# AiiDA Team Meeting 2024-08-08
###### tags: `team meetings`
###### time: 14:00 CET
[TOC]
### Present
* Edan
* Ali
* Alex
* Seb
### Catch-up round
*Max. 3 minutes each*
Seb
* To be added
Edan:
* Workgraph notes
* Submission of workgraphs raises a "missing entry point" warning
* Per Seb, this may be ignored for now and will be resolved when `aiida-workgraph` is integrated into `aiida-core` (if that is still the plan - see general discussion below), at which point, it will be registered under an entry point
* Xing to add an entry point in the `aiida-workgraph`
* An input/output object in my workflow is treated as a `GeneralData` type. The workgraph takes care to (un)pickle it when needed
* A step in my workflow was excepting due to a failed unpickling of this data, the reason being a missing `readline` method on the file handler class (`node.outputs.retrieved.base.repository.open("results.pickle"`)
* Xing was unable to reproduce the error on a fresh install of the package
* Turns out Xing's file handler was of `io.BufferedReader` type, while mine was of `PackedObjectReader` - a `disk-objectstore` type which does not implement `readline`
* See `disk-objectstore` issue [#174](https://github.com/aiidateam/disk-objectstore/issues/174)
Alex:
* Started to work on `aiida-workgraph`, but left discussion open for next time
### General discussion
* What are the plans for AiiDA beyond v2.6?
* How should we handle `aiida-core` + `aiida-shell` + `aiida-workgraph`?
* Options (non-exhaustive!)
1. All integrated into `aiida-core`
2. Kept as separate dependencies of a "new" `aiida` package (jupyter model)
* What to do with the website?
* Move to React?
* Requires:
* discussion (some lively already)
* tech survey (Edan to present survey at the 29.08.2024 MSD group presentation)
* How to publish/maintain?
* Integrate/unify documentation
* Considering above point regarding yes/no to integration in `aiida-core`, how would documentation of dependencies be handled?