DMG STRATEGIC PLANNING DMG Confidential 1 Problem 1

DMG STRATEGIC PLANNING DMG Confidential 1

Problem #1 � Development and maintenance � Huge demand for DMG services plus focus on short-term benefits led to shortcuts in code development � More time is now spent on maintenance and support activities than developing new features � Current technology stack and code base does not support agile development � Aging code base and technology stack is not adequate to meet current and future demands DMG Confidential 2

Problem #1 (cont. ) � DMG’s technical debt � New features take longer to develop � Testing takes longer � Software deployments take longer � System has become less stable � Extremely difficult to troubleshoot issues � Code base is “brittle” � New developers take longer to ramp up � Not an attractive job for developers DMG Confidential 3

Problem #2 � Scalability � Ingest and export processes not able to handle burst traffic loads � Exponential growth in storage usage and related costs � Peak sizing can result in excess capacity DMG Confidential 4

Scope of Complexity DMG Confidential 5

MCS DMG Confidential 6

Plan of Record with MCS � DMG looking to leverage MCS to obtain asset management back-end services and storage � Migrate cine. SHARE users/assets in phases � Then migrate EAGL to MCS-DMR in one shot � � DMG focuses on maintaining customer-facing applications Expected benefits � Better scalability and performance � Increased agility to create innovative solutions � Lower long-term storage costs � Cost savings from reduced headcounts (back-end services and infrastructure) DMG Confidential 7

Assumptions Necessary DMR services will be available in MCS solution � “MCS-DMR” services are functionally and architecturally adequate � MCS can meet new SPE feature requests in a timely manner � MCS charges SPE at cost � DMG Confidential 8

Impacts of Assumptions � Focused on short-term fixes to DMR rather than implementing complete fixes that would be “thrown away” once DMR was re-platformed on to MCS � Further contributes to technical debt � Deferred work on enhancements (e. g. review and approval) if those features were available or planned in MCS DMG Confidential 9

Concerns / Risks � � � “MCS-DMR” is built on the same legacy code base as EAGL Less than 50% of DMR services are enabled in current MCS solution MCS estimates for closing the DMR and cine. SHARE feature gaps are troublesome from a timing perspective Migration costs will increase due to complexity of a phased approach The “MCS-DMR” integration is a moving target - both sides are drifting since the branch of code Will MCS and SPE priorities stay aligned? DMG Confidential 10

Alternative Plan with MCS � Migrate client applications to MCS from both EAGL and cine. SHARE as and when required features are available � Attractive because avoids a big switch over and can be started soon � Complex because of dependencies between features and because of sharing of assets � Assumes Ci UI will work for migrated users � Use MCS for relevant new workflows, e. g. review and approval DMG Confidential 11

New Digital Media Platform DMG Confidential 12

New Platform Needed to continue DMGs mission efficiently Build a new digital media platform using modern web technologies, protocols and design practices � Target cloud deployment � Leverage open source/best of breed technologies � � � Ruby on Rails, Elastic. Search, Node. js, Mongo. DB � Benefits � � Timely response to customer needs Faster development Less maintenance Continuous Delivery ○ Continuous Integration/Automated Testing ○ Continuous Deployment DMG Confidential 13

Options Replace entire stack (DMR, Eagl, etc) 1. � Complete split from MCS, development burden falls on DMG Replace application layer, keep DMR (either DMG or MCS) 2. � � � Imperfect option, likely intermediate solution Constrained by existing DMR interfaces, likely need a services layer to isolate them Allows MCS to carry on with plan of record Explore with MCS a common path to a new platform 3. � � Does MCS see same issues with code base? E. g. DMG moves application layer to new platform, MCS moves DMR to new platform. DMG Confidential 14

APPENDIX DMG Confidential 15

New Platform: Architectural Map � Insert diagram DMG Confidential 16

Benefits of New Technology Stack � � � � Faster development – less lines of code to achieve functionality Easier to maintain – Ruby and JSON Easier to ramp up new developers Better support for open source tools Better integration for external apps Better design for future demands Support for continuous integration and deployment Enhances collaboration and innovation DMG Confidential 17