Keming Labs Talks

I was not planning on speaking at this conference, but the organizers asked me to fill in for a last-minute speaker cancellation---this talk is essentially a repeat of the Øredev talk. --kevin

As programmers, we're used to seeing data in rectangular tables that are optimized for fast retrieval and processing by computer. Pictures of data, such as scatterplots, bar charts, and maps, optimize data for efficient analysis by human. SQL is powerful because it allows us to ask complex questions of our data without busying ourselves with the mechanics of iteration, aggregation, and indexing. We need the same for pictures: a grammar that allows us to express rich data visualizations without the nuisance of looping, drawing axes, and juggling legends.

In this talk, I will introduce such a grammar of graphics, implemented in Clojure. As motivation, I'll discuss the principles of effective data visualization and the insights that can come from just looking at your data. The grammar itself consists of simple data structures, maps and arrays. As a first consequence, this means the grammar can be easily used across the JVM as well as via JSON. However, there are several other, deeper ramifications of having a "data API" that will be discussed.

As programmers, we're used to seeing data in rectangular tables that are optimized for fast retrieval and processing by computer. Pictures of data, such as scatterplots, bar charts, and maps, optimize data for efficient analysis by human. SQL is powerful because it allows us to ask complex questions of our data without busying ourselves with the mechanics of iteration, aggregation, and indexing. We need the same for pictures: a grammar that allows us to express rich data visualizations without the nuisance of looping, drawing axes, and juggling legends.

In this talk, I will introduce such a grammar of graphics, implemented in Clojure. As motivation, I'll discuss the principles of effective data visualization and the insights that can come from just looking at your data. The grammar itself consists of simple data structures, maps and arrays. As a first consequence, this means the grammar can be easily used across the JVM as well as via JSON. However, there are several other, deeper ramifications of having a "data API" that will be discussed.

Panel with Kim Rees (Periscopic), Hadley Wickham (Rice University), and David Nolen (New York Times).

Advances in browser and mobile technologies have made the visualization and interaction of data on web a viable alternative to traditional tools used to visually explore data. Panelists will discuss the current state of web data visualization, as well as novel approaches made possible by recent advances.

Kevin will review three of Keming Labs commercial visualizations from the past year: a wind turbine monitoring platform, a gene variant analysis tool, and an iPad diabetes clinical reporting tool. With this context, Kevin will then wax philosophical on emerging programming languages, visualization on the web, and the role of open source software in the commercial and academic spheres. Finally, in what may be of interest to students, Kevin will also detail his transition from VisWeek 2010 workshop paper third author to full-time world-traveling high-fiving charts & graphs consultant.

Modern web browsers provide a versatile foundation for graphical, interactive applications. ClojureScript brings rich semantics to the browser and, combined with Clojure on the server, gives developers an insane amount of leverage to build novel tools. Kevin will discuss using state machines with data-driven DOM-manipulation to build visual applications on the web and iPad. He will evince how Clojure(Script)’s rich data structures and abstractions foster declarative programming, and outline some testing and development benefits from shared client/server language. Concrete examples will be drawn from data visualization applications in the energy and biotech fields.

Statistical graphics illuminate by mapping abstract data to visual primitives like dots, lines, and color. Excellent statistical graphics first and foremost show the data. Likewise, the tools for making such graphics should emphasize essential data-visual mappings and hide the implementation details of looping, painting, and layout.

Clojure's emphasis on immutability allows one to compose functions while retaining the ability to reason about their behavior. These same strengths can be leveraged to build rich, data-driven graphics composed of smaller visual pieces. Kevin will discuss using ClojureScript with the Cassowary constraint solver to construct and layout statistical graphics on the web.

Invited talk to Reed College. Kevin will discuss his experiences consulting in the design and programming fields and the pros and cons of self-employment. The talk will include an extended Q & A session for interested students.

Kevin will discuss using the D3 JavaScript data visualization library from ClojureScript. In particular, using macros to construct a ClojureScript façade that implicitly coerces types and allows one to naturally extend jQuery-style chained syntax. He will also discuss miscellaneous ClojureScript tips and rough edges, and compare ClojureScript with CoffeeScript and more traditional JavaScript functional-mixin libraries like Underscore.js.

Introduction to the newly released Clojure–to–JavaScript compiler.

D3 is a JavaScript library that binds arbitrary data to HTML & SVG elements, allowing you to build rich, interactive visualizations on the web. Rather than try to serve as a monolitic library supporting every visualization imaginable, D3 simply provides a lightweight, data-driven interface to the web's underlying HTML, CSS, and SVG technologies. Developers can use D3 immediately with their current development tools and knowledge of web standards.