualizations and insight into your spending habits, which can help you save money.

How to make a font on Ubuntu with PPAs

Thu, 19 Sep 2019 00:00:00 +0000

Microsoft released a new font called Cascadia Code. I’d prefer if I could install this with the debian package manager, so I went to figure out how to put that on Ubuntu’s PPAs.

# Install dependencies for the ppa creation process
sudo apt install -y dh-make bzr-builddeb

# Tell bzr who I am
bzr whoami "Justin Abrahms <justin@abrah.ms>"

# Download the font
wget https://github.com/microsoft/cascadia-code/releases/download/v1909.16/Cascadia.ttf

# Create a tarball
tar cf cascadia.tar ./Cascadia.ttf

# Try to build the actual package
bzr dh-make ttf-cascadia-code 1909.16 cascadia.tar

From here, we have a debian/ folder that contains lots of files we need to configure. I deleted the two README files that were generated and filled out the copyright and control files. It’s important that the name-email combination you put for yourselve in these matches the name-email combination used in your GPG key.

Whiteboard photo snapper

Sun, 10 Feb 2019 00:00:00 +0000

Six people live in my house. We don’t have strong technology overlap (some folks in the house only use simple phones and others aren’t keen on things like slack or similar). We’ve addressed this by centralizing around a large whiteboard in our kitchen space. This holds things like grocery lists, house announcements, and similar. I often do grocery shopping on my way home, but rarely remember to snap a photo of the list before leaving for work.

Serene Lake

Tue, 17 Apr 2018 00:00:00 +0000

Serene lake is a backpacking trip that I’ve done a few times. It features wonderful Oregon moss-y forests, a variety mushrooms, huckleberries, and some wonderful lakes. If you’re lucky, you can also catch some mountain views from ontop of the hike’s one ridgeline.

The hike is lollipop shaped from the entrance and clocks in at around 14 miles. It starts at Shell Rock lake. There are often car campers here, given it’s proximity to the parking lot (about a mile or less). From there, you take the trail onward to a fork in the road. The first year, we went right, which is the direction I’d suggest. It takes you to Frazier Turnaround, which is accessible by car some parts of the year. Commenters saying that it’s blocked by boulders and such has been enough to add the extra 2-3 mile hike for the Shell Rock start.

Software engineering vs other engineering

Tue, 17 Apr 2018 00:00:00 +0000

Quality Assurance (or QA) is a mechanism by which teams confirm that they’re hitting the requirements they’ve said they’d hit. One example is ensuring you’re validating your contractual obligations from a consulting standpoint. Another is making sure you meet the end user’s expectations.

A friend who does QA in the nuclear industry recently had some issues with an outside software vendor. They caused some political strife between the government and a native tribe because they didn’t follow the contractual processes. He wrote asking me if there were any similar standards to the “American Society of Mechanical Engineers QA Requirements for Nuclear Facilities Applications”, which outlines how teams build software in his industry. I didn’t have good news for him.

Stub Stewart Bike Ride

Tue, 17 Apr 2018 00:00:00 +0000

I do a yearly bike ride out to LL Stub Stewart State Park, between Banks, OR and Venonia, OR. It’s a fantastic intro to bike camping trip that’s a known quantity. The full ride is around 48 miles and requires minimal meal preparation given it’s access to food and grocery stores.

The guidance I give to my friends is: “If you bicycle around town to get around, you can do the trip no problem. If you aren’t much of a cycler but are in reasonable shape, you can do it, but your saddle will be sore. If you’re not in shape and don’t bicycle, you should strongly consider whether this is where you start cycling. You’ll probably fall behind and be tired. While the group won’t leave you, people tend to feel bad to be the last one in the line.”

Understanding the formal definition of Big-O

Tue, 17 Apr 2018 00:00:00 +0000

This is the third in a three post series. The first post explains Big-O from a self-taught programmer’s perspective . The second post talks about how to calculate Big-O .

In my original Big-O piece, I’ve made a bunch of simplifying assumptions and told you a bunch of half truths. It might seem a bit disingenuous, but really.. these half-truths are helpful because (unless you’re going after a formal CS degree) they don’t really matter.

What is an algorithm?

Tue, 17 Apr 2018 00:00:00 +0000

One of the early classes in computer science curriculum is “data structures and algorithms”. When you read computer science text books, you’ll see these words paired together. This makes the word sound more fancy and important than it should. An algorithm, while important, is not particularly fancy.

Algorithms, simply put, are how you go about doing something.

When paired with data structures, “algorithm” means how you do something using a particular data structure. One example of this is how do you put an item into a list? Well, that depends a bit on what the underlying data structure to the list is. Maybe you’re moving some pointers around. Maybe you’re doing a big shuffle of blocks of memory. The exact machinations of this are your “algorithm”.

Will binary always be the lowest level way to program?

Tue, 17 Apr 2018 00:00:00 +0000

@justinabrahms serious question: will binary always be the lowest level computer language? Is there (conceivably) a different model?
— Justin Jackson (@mijustin) December 15, 2015

Justin Jackson asked if binary will always be the lowest level language, or if that might ever change. To answer that, let’s talk about what low level means.

Binary isn’t exactly the lowest level programming language though. Binary isn’t actually a programming language at all. Binary is simply a way to convey information. In some cases, this information is the layout of logic circuits. In others, it might be media like a YouTube video or git checkout streaming to your disk.

An overview of Guice

Tue, 23 Aug 2016 00:00:00 +0000

Google’s Guice is a Java-based dependency injection framework, which means approximately nothing to people who aren’t familiar with it.

Google’s Guice is a way to build a graph of dependencies so you can instantiate complex objects made of simpler parts.

Google’s Guice is a replacement for the new keyword for many of your objects. Bizzarre, right?

In a more tangible sense, Guice is a tool that’s used in a lot of Java development. Beyond that, it positively impacts your code resulting in a reduction of boilerplate and an increase in testability.

Literate API Testing with Dredd

Thu, 12 Nov 2015 00:00:00 +0000

I’ve just started my latest project at MIT, and I’m quite excited about a new piece of technology that the team uses: apiary.io.

With Apiary, we generate a simple blueprint file (source, output) which is a full description of your API. This shows off requests and responses with a bit of prose. It’s all in markdown file and is very simple to get started. There is a tool called Protagonist which will parse this into a big JSON blob.

Altruism in the job hunt

Wed, 11 Nov 2015 00:00:00 +0000

I’ve recently accepted a job at MIT’s Office of Digital Learning which I’m quite excited about. The bulk of the work is done in open source on GitHub. We’re working on a stack I’m familiar with (Python/Django) but are also actively using newer technology I’m interested in (React/Docker).

Technology choice wasn’t the actual reason I chose this job though. I’ve been gaining a bunch of experience over the past few years in product and engineering. First at Google, where I first started to understand what product was about, then later at Sprint.ly where I got to see first-hand how it’s done by Joe Stump . The next stop on my career path would have naturally been somewhere leading an engineering team at a venture-backed startup. I certainly have enough industry connections to make that a reality (and I’m thankful for that!).

MicroConf 2015 Recap

Mon, 20 Apr 2015 00:00:00 +0000

I’ve just returned from MicroConf 2015. MicroConf, if you’re unfamiliar, is a conference for self-backed startup entrepeneurs. Most of the talks center around how to grow out a small business. The audience is split roughly evenly between consultants and product.

There were two stand-out talks for me this year. One by Patrick McKenzie and the other by Steli Efti.

The first talk of the conference was by Patrick McKenzie (or patio11 as he’s known around the internet). He talked about his original product, Bingo Card Creator (which does exactly what it sounds like), and how it’s audience was different than his next product, Appointment Reminder (which calls people to remind them of their appointment). He wasn’t able to leverage his existing audience of teachers. In terms that I use, he wasn’t able to use the assets he had (a wide network of teachers) as a superpower in his latest endeavor. Patrick also had my favorite quote of the conference.

Real-time code coverage analysis

Thu, 09 Apr 2015 00:00:00 +0000

I recently sat down to write a node service and happened upon a very helpful workflow when trying to write tests that I wanted to share.

When writing tests, one of the biggest jobs is to ensure you have a reasonable degree of coverage on the various branches in your code. This means that if you have an if statement, you want to check the case where the if is true and the case where the if is false. This gets a little more complicated if you have multiple statements in your if statement like if (foo || bar) {, but the idea is similar.

Imhotep helps you enforce your styleguide during code review

Sat, 23 Aug 2014 00:00:00 +0000

Over the holiday break, I got a long-standing side project into a working state. Imhotep is a tool that will enforce coding standards on pull requests and commits. It accomplishes this by running code linters and commenting on your GitHub pull requests with any violations.

The main motivation for imhotep is based on a trend I saw at Google. We did pre-merge code reviews and I was on a team of approximately 12 people. It became apparent that different folks gave different sorts of code reviews. There were some people who would validate syntax, code style enforcement and rubber stamp your change. These were the tactical reviewers. There were others who would identify problematic aspects of your overall design and, as a result, would teach you a great deal about how to build robust software. These were the strategic reviewers.

Inheritence versus Composition

Sat, 23 Aug 2014 00:00:00 +0000

Part of a healthy engineering culture is having a good code review process. There are many things that contribute to a [[https://github.com/sprintly/sprint.ly-culture/blob/master/Coding.mkd#code-review-as-knowledge-sharing-and-buy-in][good code review process]] , but one of those is thinking through why something should be the way it should. I recently did more writing than is usual which I’ve adapted to use as this post.

The code under review#

We start with some code used to render columns of tickets in Sprint.ly’s UI. Depending on where they are in the UI, these columns are sortable, you can drag & drop between them, or you can adjust which properties they are ordered by. The code looked something like this:

Meaning of Community

Sat, 23 Aug 2014 00:00:00 +0000

My wife attended Cascadia JS 2013 and I tagged along to join the
significant other track, which was really just an excuse to visit
Vancouver for a weekend. I had a great deal of fun getting there on
the Hacker Train, which was a merry band of JavaScripters who rented
out a train car. Along the way, I had many conversations that went
something like this.: js: “Are you excited for the conference?”

Setuptools Entrypoints Explained

Sat, 23 Aug 2014 00:00:00 +0000

Python’s setuptools has a rather interesting feature called “entry points” which facilitates plugin discovery. I’m using this in imhotep to allow linters to live as separate packages.

So how does it work?#

In your setup.py, you defined a property called entry_points which is a dictionary mapping the entry point key to a list of values. The values take the form of ~~.py = imhotep_pylint.plugin:PyLint~~. This is another mapping in string form from a user-defined string to a dotted module name (and optionally, as here, an object to import).

Thinking like a programmer: Inputs & outputs

Sat, 23 Aug 2014 00:00:00 +0000

One of the difficulties in learning to program is not the specific syntax it takes to accomplish a goal, but the thinking that went behind it. As a coworker of mine once said, “I understand what you did, but I wouldn’t have thought of doing it that way.” One method of thinking about a problem is by considering the inputs and outputs of a program or function.

Let’s talk about inputs and outputs from a function. Functions are a good example, but it could apply to modules or distributed system components too.

Why a community might want a Python Software Foundation

Sat, 23 Aug 2014 00:00:00 +0000

There have been a few discussions around the node.js community around why someone might want a foundation similar to the ones created by the Python or Django communities.

I’ll start by noting that the bulk of my involvement with these entities has been indirect at best. I am a member of the Python community, but haven’t had many direct involvements with these organizations. That said, I’ve seen the benefits of their hard work.

NodeConf 2014: A story of aspiration and community

Wed, 04 Jun 2014 00:00:00 +0000

This year, I helped mentor NodeConf 2014, which was an amazing experience. For those not familiar, NodeConf is a much different format than the average industry conference. First, instead of a standard conference center, the venue for NodeConf is an idyllic middle schooler’s summer camp in sunny California. There’s a museum of taxidermied animals, foxes and deer wandering about, and a pond for swimming adventures. Evening activities ranged from heated games of 4 square (or absurd variants like 25 square) to s’mores, from LED firework displays to late night hacking.

Pull request templates make code review easier

Wed, 04 Jun 2014 00:00:00 +0000

We do a bunch of code reviews at Sprint.ly. One of the things we’ve done to make that process a little easier is institute a code review template. This template is a free-form list of questions that every person filling out a pull request uses.

Our pull request template looks like this:

#### What's this PR do?
#### Where should the reviewer start?
#### How should this be manually tested?
#### Any background context you want to provide?
#### What are the relevant tickets?
#### Screenshots (if appropriate)
#### Questions:
- Is there a blog post?
- Does the knowledge base need an update?
- Does this add new (Python) dependencies which need to be added to chef?

Some sections get a “n/a”, such as the screenshot section for API only changes. In general though, this has proven to make reviewing a much easier process. Curious what it looks like in practice? Here’s an example from a recent pull request.

PyCon 2014 was awesome

Wed, 04 Jun 2014 00:00:00 +0000

PyCon was a ton of fun this year. Always the highlight of my conference season, 2014 was no exception.

I gave a talk this year on Big-O notation. I told the story about how not knowing it led to a few costly mistakes and then explained it so that (hopefully) no one will make the same mistake again. You can watch the video below. Thanks to everyone who came up to me after my presentation to talk about it!

Node Knockout: A chance to play & experiment

Sun, 10 Nov 2013 00:00:00 +0000

This weekend, I competed in Node Knockout, a 48-hour coding competition designed around Node.js. For my idea, I chose to write a visualization of an AST querying language called cssauron. The entry has since been taken offline.

Overview#

cssauron is a tool which maps the ideas of querying for DOM elements through CSS selectors (eg how jQuery operates), and mapping that to source code. cssauron itself is AST agnostic and there are plugins for a few options that map the concept of selectors to the relevant AST nodes. The implementation I used was cssauron-falafel which binds with the falafel library for querying javascript source.

Obseletion through Programming

Sun, 10 Nov 2013 00:00:00 +0000

Working in tech, programmers are often brought in to apply automation to processes as a means to increase productivity. This ranges from automating the creation of manufactured goods or speedily compiling reports based on Excel spreadsheets. As a younger programmer, I found myself wondering what the person who used to do this work would do with their new found time. Simply put, is my hard work getting someone else fired? Within the context of knowledge workers, I think the answer is no.

Book Review: Backbone.js Testing by Ryan Roemer

Sun, 03 Nov 2013 00:00:00 +0000

Ryan Roemer, the author of Backbone.js Testing, wrote a post to the Seattle-area JavaScript developer’s mailing list to ask if anyone was interested in reviewing his book. As I have been diving into front-end testing at Sprint.ly, I agreed to review it. I was interested in how our setup compared to the setup he was advocating.

The book#

The book itself talks a bit about the generalities of testing: how to test, what to test, and what makes a good test. It introduces a simple app which is centered around taking notes in markdown and rendering them as HTML. It is similar to the TodoMVC app, but is a bit simpler and more focused.

Selenium's Page Object Pattern: The Key to Maintainable Tests

Sun, 03 Nov 2013 00:00:00 +0000

At Sprint.ly, we’ve been investing in the longer term health of our code base. Part of our strategy for handling this involves selenium testing.

For those who may be unfamiliar, selenium tests operate by instructing a browser to click on things or type in text boxes in an order you specify. The biggest upside to this is it’s the most accurate simulation of an actual user’s experience. One of the biggest obstacles of these sorts of tests is they are brittle. To combat this, we’re using a concept called the page object pattern.

Javascript Refactoring without an IDE

Wed, 16 Oct 2013 00:00:00 +0000

There are a lot of questions that pop up when you’re changing an entry in your style guide. If you’re editing a file which hasn’t been updated, is this pull request the one to update it in? Is that a separate PR? Should I just keep with the old style for now, but use it for new files? Without a clear cross-over point, you may end up with multiple competing standards, all of which are a snapshot of what the styleguide looked like when the code was written.

A list of scary topics.

Mon, 14 Oct 2013 00:00:00 +0000

Computer Science / Engineering#

Binary Protocols
Security
Sockets Packets low-level networking
Systems Coding (aka linux syscall apis)
Compilers
Assembly
3D Graphics Programming
Animation
Binary Compatibility
File systems
Writing a virus
C++

Mathematics#

All of it past arithmetic.

Big-O is easy to calculate, if you know how

Mon, 14 Oct 2013 00:00:00 +0000

Calculating Big-O#

This is the second in a three post series. The first post explains Big-O from a self-taught programmer’s perspective . The third article talks about understanding the formal definition of Big-O .

So now that we know what Big-O is, how do we calculate the Big-O classification of a given function? It’s just as easy as following along with your code and counting along the way.

def count_ones(a_list):
 total = 0
 for element in a_list:
 if element == 1:
 total += 1
 return total

The above code is a classic example of an O(n) function. This is because we have to loop over every element that we get in order to complete our calculation. We can trace this by following along the code and counting. We’re doing a few calculations here.

Pricing books as a function of risk

Mon, 14 Oct 2013 00:00:00 +0000

I’ve started the motions for a book that teaches computer science information to people who didn’t go through a university program. This is pretty common in the field of programmers. It’s one of the few knowledge-based fields where practical experience still competes against university degrees.

Where do you publish a book like this?#

I came across leanpub, which is a really interesting ebook publisher and marketplace. You link your Dropbox account to their service and edit Markdown files in a folder they share with you. They take this and build a book for you in multiple formats.

The State of GitHub's Code Review

Mon, 14 Oct 2013 00:00:00 +0000

In a word: Stagnant#

Code review is an essential part of a team’s shipping process. It helps disseminate knowledge across teams, catches bugs (or more likely poor architectural decisions) before they bite you, and provide opportunities to educate junior engineers. GitHub, the current de facto standard for this, is letting us down.

Perhaps they’ve hit some sort of local maximum for this activity, or maybe they’re focused on providing GitHub functionality for verticals like GIS or 3D modelers, but not much has changed in the world of reviewing code in many years. Herein is my 5 item wish list for code review.

Thinking of Databases as an Excel file

Mon, 14 Oct 2013 00:00:00 +0000

I was having a conversation recently around the value of databases as opposed to flat files and the question “What is the essence of a database?” came up. I explained it as it relates to Excel files, which I think is surprisingly apt.

Excel files are logically constructed around a few principles. First, there is the file itself. This lives on a hard drive somewhere, it has a name, and there is a program that can open it. When you open it up, things are organized on multiple sheets. These sheets contain columns and rows which contain data.

Big-O notation explained by a self-taught programmer

Tue, 23 Jul 2013 00:00:00 +0000

This is the first in a three post series. The second post talks about how to calculate Big-O . The third article talks about understanding the formal definition of Big-O .

Big-O notation used to be a really scary concept for me. I thought this is how “real” programmers talked about their code. It was all the more scary because the academic descriptions (such as Wikipedia) made very little sense to me. This is frustrating because the underlying concepts aren’t actually that hard.

Finding Unused files with Require.js imports

Wed, 17 Jul 2013 00:00:00 +0000

In an effort to clean up some code, I wanted to find a list of files not imported by my app. I didn’t see anything in a brief search, so I wrote this little script. Hopefully its helpful. It finds all javascript files in directories you specify as parameters to the script, greps your git repository for references to those files and echos those that should be removed.

This script does have the possibility for false negatives (ie: things that aren’t used that we don’t mention) because we use basenames. Examples might be a foo/utils.js which is used hiding the bar/utils.js which isn’t used. You may need to run this multiple times as deleting some files will uncover new unused files.

PyATOM for static site syndication

Wed, 17 Jul 2013 00:00:00 +0000

Bitten by the bug to write more and for others, I’ve brought back syndication (via Atom rather than RSS) to the blog. I found a few hundred subscribers still checking in on me, so thanks to you all for sticking around. As I run a static site, generating RSS can be a bit of a pain.

After a brief search of not finding anything for generating RSS or Atom for static blogs, I started writing something in Go which would take a YAML file and dump it into an ATOM format. A bit demoralized with the state of YAML parsers (nothing seemingly working), I swapped over to python. I was about to write a naive Atom dumper with a small templating language, when I came across PyAtom. Thanks to Python’s readable syntax, it makes for a fantastic feed writing format. As an example, this is the atom feed of this blog with the last entry populated.

PyATOM for static site syndication

Wed, 17 Jul 2013 00:00:00 +0000

setup.py, distutils and testing.

Wed, 17 Jul 2013 00:00:00 +0000

In releasing the web framework I’ve been working on, I went about providing the things any upstanding python project does: a CI configuration file, a setup.py and some tests.

The setup.py file was simple enough. PyCharm wrote one automatically. I was surprised to learn that when using distutils defining dependencies in your setup.py doesn’t install them with easy_install or pip. Its just a bit of metadata that is sent upstream.

The other gotcha is that I wanted single-command access to running my tests. The easier tests are to run and write, the more likely they are to be run and written. From setuptools land, a simple ~~python setup.py testand asetup.py~~ entry was all it took to get started. Unfortuantely, there is no specific information about setting this up in either the distutils2 documentation or in the packaging guide for Python. When attempting to cargo-cult my setup.py together, I stumbled upon simplejson’s setup.py and came across a custom test runner. You add a cmdclass keyword argument to your setup.py with a dict value mapping the command to the correct distutils.core.Command object to run.

Why Pisces?

Wed, 17 Jul 2013 00:00:00 +0000

Pisces is a Python web framework that was written with testing in mind. It was birthed as a reaction to the typical workflow in a Django project, which I’ve come to dislike as it encourages you to make your code tightly coupled and less testable.

What’s wrong with Django?#

The majority of my issues with Django can be summed up with one code block pulled from the Sprint.ly code base which is indicative of how I’ve seen things work elsewhere.

Securing your email

Wed, 03 Jul 2013 00:00:00 +0000

With the recent news surrounding privacy implications here in the US, I’ve taken it as an opportunity to encrypt and sign more emails that I send.

Motivations#

The motivations I have behind sending encrypted email aren’t to hide secrets, as I don’t really have anything to hide from the government. That said, I dislike the act of surveillence without cause in itself. I believe the simple act of passively recording all information is an inherintly immoral one and is easily corruptable. Couple this with recent statements like:

Meaning is not zero-sum

Thu, 30 May 2013 00:00:00 +0000

Rick Turoczy wrote an article about the role of meaning in the Work-Life balance equation. Rick suggests that finding meaning in your life (in his words “an occupation that fulfills you”) is a substitute for a work-life balance. I happen to fundamentally disagree with his point. Unenjoyable activites lead to a poor life

From the post:

It’s not a problem of how much time you spend with meaning, it’s about how much time you spend without meaning.

Converting from MySQL to Postgres

Tue, 21 May 2013 00:00:00 +0000

In playing with GitStreams, I’ve made the switch from MySQL to Postgres. The reason for the switch was mostly shaming from [[http://about.me/jdunck][Jeremy Dunck]], Selena Deckelmann and a few others. The results, however were quite surprising.

Storage Efficiency#

Operating on a budget $20/month Linode server (affiliate link), disk space is precious to me. In using MySQL, my data files were hovering just under 8GB for a couple of million rows of mostly-JSON data. After the migration, my size-on-disk dropped to 3GB. This is quite exciting for me as I’ve 2x’d the lifetime of this server. You can see the [[https://github.com/justinabrahms/go-gitstreams/blob/fc41957fd03224953a85cfdfec736e4cf3e62b3c/schema.sql][MySQL schema]] and the Postgres schema for comparison purposes. The main table is streamer_activity .

Ignore Django, write testable code

Tue, 21 May 2013 00:00:00 +0000

One of the best talks at PyCon 2013 was Gary Barnhardt’s Boundaries talk. One aspect of the talk was avoiding mocking by making the data interchange format based around value types. This is distinctly separate from entity types (where the identity of the object is more than the value). The canonical example value types is money. A $5 bill is a value type because it can be evenly exchanged with any other $5 bill. Entity types on the other hand are closely correlated with their identity. Even if a user has all the same preferences as another user, it’s profile is distinct.

Project Structure for Advanced Beginners

Tue, 12 Mar 2013 00:00:00 +0000

I’ve recently volunteered to help out Code Scouts, a non-profit organization tasked with helping people learn programming and increase diversity in the field. One of the other mentors, Peter Banka, had a very interesting take on project-based work.

Programming projects as an MMORPG#

Peter’s idea was simple. Instead of requiring everyone to know everything, let them specialize. The way he explained it was in terms of looking for a quest group in an MMORPG. It takes a fighter and a wizard and a few more to complete quests. That doesn’t mean one person needs to know all things (at first, especially). The smaller the surface area of things people need to learn before they’re meaningfully contributing, the more engaged people are. I’d suspect this positively correlates to fewer people churning out of the program or generally being frustrated.

Why I changed my name

Thu, 01 Nov 2012 00:00:00 +0000

Justin {Lilly,Abrahms}#

I got married, my wife and I chose to change our names. My new name is Justin Abrahms. As I expect there will be several people who have questions mostly around the question of “Why?”. So..

Why?#

I’ve been sort of interested in changing my name conceptually. I have a fascination with weird polish names like Poliskiewicz (Po-lis-ko-witz), Kowalski (Ko-wal-ski), or my good friend Andrew’s last name. I’ve joked with Tracy for years that we would change our names when we eventually got married. When the time came, we started really exploring the name thing. She asked “Well, what about my name, Hinds?”. Pffft. I just couldn’t even entertain the idea. Then it hit me: If I find it preposterous to take my wife’s name, why would I ask her to take my name?

Cellphone Plans for Geeks

Sun, 23 Sep 2012 00:00:00 +0000

Same stuff, less money#

In an effort to be more frugal, I’ve recently altered my cellphone plan to one in which I get unlimited data and text messages for $30/mo. When I say frugal, I don’t mean cheap. The goal here isn’t to spend as close to 0 dollars as possible, but rather to minimize the expense things to the point feasible, given your values. I value having a smartphone with maps (specifically with bicycling directions) quite a lot, so I’m willing to spend money for it. Eric Holscher (the only person I know with a feature phone) values knowing how to get around the city without needing a phone, so his calculation is different than mine. This is the essence of frugality. Now its just a matter of minimizing the expense such that I can still get my maps any time I want them.

Gittip: New, Interesting, Important

Sun, 23 Sep 2012 00:00:00 +0000

Great possibilities, solid direction#

Gittip is a service for donating money to people (initially, open source developers, but could be for anyone) with no strings attached. Just keep being awesome. It’s being lead by [[http://whit537.org/][Chad Whitacre]]. While the service is still a few months old, folks seem to be using it and deriving benefit. All successful markers for the system.

How it works#

Donations made through gittip are anonymous in that recipients don’t know who is giving them the money. This is meant to facilitate the idea that the money isn’t for any particular work. This isn’t “because you fixed that bug I had” or “so you WILL fix that bug I have”. It’s meant as ongoing encouragement for doing good things. The ongoing portion of this is rather important. This isn’t a tip in the sense that it is a one-time payment. You allocate some amount of money (1, 3, 6, 12, or 24 USD) to be paid out weekly to a person.

Personal Philosophy

Sat, 15 Sep 2012 00:00:00 +0000

Beliefs in action#

I’ve been doing a bit of thinking lately about my own personal values and how they drive my decisions on a day-to-day basis. I think, to some degree, this is normal when going through a big move and starting at a new job. You’re faced with people who don’t know much about you. It also draws into focus some of the differences between you and others. As such, I thought I would list a few things that I personally believe both in hopes that you might learn something of me and I might learn something of myself.

Literate programming with org-babel

Sun, 15 Jan 2012 00:00:00 +0000

As part of my ongoing love affair with emacs, I’ve started annotating my dotfiles with org-babel. This may on the surface sound syncing these things would be a tedious task, but thanks to [[http://en.wikipedia.org/wiki/literate_programming][literate programming]], its not too bad.

Literate programming?#

Literate programming is a concept in which you write prose and embed in it source code. You can see examples of it in my article on git. It has the benefit of being written for humans in the first place. I haven’t spent enough time with it to know if this is true, but I suspect it also makes updating the documentation with a given change a little easier as well.

EasyMock for mere mortals

Thu, 04 Aug 2011 00:00:00 +0000

I’ve found, during my brief experience with EasyMock, that trudging through the documentation makes it terribly difficult to get things done. From my reading, I wrote a handful of tests which use the reset(), replay(), verify() dance. When I changed trivial things in my implementation, all of my tests broke. “We expected you to call uselessMethod() but you didn’t!”

Simply fulfilling an interface#

Simply put, verify() is overrated. It turns out that the majority of what I want out of easymock is for it to override all of the interface’s methods with OperationNotSupported exceptions, except for a few precious methods, which I can provide a bit of implementation. This is done by just simply not calling verify() at the end of the method. While you don’t assert that X method was called Y times, chances are, you really don’t want to do that anyway. It ties you terribly close to your implementation, making your tests brittle.

Django at 30,000ft: A Manager's View

Mon, 25 Jul 2011 00:00:00 +0000

/Note: This article was originally published in the September 2008 edition of Python Magazine. It is published here edited only for formatting./

Django is a full-stack web framework based in Python that allows you to achieve fast turnaround times. “Full-stack web framework” basically means it comes with tools necessary to access information in your database(s), report that information back to the user and many of the other functions you need on an everyday basis when dealing with web applications.

Vim & Python: Making yourself at home

Mon, 25 Jul 2011 00:00:00 +0000

You might look at Python code every day, but what is the lens through which you view code? It’s your text editor. This article was written for Python Magazine and was first in a series on Python development environments. Here, we look at how Vim users can boost their productivity by having Vim tell you as much as possible about your program, while you’re writing it.

As developers, we spend a majority of our productive day interacting with code. How we interact with this code is often the last thing on our mind. Taking a step back and looking at our daily habits can often yield surprising results in productivity, and is a healthy thing to do once in a while.

Writing a static site generator using org-mode.

Thu, 14 Jul 2011 00:00:00 +0000

This site is now generated through org-mode, an emacs library which is used for outlining. The generation of the HTML lies in the export functionality of outlines. The benefits of this system is that its easy, uses a tool that I’m already familiar with, and extensible.

Motivations#

My old site was written in Django and was consuming a fair amount of system resources. This has little to do with Django itself, but more with my hosting setup. In an effort to simplify, I’ve changed hosting and, with this, changed what generates my website. Part of this change is removing many of the “expensive” operations in a server, such as python interpreters that run per site. Now, while sustaining 150 requests per second, the hosting setup only consumes 4% of my 64mb of ram allotment.

6 things I learned about setuptools

Fri, 17 Jun 2011 00:00:00 +0000

While building pyvcs and django-vcs over 4th of July weekend with Alex Gaynor [fn:0] , there were a few interesting lessons learned, mostly revolving around how setuptools works. As Jacob mentions in his [[http://jacobian.org/writing/django-apps-with-buildout/][post on buildout]] , you mostly have to cobble together the usage of setuptools from other setups. This works well at about 90% of the case. There are a few gotchas, however.

find_packages(‘directory’) will look under directory for a list of python modules to install. For our applications, we actually just wanted find_packages()
You can’t easily specify multiple authors for a project. We settled on author = “First Author, Second Author”
Its a good idea to version target your requirements. Don’t require “pyvcs”, require “pyvcs>=0.1”.
If your project contains django templates, set the zip_safe property to False.
The register or sdist upload setup.py commands actually check that your project installs. For that, its likely better to use a virtualenv and test the installation.
package_data is used to includes docs (or in our case, templates) with the project.

[fn:0] You can find the announcement here.

Bash: Your forgotten friend (part 1)

Fri, 17 Jun 2011 00:00:00 +0000

In the name of meta-productivity, or improving productivity by improving process, I’ve come to learn and love Bash. For those who aren’t familiar, Bash is the “scary” black terminal in Mac and Linux. Its often recognized as the home of true geeks a la Wargames. From here out, I’ll assume a basic knowledge of bash (what “export” and similar commands do). In an effort to spread some of the joy of Bash, I’ve included a few handy features and tricks that help me out day to day.

Bash: Your forgotten friend (part 2)

Fri, 17 Jun 2011 00:00:00 +0000

Aliases are one of the core portions of Bash that will help you become more productive almost immediately. To define an alias, the command is simply:

alias short='long command goes here'

Super simple! To save aliases, however, you’ll need to put them in your .bash_profile so they load automatically. To see a list of currently defined aliases, type alias. Here are my aliases, chunk by chunk, and a brief bit about what they do.

Bash: Your forgotten friend (part 3)

Fri, 17 Jun 2011 00:00:00 +0000

For the final installment of Bash: Your forgotten friend (see: part one and part two ), I’ll be going over the functions I use to make life easy. For those unfamiliar, functions aren’t as scary as they sound. They’re basically things that are too complex for simple aliases but aren’t worth writing an entire bash script to do. This doesn’t, however, mean that they aren’t powerful. You can pack a lot of functionality into these small functions.

Customizing Django Forms Multiwidgets

Fri, 17 Jun 2011 00:00:00 +0000

This once contained slides to a talk I gave to django-nyc. While I don’t know where those are anymore, you can find the resulting code demonstration on github.

Developing a Personal Brand

Fri, 17 Jun 2011 00:00:00 +0000

I’ve spent the last year in classes at AiO, and they have really brought me a better understanding on the importance of things like branding. It provides a consistent feeling across your site and other products, such as business cards, resumes, and invoices. It offers people who interact with you assurance that they’re dealing with the same person as before through the use of visual touchstones. A lot of people get the concept of branding wrong, though. Branding isn’t about putting your logo on everything you own. That’s called making something garish. Branding is just using a consistent style across everything: the same font, the same colors, even something as intangible as the same sense of humor in your writing. It all plays a part in your brand.

Django Portfolio Application

Fri, 17 Jun 2011 00:00:00 +0000

As I near the end of my run with AiO, I’ve been asked to come up with a portfolio. This has been pretty time consuming as there have been several rounds of critiques as I went through the process. The last stage of this journey, at least as far as the school is concerned, is the creation of an interactive portfolio. While the class will be doing theirs in Dreamweaver (eew), I’ve elected to create a reusable application for django to handle this. This is basically a mind dump of what it should include in hopes someone will spot something I’ve missed.

django-voting: A brief tutorial

Fri, 17 Jun 2011 00:00:00 +0000

django-voting is a reusable application which gives you reddit-style voting capabilities on generic objects. The application, itself, is a very well documented piece of software. Then again, that’s not saying a lot considering its from Jonathan Buchannan. All of his projects are famously documented.

Installation#

The installation of this application was standard fare for a google code app. SVN Checkout the application to a point on your python path and that’s about it. For me, the command was: ::

Django, PyISAPIe and IIS

Fri, 17 Jun 2011 00:00:00 +0000

For those poor souls who are forced to deploy Django on IIS, I feel your pain. I spent the better part of my morning fixing a broken PyISAPIe. To shortcut your trek, here’s what I’ve found.

Signals have changed since the last release. To fix this, I altered the call to signals to directly call the signals instead of importing the signals module. I then updated the signal calls themselves to reflect the new style.

Domain Sniffing Middleware

Fri, 17 Jun 2011 00:00:00 +0000

My first project for the new job has been creating an internationalized version of the website. I think it also serves as a good test run for how django works with the infrastructure that’s in place. Instead of having two separate django installations, each with mirrored files (save settings.LANGUAGE_CODE), I wrote a quick middleware which will add a session key for django_language, which is used by the locale middleware to determine which version of the site to show you, based on the domain you’re accessing.

Emacs and Transmit: Remote edit Fix

Fri, 17 Jun 2011 00:00:00 +0000

So I’m a huge fan of Panic’s Transmit application, as referenced by my favorites post. Its ease of use makes editing remote files not suck. One very cool feature of Transmit is its ability to edit remote files with any sort of editor. This works famously with both Textmate and Vim, but didn’t quite work for me in emacs. After sending an email to the fellows at Panic, I got the fix in about 20 minutes. Those guys completely rock (go buy their software!).

Emacs: Being Productive

Fri, 17 Jun 2011 00:00:00 +0000

As not to give false ideas of what this post is about, let’s lay it out up front. I’m going to show you how to be able to get work done in emacs. You’re not going to be an emacs guru after reading this. You’re not even going to be all that productive. That being said, you will be able to accomplish some work.

Configuration for an Effective Session#

Following Steve Yegge’s epic post, there are a few things in there that you might question. I know I did. He suggests swapping the places of capslock and control. At first this just seemed pointless. I ignored his suggestion and after 2 days, I got a case of emacs pinkie which is pain from over use. Taking his suggestion, however, has proven to be quite helpful. It keeps my fingers on the home row a majority of the time which allows for speedier typing. Another change he suggested, C-w as kill-word, seemed to be another one of those small gains that wasn’t worth doing. Turns out, I still don’t. Kill-word itself is quite useful, but remapping C-w makes using other’s emacsen painful. I’ve tried to stop customizing the built-in shortcuts for portability. The only one I can’t live without is C-M-hjkl for navigation between buffers.

Emacs: Introduction & Installation

Fri, 17 Jun 2011 00:00:00 +0000

A (very) Brief Background#

So emacs is one of those programs that never dies.. it just looks that way. Development for emacs began in the 70’s and is still active, making it somewhat of a big deal in terms of computer history. Emacs stands for Editor MACroS and was originally designed for some archaic system I’ve never heard of. If you’re really that interested, check out the wikipedia entry. The plural form of emacs is emacsen. There is a very popular fork of emacs called Xemacs, which forked in 1991 apparently due to emacs’s glacial release cycle. Over time, they’ve converged on features (I think), and mostly serve as a headache to people wanting to support both. Enough about that.. On to the important stuff.

Emacs: The Mindset, A contextual shift

Fri, 17 Jun 2011 00:00:00 +0000

There were a few insights I got when I switched from vim to emacs. Hopefully they’ll help you get a leg up.

Meta vs. Ctrl#

Learning the difference between meta and ctrl was my first aha moment when I started using emacs. There is a definite difference in the general functions of the control key and the meta key in reference for navigation and text manipulation.

Fixing your models subdirectory

Fri, 17 Jun 2011 00:00:00 +0000

For those of you who keep your models in a subdirectory, you may have noticed that you can’t properly load fixtures. This is because Django checks for the location of your models file to determine your app directory. To fix this, move your models directory to something like modeldir and create a models.py which imports the models from inside.

Before:

myapp
|__ views.py
|__ __init__.py
|__ models/
|____ __init__.py
|____ mymodel1.py
|____ mymodel2.py

After:

Google account suspended: A post mortem

Fri, 17 Jun 2011 00:00:00 +0000

A few weeks back, I had my Google account disabled. If you’re anything like me, your first reaction is a lackadaisical, “That sucks.” Then the realization of what that really means starts to creep in to your gut, making you nauseous. No more email. No more iGoogle homepage. No more feed reader. Next year’s vacation spots you had mapped out in a comparison chart on Google Docs? Yep, those are gone too. More Nausea.

Importance of Documentation

Fri, 17 Jun 2011 00:00:00 +0000

I think there is a bit of a pain point for me currently with regards to open source software: documentation. I’ve been in a position where I need to create a GUI application for HTML generation with a bit of database interaction. Nothing too serious, but a bit of help wouldn’t be a bad thing.

The first result google brought to me was dabo, a gui application framework based on wxPython. Cross platform GUI apps with the ease of development provided by a framework? Sign me up! Then came my impasse. No download link. I traversed their wiki for a solid 15-20 minutes trying to find a way to download this awesome application. No results, it would seem.

Javascript unit tests with JS Test Driver

Fri, 17 Jun 2011 00:00:00 +0000

The general idea for JS Test Driver is that it allows you to run javascript unit tests. While this is nothing new, there are some interesting bits. JSTD provides XML output of tests running, which can be used for CI integration and test coverage. It can also run your tests across multiple browsers.

JS Test Driver Installation#

JSTD’s only dependencies are java and a browser, which makes it very easy to get it setup. I’m assuming you are on an Ubuntu machine, if you aren’t, you’re on your own.

Macports and PYTHONPATH

Fri, 17 Jun 2011 00:00:00 +0000

Setting up my new macbook, I’ve had to go through the joy and pleasure that is macports. Overall, it wasn’t so bad. A few missing packages here or there. Everything was going fine until I tried to install Fabric. Its dependency on pycrypto kept failing. I had installed pycrypto via macports and no luck. Turns out there are two paths under opt which need to be added to the python path in order to get it to work.

Most Used Programs: An Index

Fri, 17 Jun 2011 00:00:00 +0000

/Note: This represents a snapshot in time. I may or may not still be using these pieces of software./

Attempting to index those programs you use every day is, in my opinion, a worthwhile task, if only in hopes of getting suggestions for X program being better than Y.

Browser#

I’m a firefox junkie. There are few times when I’ve had to turn to other browsers outside cross-browser testing. Webkit offers some nice speed increases (especially where javascript is concerned), but firebug, delicious plugin and greasemonkey really make this the browser for me.

NameError: global name 'log' is not defined

Fri, 17 Jun 2011 00:00:00 +0000

I’ve spent a portion of this evening trying to get caught up with the crazy amounts of development that have been going on with the oebfare blogging engine over the past few weeks. I have a nice little Ubuntu VM setup that works really nicely, but when gathering the requirements for the projects, I hit a snag. I apparently had a borked setuptools (installed via apt) which resulted in the error::

Piping to stdout when using BuildBot

Fri, 17 Jun 2011 00:00:00 +0000

Just a quick tidbit for anyone googling like I have been all morning:

If you’re attempting to pipe a command in a buildbot script, you need to issue it as a single string, not a list of strings. By using a single string, buildbot uses a shell, otherwise it doesn’t. The > necessary for output redirection is a shell thing, which will only work in this method.

To recap:

f1.addStep(ShellCommand(command=['/home/www/bin/django', 'build_solr_schema', '>', '/etc/solr/conf/schema.xml'],
 description=['building', 'schema'], descriptionDone=['schema','built']))

becomes:

psycopg2 Internal Error

Fri, 17 Jun 2011 00:00:00 +0000

For those who’ve hit the following complaint with django (like I just did):

psycopg2.InternalError: current transaction is aborted, commands ignored until end of transaction block

You need to set up PostGIS and ensure that the table you’re looking at has been created from the template_postgis template. Just a warning, check your database logs for more info.

Speedy Google Code Checkouts

Fri, 17 Jun 2011 00:00:00 +0000

Here’s a handy bash shortcut for automatically checking out a google-code project.

# SVN ALIASES
export GOOGLE_CODE_CHECKOUTS="$HOME/Code/trunks"
gcco () {
 svn checkout http://$@.googlecode.com/svn/trunk/ $GOOGLE_CODE_CHECKOUTS/$@
}

Top 5 Bookmarklets

Fri, 17 Jun 2011 00:00:00 +0000

Insert jQuery : This bookmarklet is amazing if you’re a firebug user. It allows you to use jquery in firebug, even when the site doesn’t support it. I use it for things ranging from checking a bunch of checkboxes or checking the number of uses for a particular css class. You don’t know that you’re missing it until you don’t have jQuery at your fingertips. (The bookmarklet is towards the middle)

Using Database views with Django

Fri, 17 Jun 2011 00:00:00 +0000

Database views will make their appearance in the upcoming Django 1.1 release. Having used them for my latest project, there are a few gotchas when it comes to using them.

In our project, we were using South for our database migrations. To ensure our database views were created, we added them to a south migration using db.execute. This worked very well for 90% of our use case. The biggest hurdle for this ended up being testing. When syncdb runs, your unmanaged models (ie: database views) aren’t generated. There were a few options available to us at this point.

Using test generators to save time

Fri, 17 Jun 2011 00:00:00 +0000

Test generators are an amazing tool when you are testing small variations in the same code path. It uses generators[fn:0] which yield callables[fn:1] that are called as the test suite runs. For example, imagine you’re testing a rounding library.

class TestRound(object):
 def _check_rounding(self, test, expected):
 assert round(test) == expected

 def test_rounding(self):
 for x, y in [(1, 1),
 (1.9, 2)]:
 yield self._check_rounding, x, y

# VERSUS

class TestRound(unittest.TestCase):
 def test_rounding_same(self):
 self.assertEquals(round(1), 1)

 def test_rounding_one_dot_nine_to_two(self):
 self.assertEquals(round(1.9), 2)

While these two examples look mostly similar for 2 cases, imagine what ten cases would look like, or 50. For a much more real-world example, imagine a test generator which will crawl your webapp’s url routing setup and assert that each of them return a 200 status code.

Versioning Production Settings

Fri, 17 Jun 2011 00:00:00 +0000

In deploying my latest side project, I came across a conundrum. How do you manage production specific settings files (complete with passwords, API keys and other secrets), while still publishing the source code?

Until now, I’ve had unversioned production settings sitting around the file system. While this hasn’t failed me yet, its only a matter of time. Sensing an opportunity to derail a perfectly good deployment for another half hour, I’ve come up with what I think is a seemingly solid way to keep those production settings around: git submodules.

Vim Screencast: % Operator

Fri, 17 Jun 2011 00:00:00 +0000

Here’s another in my series of vim screencasts. This one talks about the various uses of the % operator. You can find the video at http://showmedo.com/videotutorials/video?name%3D3160020 (likely dead)

In non-editing mode, the % operator swaps between (), [] or {} style braces. This is super helpful for PHP style syntax that wraps large portions of code in these. The real power of the % operator comes in Ex mode, however.

Ex mode is the mode that is triggered by the colon ( : ). It prefaces such commands as :w[rite] or :q[uit]. In this mode, the % sign represents your current working file. You can use it to great effect with the ! operator which escapes to the shell temporarily. This allows commands such as::

Vim Screencast: An Introduction

Fri, 17 Jun 2011 00:00:00 +0000

In my latest screencast, I’ve gone a bit longer than normal (about 7 minutes) in order to cover an introduction to vim. In this screencast, I give a brief overview of modal editing and what that means. I then go over how to move around in vim, how to enter editing mode in various and interesting ways. After that, how to cut / copy and paste were next on the list. Finally, I showed how to save and exit a vim file. You can find the video at http://showmedo.com/videotutorials/video?name%3D3160030%3BfromSeriesID%3D316 (Unfortunately dead since original publishing).

Vim Screencast: Macros

Fri, 17 Jun 2011 00:00:00 +0000

Having tried for literally days to get a legible screencast uploaded and online to any of a number of video services, I’ve decided to link to the actual file itself. This is a very quick and simple screencast of how to use macros in Vim. You can find the video at showmedo.com/videotutorials/video?name=3160000&fromSeriesID=316 (likely dead).

Vim Screencast: Selections

Fri, 17 Jun 2011 00:00:00 +0000

This is a relatively short screencast on a few different ways to select text in Vim. Vim’s visual mode is a very powerful feature of the editor which allows you to select data in different, yet meaningful, ways.

The screencast goes over normal visual mode, visual block mode, and text object selection. The relevant helpfiles are::

:h visual-mode
:h v_b_i
:h text-objects

As a developer, its as important to know your editor as well as a carpenter knows his hammer. You can find this video at http://showmedo.com/videotutorials/video?name%3D3160010 (likely dead).

Vim: My new IDE

Fri, 17 Jun 2011 00:00:00 +0000

On Michael Trier’s suggestion, I’ve decided to make Vim my primary editor once again. Having recently switched from Textmate, there were a few things that I needed to get up and running to stay productive. Here they are in no particular order.

NERDtree#

The NERDtree plugin functions much the same way that TM’s project drawer works. It allows a hierarchal view of a folder structure. From this view, you can open files in new tabs, in the opposite window, even in split views. The only downside I’ve come across yet is the inability to perform file operations from within NERDtree. That being said, TM’s implementation of this was always felt a bit wonky to me as well.

Virtualenv Wrapper Helper

Fri, 17 Jun 2011 00:00:00 +0000

A bit inspired by all the awesome stuff happening at PyCon, I’ve done a bit of hacking on my shell to improve the use of virtualenvwrapper. Remembering to checkout the correct virtualenv is a bit of a pain, especially when you move around your filesystem as much as I do. To fix this, I’ve written a bit of code for my bash prompt to ease the pain. It works by looking for the existence of a .venv file every time you cd into a directory. If it exists, it will attempt to workon the contents of the file, which should be the project name. For instance, for my oebfare project, the file contains the name of my virtualenv. Due to this, the following is possible:

My motivations for choosing emacs

Wed, 15 Jun 2011 00:00:00 +0000

The EmacsVsVi wiki page on the PythonInfo wiki describes emacs as

Among technically inclined UNIX*Linux users vi*vim tends to be
preferred more by sysadmins while emacs/xemacs tends to be favored
by programmers.

In an effort to “know your enemy” and whatnot, I figured I would give emacs a shot. I’ve tried the tutorial a few times before and typically got about 1/2 way through before giving up on the control and alt craziness. This time, however, I actually managed to get through the tutorial.

GTD Excel File

Fri, 04 Mar 2011 00:00:00 +0000

Hi. I’m guessing you’re looking for the GTD excel file. Unfortunately, this file is no longer available.