Code[ish]

Nowadays, the internet is so huge that it can be hard for people to find others who share their niche interests. But when they do find that rare kindred spirit, it can feel like a magical moment. Lynn Fisher and design agency &yet have been exploring ways to help people build community around their passions (which can sometimes be a little “weirdâ€). The team launched a project called “Find Your Weirdos†that incorporates different tools, sites, and techniques for helping people connect with their fellow weirdos. Their project also helps companies connect with customers through niche interests.

Inspired by the Weirdos project, the &yet team envisioned ways to help Heroku developers connect and the Wicked CoolKit was born. The kit harkens back to the earlier days of the internet, when simple, fun web widgets and tools helped people connect without all the noise of today's mega social platforms. The initial version of the kit offers a new take on a few nostalgic web widgets, including:

• Developer trading cards â€" Echoing the retro joy of collecting baseball cards or playing card-based games, this widget allows developers to create their own profile card. They can specify their personal bio, coding skills, niche interests, “feats of strength,†and more, and share it within an elegantly designed UI.

• Themed stickers â€" A perennial favorite, stickers are a colorful way to identify interests, such as baking or woodworking. Users can download stickers to use as they wish, or add a sticker to their trading card that links to other people’s cards that have the same sticker.

• Webring â€" Years ago, fans and friends would use a webring to share a collection of websites dedicated to a specific topic. The kit brings the old school webring into the modern context and allows people to easily share and access web resources.

• Hit counter â€" Everyone wants to know how many visitors came to their site. The old-fashioned hit counter is a fun way to track and display page visits. The higher the number, the more likely people will want to engage with the site (and the developer behind it).

The Wicked CoolKit is fully open source and available to use.

• Lynnandtonic.com â€" Lynn’s personal website.
• wickedcoolkit.com â€" Home of the Wicked CoolKit
• Show, don’t tell â€" the story behind the Wicked CoolKit.
• Find.yourweirdos.com â€" a series of essays on how companies connect with customers through sharing mutual niche interests.
• Face.camp â€" an app that connects to Slack for people to capture and post animated gifs.
• Wegotchu.cards â€" digital cards that people can pass around and sign.

Corey Martin leads the discussion with two developers about production incidents they were personally involved in. Their goal is to inform listeners on how they discovered these issues, how they resolved them, and what they learned along the way.

Ifat Ribon is a Senior Developer at LaunchPad Lab, a web and mobile application development agency headquartered in Chicago. For one of their clients, they developed an application to assist with the scheduling of janitorial services. It was built with a fairly simple Ruby on Rails backend, leveraging Sidekiq to process background jobs. As part of its feature set, the app would send text messages to let employees know their schedule for the week; these schedules were assembled by querying the database several times, fetching frequencies and availabilities of workers. Unfortunately, a client noticed a discrepancy between how many notices were being sent out, versus how many jobs they knew they had: of the 400 jobs total, only 150 had notifications. It turned out that all of the available database connections were being exhausted--but that was only half of the issue. Sidekiq was attempting to process far too many jobs at once, and each of these jobs were responsible for connecting to the database, exhausting the available pool. The solution Ifat settled on was to reduce the number of parallel jobs processed while increasing the number of connections to the database. From this experience, she also learned the importance of understanding how all these different systems interconnect.

Christopher Ostrowski is Chief Technology Officer at Dutchie an e-commerce platform for the cannabis industry. One Christmas Eve, while celebrating with his family, Chris began receiving pager notifications warning him about some sluggish API response times. Since it didn't really have any significant end user impact, he ignored it and went back to the festivities. As the night went on, the warnings became significant alerts, and he pulled together a response team with colleagues to figure out what was going on. By all accounts, the website was functioning, but curiously, the rate of orders began to drop off. Through some investigation, they realized what was going on. Customers' order numbers were assigned a random, non-sequential six digit numbers. Dutchie was about to track its one-millionth order, a huge milestone. Before any orders are created, though, the app generates a six digit number, and tries to create one that doesn't already exist. The database was constantly being hit, as less and less six digit numbers were available for use. The solution ended up being rather simple: the order number limit was increased to nine digits. Although they had monitoring in place, the data was set up as an aggregate reporting; even though the "create order" API was slow, all of the others were low, keeping the average within tolerable levels. Christopher's solution to avoid this in the future was to set up more groupings for "essential" API endpoints, to alert the team sooner for latency issues on core business functionality.

• LaunchPad Lab is a web and mobile application development agency
• Dutchie is an e-commerce platform for the cannabis industry

Rick Newman is a Director of Engineering at Salesforce Heroku. He's joined by Marco Faella, a professor of advanced programming and author of "Seriously Good Software." In Marco's view, there are of course several ways ways to characterize "good" software. Excellent software that goes above and beyond correct functionality includes code that is readable, robust, and performant. Each of these have different importance, depending on context. Robust software, for example, includes addressing issues with scalability, but only if one expects the software to be in such a high availability environment.

It's important to address these requirements from the beginning, when the software architecture is being mapped out. Marco gives the example of developing software for an external client. This client might know all the business logic and how it ought to function, but addressing the code's future evolution and maintenance are just as important, and whose responsibility lands squarely in the hands of the developer.

It can also be worthwhile to make an investment in education, learning about algorithms, data access, and other key concepts in the world of computer science. Such a foundation would allow one to adapt to the changing conditions of programming, whether those are caused by new hardware or modifications in the languages themselves.

• "Seriously Good Software" is Marco's book on the subject of writing strong code -- get a 40% discount with the code podish19

Host Greg Nokes is a distinguished technical architect with Heroku. His guests are Alli McGee, a product manager, and Lewis Buckley, a senior application engineer, from BiggerPockets. BiggerPockets was founded 16 years ago to educate non-professionals about real estate investing.

As a self-funded company, it's critical for BiggerPockets to create products that customers will pay for. One way they achieve this product/market fit is by building cross-functional teams that are user-focused. All product teams have a project manager, tech lead, and designer that work closely together. This design-led approach allows teams to collaborate with representation from users, technology, and design.

As the PM on one of these teams, Alli lives at the intersection of what can we do for business, what can we do from a technology perspective, and what can we do for the user. She advocates for the customer, bringing knowledge of what customers want, what problems they are facing, and how they have interacted with prototypes in usability studies. Alli also advocates for the business to be sure products make money. Finally, Alli advocates for developers to make sure the project is technically feasible and won't cause technical debt.

Another way BiggerPockets creates market fit is by creating Minimum Lovable Products—the smallest cheapest thing they can build that people love. With their current product, BP Insights, Alli and Lewis used this strategy to create the first iteration of a product that provides insight into local real estate markets. They then tested the product with users, iterated, and slowly built out a more fully formed offering.

For their tech stack, BiggerPockets is built on a Ruby on Rails monolith. While some in the industry say Ruby on Rails' time is over, Lewis argues that it has been a great choice, as using a well-known stack has allowed them to worry less about the technology and focus more on building value for users. The BP Insights product was built on this monolith using a massive data set of nearly every property in the US. BiggerPockets imported the data to an Amazon S3 bucket and eventually copied the data to Amazon Redshift for querying.

• BiggerPockets

Robert Blumen is a DevOps engineer with Salesforce, and he's joined in conversation with Andrzej Ludwikowski, a software architect at SoftwareMill, a Scala development shop. Andrzej is introducing listeners to the concept of event sourcing against the more traditional pattern of CRUD, which stands for create-read-update-delete. CRUD systems are everywhere, and are most typically associated with SQL databases. In comparison, event sourcing is a simply a sequential list of every single action which occurred on a system. Whereas in a database, a row may be updated, erasing the previous data in a column, and event source system would have the old data kept indefinitely, and simply record a new action indicating that the data was updated. In a certain sense, you can get the state of your system at any point in time.

Each architectural pattern has its pros and cons. For one, an event source system can make it easier to track down bugs. If a customer notes an issue an production, rather than pouring through logs, developers can simply "rewind" the state of the application back to some earlier event and see if the faulty behavior is still there. On the flip side, since the event stream is immutable, fixes to previous data needs to be made at the end of the stream. You can modify old events or insert new ones into the flow.

CQRS, or Command Query Responsibility Segregation, builds on top of event sourcing. The idea is to separate the part of the application responsible for handling commands and writes from the part responsible for handling queries and reads. This separation is not only on a software level (different repositories and different deployments), but also on the hardware level ( different hosts and different databases). The motivation for this is to be able to scale each part independently. Maybe your app has more writes than reads, and thus requires different computing power. It allows for a separation of concerns, and can make overall operations more efficient, albeit at a complexity cost. Andrzej is quick to note that event sourcing and CQRS divisions are not necessary for every application. Teams, as always, need to understand how the data flows in their application and which architectural pattern is most efficient for the problems they are trying to solve.

• SoftwareMill is a Scala development shop
• Martin Fowler gives a brief run-down on event sourcing
• On the SoftwareMill blog, Andrzej has a blog posts on entry-level event sourcing, keeping your domain clean in Event Sourcing, and the best serialization strategy for Event Sourcing
• CQRS.nu provides more educational resources
• Andrzej Ludwikowski's web site and talk Event Sourcing: What could go wrong?

Luke Kysow is a software engineer at HashiCorp, and he's in conversation with host Robert Blumen. The subject of their discussion is on the idea of a service mesh. As software architecture moved towards microservices, several reusable pieces of code needed to be configured for each application. On a macro scale, load balancers need to be configuring to control where packets are flowing; on a micro level, things like authorization and rate limiting for data access need to be set up for each application. This is where a service mesh came into being. As each microservice began to call out to each other, shared logic was taken out and placed into a separate layer. Now, every inbound and outbound connection--whether between services or from external clients--goes through the same service mesh layer.

Extracting common functionality out like this has several benefits. As containerization enables organizations to become more polyglot, service meshes provide the opportunity to write operational logic once, and reuse it everywhere, no matter the base application's language. Similarly, each application does not need to rely on its own bespoke dependency library for circuit breakers, rate limiting, authorization and so on. The service mesh provides a single place for the logic to be configured and everywhere. Service meshes can also be useful in metrics aggregation. If every packet of communication must traverse the service mesh layer, it becomes the de facto location to set up counters and gauges for actions that you're interested in, rather than having each application send out non-unique data.

Luke notes that while it's important for engineers to understand the value of a service mesh, it's just as important to know when such a layer will work for your application. It depends on how big your organization is, and the challenges you're trying to solve, but it's not an absolutely essential piece for every stack. Even a hybrid approach, where some logic is shared and some is unique to each microservice, can be of some benefit, without necessarily extracting everything out.

• HashiCorp helps automate infrastructure configuration
• Consul, Linkerd, Istio, and Kuma are several open source components for service meshes and control planes
• Mastering Service Mesh: Enhance, secure, and observe cloud-native applications with Istio, Linkerd, and Consul by Anjaki Khatri and Vikram Khatri
• Consul service mesh
• What's a service mesh? And why do I need one? by William Morgan offers additional advice on when to use service meshes
• The Service Mesh: What Every Software Engineer Needs to Know about the World's Most Over-Hyped Technology
• What is a Service Mesh? - a blog from NGINX

Rick Newman interviews Mikolaj Pawlikowski, who recently wrote a book called "Chaos Engineering: Crash test your applications." The theory behind chaos engineering is to "break things on purpose" in your operational flow. You want to deliberately inject failures that might occur in production ahead of time, in order to anticipate them, and thus implement workarounds and corrections. Typically, this practice is often used for large, distributed systems, because of the many points of failure, but it can be useful in any architecture.

One of the obstacles to embracing chaos engineering is finding high level approval from other teammates, or even managers. Even after the feature is a complete and the unit tests are passing, it can be difficult to convince someone that some resiliency work needs to continue, because there's no visible or tangible benefit to preparing for a disaster. Mikolaj suggests that people clearly lay out to other colleagues the ways a system can fail, and the impact it can have on the application or business. Rather than try to fear monger, it can be useful to point to other companies' availability issues as words of caution for their teams to embrace. Mikolaj also says that chaos engineering doesn't need to focus solely on complicated problems like race conditions across distributed systems. Often, there's enough low hanging fruit, such as disk space running out or an API timing out, that can be useful to consider fixing.

The chaos engineering mindset can also extend beyond pure software. If you think about people working across different timezones as a distributed system, you can also optimize for failures in communication before they occur. Everyone works at a different pace, and communication issues can be analogous to a network loss. Rather than fix miscommunications after they occur, establishing shared practices (like writing down every meeting, or setting up playbooks) can go a long way to ensuring that everyone will be able to do their best under changing circumstances.

• Mikolaj's book is called Chaos Engineering: Crash test your applications -- get a 40% discount using the code podish19
• powerfulseal is a testing tool for Kubernetes clusters
• Mikolaj distributes the Chaos Engineering Newsletter
• Conf42 is a conference focusing on high-level computer science
• ChaosConf is the world's largest Chaos Engineering event
• Awesome Chaos Engineering is a curated list of Chaos Engineering resources

Robert Blumen is a DevOps Engineer at Salesforce, joined by Ev Haus, Head of Technology at ZenHub. Together, they're going over a critique over several methodologies when writing code as part of a large team. First, there's DRY, which stands for Don't Repeat Yourself. It's the idea that one should avoid copy-pasting or duplicating lines of could, in favor of abstracting as much repeated functionality as possible. Then, there's DAMP, or Don't Abstract Methods Prematurely, which is somewhat in opposition to DRY. It advises teams to not create abstractions unless they are absolutely necessary. Last on the list is WET, or Write Everything Twice. This is the idea to embrace duplication whenever possible.

Ev notes that, like many programming absolutes, the success of each strategy depends entirely on the context. DRY, for example, sounds like a really good idea, until it happens everywhere. Suddenly, a chunk of code becomes difficult to reason, as a developer jumps around various method definitions to piece together a flow. DAMP often makes sense as a counterpart to DRY, because if you abstract too early in your codebase, you may find yourself overloading methods or appending arguments to handle one-off cases. DRY is typically best suited for testing environments, where an absolutely reproducible set of explicit steps is often preferable in order to quickly understand what is occurring.

No matter the strategy you use, the core tenant is to solve the problem first. Try to accomplish the goal you need to, whether that's adding a feature or squashing a bug. Don't over optimize until you've finished what you need to, and don't think too far into the future about all the possible edge cases. The rest of the balance comes with experience. Some duplication is bad, but not all of it. Figuring out the absolute perfect solution is unlikely, so you've got to put the code out into the real world to find out what works. After that, bake some flexibility into your processes to adjust hot code paths or refactor them when needed!

• ZenHub is an agile project management tool for GitHub
• Wikipedia's definition of DRY
• "Using DRY, WET & DAMP code" is Ev's article on different coding methodologies
• Codewars is a website with programming puzzles and challenges
• The Pragmatic Programmer by Dave Thomas is a popular book highlighting some of these concepts
• DRY code, DAMP DSLs by Jay Fields and DRY vs DAMP in Unit Tests by Vladimir Khorikov are more write-ups on the subject

Host Joe Kutner is an architect working at Salesforce, and his guest is Cornelia Davis, the CTO of Weaveworks, a platform for infrastructures. Cornelia argues that most companies building complex web-based applications are doing so without fully understanding the unique operational challenges of that environment. Even several well-known patterns, such as adding circuit breakers or retry patterns, are not standardized across the industry, and certainly not across languages, let alone in frameworks and other easily consumable dependencies. In many cases, there are over reliances on infrastructure availability that only become obvious once a problem occurs. Cornelia gives the example of a massive AWS outage that occurred several years ago. For many companies lacking redundancy contingencies, their applications were offline for hours, through no fault of their own.

Another potential conflict between operational patterns and software design emerges around container-based lifecycles. If you have a new application configuration that you want to deploy, Kubernetes, which is designed to be stateless, encourages you to simply get rid of a pod and start up a new one. But it's entirely possible that there's some running code that doesn't know how to pick up these new changes, or even a service which can't recover from unexpected downtime. Considering these issues is the difference between running the cloud and being truly cloud native.

To the industry's credit, Cornelia does see more platforms and frameworks adopting these patterns, so that teams don't need to write their own bespoke solution. However, it's still necessary for software developers and operational engineers to know the features of these platforms and to enable the ones which make the most sense for their application. There is no "one size fits all" solution. As the paradigms mature, so too does one's knowledge of the interconnected pieces need to grow, to prevent unnecessary errors.

• The twelve-factor app
• Cornelia wrote a book called Cloud Native Patterns (get a 40% discount using the code podish19)

Hailey Walls is a Customer Solutions Architect with Heroku, and she's engaged in a conversation with Paul Orland, the founder of Tachyus and author of Math for Programmers. Paul took graduate level math classes, and even ended up with a Master's degree in Physics, but even he admits that he comes down with his own kind of math anxiety. Now, he works as a programmer, building predictive models, but he encounters many engineers who don't have a basic understanding of fundamental math concepts, like calculus or linear algebra. Seeking to rectify this, he wrote a book called Math for Programmers, which methodically explains mathematical concepts using real-world examples. He hopes to be able to teach math to many more people.

Paul emphasizes that, although thinking of mathematics can be intimidating, it's not different than working on any other skill. If you decide to go weight lifting, you start with a 10 pound weight, then a 15 pound one, and on and on. Similarly, with math, if you train on problems that are simpler, future problems will build upon the techniques you've honed. The appeal for gaining math skills is almost analogous to that of programming: there is always a right and final answer. Just as a compiler determines how a program works and whether a syntax is valid, taking in input and producing output, so too is math deterministic. Fundamentally, better mental acuity with math can help teach you how to consider the behaviors of complicated systems.

For people interested in studying math more closely, Paul advises students to not be discouraged by problems which appear hard. It can be best to pick a problem that you are naturally interested in, which will lead to a general willingness to try and solve it. Similarly, he'll also take a math concept and turn it into a program, which has helped him reason about flow and patterns much more clearly in the past.

• Tachyus helps energy producers optimize their production
• Math for Programmers is the book Paul wrote -- listen for a 40% off code