Tim Berglund

Presentations

Decision Making in Software Teams

Alistair Cockburn has described software development as a game in which we choose among three moves: invent, decide, and communicate. Most of our time at No Fluff is spent learning how to be better at inventing. Beyond that, we understand the importance of good communication, and take steps to improve in that capacity. Rarely, however, do we acknowledge the role of decision making in the life of software teams, what can cause it to go wrong, and how to improve it.

In this talk, we will explore decision making pathologies and their remedies in individual, team, and organizational dimensions. We'll consider how our own cognitive limitations can lead us to to make bad decisions as individuals, and what we might do to compensate for those personal weaknesses. We'll learn how a team can fall into decision-making dysfunction, and what techniques a leader might employ to healthy functioning to an afflicted group. We'll also look at how organizational structure and culture can discourage quality decision making, and what leaders to swim against the tide.

Software teams spend a great deal of time making decisions that place enormous amounts of capital on the line. Team members and leaders owe it to themselves to learn how to make them well.

Complexity Theory and Software Development

Some systems are too large to be understood entirely by any one human mind. They are composed of a diverse array of individual components capable of interacting with each other and adapting to a changing environment. As systems, they produce behavior that differs in kind from the behavior of their components. Complexity Theory is an emerging discipline that seeks to describe such phenomena previously encountered in biology, sociology, economics, and other disciplines.

Beyond new ways of looking at ant colonies, fashion trends, and national economies, complexity theory promises powerful insights to software development. The Internet—perhaps the most valuable piece of computing infrastructure of the present day—may fit the description of a complex system. Large corporate organizations in which developers are employed have complex characteristics. In this session, we'll explore what makes a complex system, what advantages complexity has to offer us, and how to harness these in the systems we build.

Getting Started with Grails

Grails is emerging as a standard JVM web framework in environments ranging from startups to the enterprise. It's a full-stack solution build on rock-solid components, fully relying on convention over configuration, and using the best application language the JVM has yet seen: Groovy. This is the place to be for web apps on the JVM.

In this introductory talk, we'll get a whirlwind introduction to Grails, visiting seven things you need to know about the framework to get started.

1. Programming in Groovy
2. Creating an app
3. Interacting with the database
4. Building your UI
5. Processing web requests
6. Tapping the huge plugin community
7. Deploying to production

Database Refactoring with Liquibase

Most teams manage database change using an ad-hoc system of SQL migration scripts manually applied to various development, staging, and production servers. Some even contrive automated processes, but rarely does this surplus build engineering deliver value directly to the customer. We should be writing applications, not build tools.

In this session, we'll take a look at a ready-to-use, open-source database refactoring tool called Liquibase. Liquibase enables developers to make database changes with confidence, share those changes in a predictable way with other team members, and apply them to automated QA builds, staging servers, and production environments. It provides a credible path to agile database development, and it integrates well into popular build tools. It's a key enabler of the culture of database responsibility that most teams are missing.

Database Refactoring Workshop

Take one ugly legacy schema, a toolbox full of simple database refactorings, and a world-class schema refactoring tool, and you've got 90 minutes of workshop that will equip you to bring a culture of database responsibility to your team.

In this workshop, we'll start with a live schema in need of some help, and slowly improve it in a controlled fashion using Liquibase. We'll see how to create and alter tables, add constraints, drop columns, control changes in stored procedures, and more. You should come away with a solid understanding of how to use the tool and how to integrate it into your team's development, build, and deployment processes.

Attendees should already have a conversational understanding of Liquibase, or have attended Scripting the Schema with Liquibase session. Please bring a laptop or be prepared to pair with a friend.

Cassandra: Radical NoSQL Scalability

Want to go deep on a popular NoSQL database? Cassandra is a scalable, highly available, column-oriented data store in use at Netflix, Twitter, Reddit, Rackspace, and other web-scale operations. It offers a compelling combination of a rich data model, a robust deployment track record, and a sound architecture, making it a good choice of NoSQL databases to study first.

In this session, we'll talk about Cassandra's data model, look at its query idioms, talk about how to deploy it, and look at use cases in which it is an appropriate data storage solution. We'll study its origins in the Amazon Dynamo project and Google's BigTable, and learn how its architecture helps us achieve the gold standard of scalability: horizontal scalability on commodity hardware. You'll leave prepared to begin experimenting with Cassandra immediately and planning its adoption in your next project.

A Thorough Introduction to Gradle

Gradle. Another build tool? Come on! But before you say that, take a look at the one you are already using.

Whether your current tool is Make, Rake, Ant, or Maven, Gradle has a lot to offer. It leverages a strong object model like Maven, but a mutable, not predetermined one. Gradle relies on a directed acyclic graph (DAG) lifecycle like Maven, but one that can be customized. Gradle offers imperative build scripting when you need it (like Ant), but declarative build approaches by default (like Maven). In short, Gradle believes that conventions are great -- as long as they are headed in the same direction you need to go. When you need to customize something in your build, your build tool should facilitate that with a smile, not a slap in the face. And customizations should be in a low-ceremony language like Groovy. Is all this too much to ask?

Gradle has received the attention of major open source efforts and has chalked up significant conversions by the Spring Integration, Hibernate, and Grails projects. What do these technology leaders see in this bold new build tool? They see not only a better way to build Java applications, but an extensive ecosystem of connecting to existing Ant and Maven build files while expanding the horizon of test, CI, and deployment automation in an easy manner. Join us for 90 minutes and let us take you on this same walk of discovery of the most innovative build tool you've ever seen'.

Lightweight Web Apps with Ratpack

Ratpack is a hyper-lightweight, Groovy-based web framework for developing and deploying simple apps in a hurry. Like its high-achieving cousin Gaelyk, it provides Groovy developers with a way to create web apps without days of iteration zero setup time.

In this talk, we'll look over Ratpack's very simple structure and live-code a small, practical example application. We'll look at how to evolve simple controller logic, how to manage templates, how to persist data, and how to deploy Ratpack applications to the web. The Java world needs ways to build small applications in a hurry, and Ratpack is the latest way to do it!

Clojure Workshop (Bring a Laptop)

Clojure is earning attention as one of the most innovative languages of the JVM in use today. It draws the attention of elite developers with features like software transactional memory and its commitment to functional programming paradigms, and has clean Java interoperability besides. But beyond the buzzwords is a syntax that many developers find intimidating. There is no reason for this to be a barrier!

In this hands-on workshop, you'll write a series of small Clojure programs to solve ordinary programming problems like counting list elements, processing strings, and querying databases. Your instructor will explain each exercise to you in advance, giving you a chance to code it the provided LabREPL before showing you the solution. Bring a laptop, and be prepared to pair with a friend! You'll see that Clojure is not as hard as it might look.

NoSQL Smackdown 2012

Alternative databases continue to establish their role in the technology stack of the future—and for many, the technology stack of the present. Making mature engineering decisions about when to adopt new products is not easy, and requires that we learn about them both from an abstract perspective and from a very concrete one as well. If you are going to recommend a NoSQL database for a new project, you're going to have to look at code.

In this talk, we'll examine three important contenders in the NoSQL space: Cassandra, MongoDB, and Neo4J. We'll review their data models, scaling paradigms, and query idioms. Most importantly, we'll work through the exercise of modeling a real-world problem with each database, and look at the code and queries we'd use to implement real product features. Come to this session for a thorough and thoroughly practical smackdown between three important NoSQL products.

ClojureScript

Clojure has recently been gaining attention as one of the most innovative languages of the JVM in current use, and it has mostly found a home on the server. In parallel, JavaScript has ascended to the position of the most important language of the web, and until recently it has lived only on the client. Few observers looked at the world of web development and predicted that these two would get together. Happily for us, they have!

ClojureScript is a dialect of Clojure that compiles to JavaScript, and targets the JavaScript runtimes of the web as a deployment environment. It offers the unparalleled expressiveness of Lisp, the performance and space efficiency of the Google Closure Compiler, interoperability with the in-browser object model, natural integration with server-side Clojure applications, and an insanely productive in-browser REPL. In a time of proliferating JavaScript extensions and client-side development frameworks, you owe it to yourself to take a look at this compelling vision of how client-side web development should be done.

Connected Data with Neo4j

Neo4j is an open-source, enterprise-class database with a conventional feature set and a very unconventional data model. Like the databases we're already used to, it offers support for Java, ACID transactions, and a feature-rich query language. But before you get too comfortable, you have to wrap your mind around its most important feature: Neo4j is a graph database, built precisely to store graphs efficiently and traverse them more performantly than relational, document, or key/value databases ever could.

Neo4j is an obvious fit to anyone who thinks they have a graph problem to solve, but this is not many people. It turns out that the most interesting property of Neo4j is its architectural agenda. It wants you to think of the entire world as a graph—as a set of connected information resources. Steeped in the thinking of resource oriented architecture, this NoSQL database wants to change the way you look at your world, and unlock new value in your data as a result.

Hadoop

When you want to measure fractions of a millimeter, you get a micrometer. When you want to measure centimeters, you get a ruler. When you want to measure kilometers, you might use a laser beam. The abstract task is the same in all cases, but the tools differ significantly based on the size of the measurement.

Likewise, there are some computations that can be done quickly on data structures that fit into memory. Some can't fit into memory, but will fit on the direct-attached disk of a single computer. But when you've got many terabytes or even petabytes of data, you need tooling adapted to the scale of the task. Enter Hadoop.

Hadoop is a widely-used open source framework for storing massive data sets in distributed clusters of computers and efficiently distributing computational tasks around the cluster. Come learn about the Hadoop File System (HDFS), the MapReduce pattern and its implementation, and the broad ecosystem of tools, products, and companies that have grown up around this ground-breaking project.