book

Breaking Data Science Open

by Michele Chambers, Christine Doig, Ian Stokes-Rees

May 2017

Beginner to intermediate

86 pages

1h 56m

English

O'Reilly Media, Inc.

Read now

Unlock full access

Open Source Software and Benefits of Open Data ScienceThe Future of the Open Data Science Stack
Recursion PharmaceuticalsTaxBrainLawrence Berkeley National Laboratory/University of Hamburg
Dynamic, Not Static, InvestmentsData Lab EnvironmentTeam Management, Processes, and ProtocolsData Services and Data Lifecycle ManagementInfrastructure and Infrastructure OperationsExecutive Sponsorship ResponsibilitiesGovernanceProvenanceReproducibility
TeamTechnologyMigration
What the Open Data Science Community Can Do for YouThe Power of Open Data Science LanguagesWhy R?Why Python?Why Scala?Why Julia?Established Open Data Science TechnologiesNotebooks and NarrativesHadoop/SparkAnacondaH2OpandasScikit-LearnCaretShinyDaskEmerging Open Data Science Technologies: Encapsulation with Docker and condaOpen Source on the Rise
How to Bring Open Data Science to the EnterpriseGovernanceCollaborationOperationsBig DataMachine Learning and Artificial IntelligenceInteractive Dashboards and AppsSelf-Service Analytics
How Collaborative, Cross-Functional Teams Get Their Work DoneData Science Is a Team SportCollaborating Across Multiple ProjectsCollaboration Is Essential for a Winning Data Science Team

Self-Service Data ScienceMeet Me Where I AmMake It Dead SimpleMake It Intuitively ObviousSelf-Service Is the Answer—But the Right Self-Service Is Needed
What Data Scientists and Developers Bring to the Deployment ProcessThe Traditional Way to DeploySuccessfully Deploying Open Data ScienceAssets to DeployProcesses to DeployOpen Data Science Deployment: Not Your Daddy’s DevOps
Models As Living, Breathing EntitiesThe Data Science LifecycleBenefits of Managing the Data Science LifecycleData Science Asset GovernanceModel Lifecycle ManagementThe Champion-Challenger ModelGetting Over Hurdle RatesOther Data Science Model Evaluation RatesKeeping Your Models Relevant

Content preview from Breaking Data Science Open

Chapter 5. Data Science Executive Sponsorship

In the previous chapter, we’ve learned why Open Data Science matters. It has the potential to dramatically transform the way organizations pursue critical strategic initiatives, as well as the way they track and measure success.

These enabling characteristics of Open Data Science, coupled with its widespread grassroots adoption, have a downside. Individuals within an organization can now make technology decisions that do not have up-front costs and can therefore bypass the established review processes—covering technical suitability, strategic opportunity, and total cost—that have traditionally been required of proprietary technology. Even worse is that different individuals may make conflicting technology decisions that come to light only once projects have made significant progress. Thus, to ensure that the short- and long-term business outcomes of adoptixng Open Data Science are aligned with the company’s strategic direction, executive sponsorship is essential.

This might sound like the typical IT-projects-need-executive-sponsorship soapbox. But keep in mind that we’re talking about making room in the enterprise IT landscape for a new world where Open Data Science connects with new and existing data to inform everything from day-to-day micro decisions to occasional strategic macro decisions. Open Data Science introduces new risks that are mitigated by appropriate executive sponsorship: