Wikimedia Technology/Goals/2020-21 Q1

Wikimedia Technology Goals, FY2020–21, Q1 (July - September 2020)

Technology Department Team Goals and Status for Q1 FY20/21 in support of the Medium Term Plan (MTP) Priorities and the Annual Plan for FY20/21 (to be published)

MTP-Y2: Platform Evolution - Evolutionary architecture

Objective Manager: Corey Floyd

Objective

Software platforms with integrated machine learning, rich media, and structured data components, and associated tooling for internal and external development and reuse of code and content.

Key Results

Key Results Manager: Kate Chapman

KR1 Reached level 3 of the Architecture Maturity Model (AMM) for at least 85% of capabilities by Q4 (https://www.mediawiki.org/wiki/Wikimedia_Architecture_Team/Maturity_model)

Q1 Goal:

* Publish and begin using the architecture process so that teams are able to understand how to engage with architecture to better accomplish their goals.

KR2 Engineers have an improved understanding of current architecture and target architecture, enabled by the outputs of reaching level 3 for the AMM Architecture Development capability, "Gap analysis, modeling and iterative strategy for reaching a target architecture are completed”, measured quantitatively through surveys in Q1, Q2 and Q4.

KR3 Four (4) new service components in production delivering new capabilities, enabling and adopting industry standard best practices for architecture, engineering and deployment, allowing for the mitigation of risks for both development teams and operational stakeholders, building trust in our development processes. 1 service in by end of Q2, 2 by end of Q3, and 4 by end of Q4.

Q1 Goals:

* Define components of target architecture and present to technology leadership.

* Survey engineering in Product and Technology to obtain a baseline of system architecture understanding with a 65% response rate.

* Public architecture overview for target architecture and obtain support from senior individual contributor leadership and engineering management leadership.

MTP-Y2: Platform Evolution - Content Integrity

Objective Manager: Leila Zia

Objective

The Wikimedia movement has the tools and knowledge to identify and respond to abuse, misinformation, and disinformation campaigns, in order to more efficiently and effectively improve the quality of content, defend the projects and retain the public’s trust.

Key Results

Key Result Manager: Leila Zia

KR1: Develop 1 model to identify misinformation (by Q3) and deliver 4 milestones towards 1 model to understand the diffusion of content on the Wikimedia projects (by Q4).

Key Result Manager: Chris Albon

KR2: Complete 4 product integrations (e.g. link recommendations) of internally built machine learning models and services for improving content integrity and deploy 7 new community-driven models that extend our existing content integrity approach.

Q1 Objective: Deploy the next iteration of ORES as a scalable, production machine learning model management and deployment system.

Q1 Goal:

* Publish a strategic plan for ORES's next iteration

Q1 Objective: Support Scoring for external data loads into elasticsearch

Q1 Goal:

* Index external data requested by scoring into elasticsearch

Key Result Manager: Leila Zia

KR3: Coordinate and report quarterly on the disinformation activities across teams, establish a community support forum (by Q3), and participate in 10 external relationships (by Q4) including academia and industry partners.

ERF-Y2: Technology Dept - Front Line Defenses

Objective Manager: Mark Bergsma

Objective

Our infrastructure and data are staffed, secured and provisioned appropriately in each area to successfully prevent or handle malicious attacks, the unavailability of one system component, or the unavailability of a staff member.

Key Results

Key Result Manager: Faidon Liambotis

KR1: We will plan (by end of Q3), build (Q4) a second EU data center, to better serve EMEA users (by end of FY21-22 Q1).

Key Result Manager: Mark Bergsma

KR2: The number of staff members solely responsible for a system or piece of software in production is reduced by 50%, from 8 to 4, by end of Q4.

Q1 Goals:

* Start the hiring process for two current SRE SPOF IC positions to be hired in Q2

* Research and implement Knowledge Sharing Activities to reduce single points of failure by 50%

Key Result Manager:John Bennett

Objective: Implement a threat identification and risk treatment program (Q1) and deploy effective countermeasures for the top 2 threats and risks per quarter thereafter.

Q1 Goal:

* Implement Iterative Workflows Teamwide

Key Result Manager: Mark Bergsma

Objective: All canonical, non-derived project data in production can be recovered using the centralized, standard backup infrastructure

Q1 Goal:

* Database provisioning/backup service is well provisioned to accommodate for growing data, redundancy and performance needs

ERF-Y2: Technology Dept - Technical Community Building

Objective Manager: Birgit Müller

Objective

Our technical community is thriving and has a clear, consistent means to discover, build, and deploy applications that support community workflows, invent new forms of content creation and consumption, and leverage Wikimedia's APIs and data beyond the core wiki experience.

Key Results

Key Results Manager: Birgit Müller

KR1: Communities find the tools that they need through the new Toolhub catalog system (technical plan Q1, working prototype Q2, MVP Q4), and developers create more diverse solutions using a user-extensible, container-based default deployment process (working prototype Q3, running at least 1 workflow by Q4) in Toolforge.

Objective: Develop and implement the Toolhub catalogue system (MVP)

Q1 Goal:

* Define the technical plan for Toolhub

KR2: A 10% increase in number of tool maintainers (4% by end of Q2) speaks to a vibrant ecosystem of technical contributors, supported by an iterative model and practice of community and capacity building (Q1 draft, Q4 MVP) which has been refined through 3 initiatives in focused outreach to technical communities (Q1-Q4).

Objective: Develop an iterative model and practice of technical community and capacity building

Objective: [DRAFT] Initiate focused outreach to technical communities

Objective: [DRAFT] Increase visibility & knowledge of technical contributions, services and consumers across the Wikimedia ecosystem

KR3: An organization strategy for key technical documents informs a continuing roadmap based on a process of drafts (Q1) refined by consultation with staff (Q2) and community input (Q3) and a prototype of a single entry point to lower barriers finding existing documentation (Q4).

Q1 Goal:

* [DRAFT] Draft organization strategy, plan and process for key technical documents

ERF-Y2: Technology Dept - Production Quality

Objective Manager: Erika Bjune

Objective

Engineering teams at WMF have a shared understanding of development velocity, production health, and code quality, and they develop commitments and workflows for improving overall code health.

Key Results

Key Result Manager: Mark Bergsma

KR1: Evangelize, implement tooling for (by end of Q1) and define and implement Service Level Objectives and Error Budgets for our top 10 services and systems (2 in Q2, 4 in Q3, 4 in Q4) and report and iterate on them with product owners on a quarterly basis to optimize the balance of speed of innovation and reliability.

Q1 Goal:

* Evangelize and explain the concepts of SLOs and Error Budgets to product owners, technical teams and other stakeholders

Key Result Manager: Greg Grossmeier

KR2: Educate teams on and define Code Health Objectives for our production-deployed code repositories (20% in Q1, 40% in Q2, 70% in Q3, 100% in Q4) and report and iterate on them with code owners on a quarterly basis.

Q1 Goals:

* Facilitate 3 workshops on refactoring practices during Q1 2020

* Provide better understanding of TDD to anyone who contributes to WMF code.

Key Result Manager: Greg Grossmeier

KR3: Evolve and replace our Continuous Integration and Review systems with 1 project moved by end of Q2 and 5 additional projects moved by Q3. In parallel, 70% of Wikimedia developed application layer production service traffic is served by the Deployment Pipeline in Q3 and 95% by Q4.

Objective: Evolve and replace our CI system. Work to ensure out code review system is right for us.

Objective: Increase production traffic served by the Deployment Pipeline

Objective: Wikimedia deployments are safe, modern, and reliable

Objective: Wikimedia uses best-in-class collaboration and development tooling