Summary: This epic tracks a collection of tickets we need to implement before releasing an RC of Node Driver V4. Cost in Eng Weeks: Bucket Epic for tracking purposes - cost is fragmented and intermittent. Engineer(s): Neal Beeken, Eric Adum, Daria Pardue, Durran Jordan 2020-06-30: Setting target date to 2021-07-09 The team is chewing through the small remainder of tickets in this epic. The team has closed 5 tickets, is currently reviewing 2 with a 3rd ready for review, and otherwise there is a single ticket that sits high in our scheduled queue to create an upgrade guide for V4. Cost in Eng Weeks: Bucket Epic for tracking purposes - cost is fragmented and intermittent. Engineer(s): Neal Beeken, Eric Adum, Daria Pardue, Durran Jordan 2020-06-16: Setting target date to 2021-07-02 Ran into some issues with tricky existing error code but now actually finishing up error type standardization in the driver ( NODE-3291 ) BSON-EXT support reintroduced to 4.0 ( NODE-2944 ) Modeling query types and brought in community types test (NODE-3297) Update modeling blocked but ready after this ( NODE-3274 ) Better types for mapped cursors ( NODE-3283 ) Merged all errors are ServerErr or DriverErr subclasses ( NODE-3291 ) Final reviews pending on merging typescript community type tests and corresponding fixes for incorrect/missing typings (NODE-3297) Cost in Eng Weeks: Bucket Epic for tracking purposes - cost is fragmented and intermittent. Engineer(s): Neal Beeken, Eric Adum, Daria Pardue, Durran Jordan 2020-06-02: Updating target date to 2021-07-02 The team continues to make strong progress on this work, some recently completed highlights: Typescript generics work, users can parameterize their collections to be of a type. ( NODE-3115 ) Typed Eventemitter, users can get hinting on their event listeners. ( NODE-3132 ) Port a monitor timing change to 4.0 ( NODE-3197 ) Enum export standardization ( NODE-3275 ) Remove callback/strict mode from Db.collection ( NODE-2752 ) There are a number of tickets in review including reintroduction of bson-ext support in v4 ( NODE-2944 ), strongly typed cursors with transform functions ( NODE-3283 ), and use of community type tests (NODE-3297). While the team has been able to direct more focus onto these tickets in recent weeks, we still have Load Balancer support for 3.7 and 4.0 open and we need to shift some attention to backporting Versioned API functionality to 3.7. Our prioritization of these tasks is Versioned API in 3.7 > Load Balancer in 3.7 > V4 RC Tickets > Several small tickets for Versioned API in 4.0 > Load Balancer in 4.0. Our goal is to cut a 3.7 release for essential 5.0 compatibility as well as a V4 RC with all breaking changes & essential 5.0 compatibility as soon as possible. Any non-essential 5.0 compat work for exposing new functionality can be iteratively released in 4.x minor versions to provide value to users as rapidly as possible. Timeline is generally starting to feel tight and we need to be hypercritical about what we are prioritizing. Cost in Eng Weeks: Bucket Epic for tracking purposes - cost is fragmented and intermittent. Engineer(s): Neal Beeken, Eric Adum, Daria Pardue, Durran Jordan 2020-05-17: Updating target date to 2021-06-04 Since serverless private beta and skunkworks, the team has been able to focus in on this work and is making great progress. The remaining implementation is in flight except for NODE-3272 and NODE-2752 , otherwise there is just some documentation followup to do. The team is planning to tag V4 as our 5.0 compat release and we are aiming to tag an RC by early June to try and garner some feedback from the community. We will be backporting some crucial 5.0 compatibility changes to 3.7 in order to allow users to leverage 5.0 serverless functionality without the need for a major upgrade of their driver. We are aiming to release a beta of 3.7 by mid-June. Cost in Eng Weeks: Bucket Epic for tracking purposes - cost is fragmented and intermittent. Engineer(s): Neal Beeken, Eric Adum, Daria Pardue, Durran Jordan 2020-05-04: Updating target date to 2021-06-04 Since serverless private beta and skunkworks, the team has been able to begin focusing more attention to these breaking changes. Aiming to wrap up load balancer so we can focus in as much as possible (other than user support tickets). Aiming to wrap up load balancer so we can focus in as much as possible, though we will also plan to pick up Avoiding Connection storms due to its high priority nature. The team is planning to tag V4 as our 5.0 compat release and we are aiming to tag an RC by early June to try and garner some feedback from the community. We will be backporting some crucial 5.0 compatibility changes to 3.7 in order to allow users to leverage 5.0 serverless functionality without the need for a major upgrade of their driver. We are aiming to release a beta of 3.7 by mid-June. Cost in Eng Weeks: Bucket Epic for tracking purposes - cost is fragmented and thus far intermittent. Engineer(s): Neal Beeken, Eric Adum 2020-04-06: Updating target date to 2021-05-07 With the serverless private beta release tagged, the team can begin directing more of their focus to this assortment of breaking changes for the 4.0 version of the driver. We have coordinated with the docs team and have decided that our GA release of 4.0 will be our Mongo 5.0 compat release in July. We will still target cutting RC versions ahead of time in the hope of receiving feedback from early adopters. The first RC date is less defined at this juncture, as we are just now able to start directing more undivided attention to this effort, but tentatively setting the end date to early May as an aspirational target. Cost in Eng Weeks: 5 Original | 3 To Date | 6 Est Final 2020-03-23: Updating target date to 2021-04-23 The team is unable to maintain pure focus on this epic due to high influx of user issues along with competing serverless private beta work. Following serverless private beta we will all aim to focus in on this epic. End date is a rough estimate as this is a bucket epic of many distinct tickets which have not been independently evaluated. Cost in Eng Weeks: 5 Original | 2 To Date | 6 Est Final 2020-03-09: Updating target date to 2021-03-19 Nothing much to report as the team has been focused on other high priority common drivers work and urgent help tickets. The whole team is planning to swarm on this collection of tickets after they finish Load Balancer, Unified Test Format, and Versioned API, all close to wrapping up. Cost in Eng Weeks: 5 Original | 2 To Date | 5 Est Final 2020-02-23: Updating target date to 2021-03-19 Neal has finished deprecating a number of items, but is otherwise pausing on this to finish Unified Test Format. The whole team is planning to swarm on this collection of tickets after they finish Load Balancer, Unified Test Format, and Versioned API, all set to wrap up in the next couple weeks. Pushing out the date as the aforementioned projects' timelines have expanded due to user support and significant review feedback. Cost in Eng Weeks: 5 Original | 1 To Date | 5 Est Final 2020-02-09: Setting target date to 2021-03-02 Neal has begun ticking things off of our deprecation checklist ( NODE-2317 ) while wrapping up Unified Test Format and waiting on reviews.