The explosion of interest in data science in recent years has led to a resurgence of programming languages, tools, and techniques that help scientists and researchers explore data-driven and scientific concepts and communicate their findings; however, few tools have allowed them to do this via web browser. In response, on Tuesday, Mozilla unveiled Iodide, an experimental tool aimed at helping data scientists write interactive documents using web technologies, within an already familiar, iterative workflow.
Iodide acts as a programming environment for creating living documents within the web browser, removing friction from these workflows by bundling the editing tool with the clean readable document, according to a Mozilla blog post. This allows workers access to both a clean document to show non-technical audiences, and easy access to the underlying code for other data scientists–differentiating the tool from IDE-style environments, which output documents like .pdf files that are removed from the original code, and cell-based notebooks that mix code and presentation components, the post said.
SEE: Big data policy (Tech Pro Research)
The tool offers scientists a “report” in the form of a web page they can fill with their content, along with tools for exploring data and modifying the report until it is ready to share. Then, they can send colleagues a direct link to the finalized report. If those colleagues want to make changes to the code, they can do so from that link, or can create another version to make changes.
Iodide is still in an alpha state, and the announcement represents an early soft launch for Mozilla to gain feedback from the data science community. While it is usable, this release should not be used for critical work, the blog post cautioned.
In its current simple backend state, workers can use Iodide to save their work online, look at work done by others, and fork and extend notebooks made by other users. Mozilla hopes to add in Google Docs-style comment threads and simultaneous editing, and the ability to suggest changes to another user’s notebook, similar to Github pull requests.
To learn more about how to become a data scientist, check out this TechRepublic cheat sheet.