# Feature OP-18120 - Improve perceived performance of project selector

Project: OpenProject
Status: closed
URL: https://community.openproject.org/work_packages/OP-18120

## Planning

| Attribute | Value |
| --- | --- |
| Date |  |
| Priority | Medium |
| Target versions | 17.0.0 |
| Sprint | 17.0.0 |
| Backlog bucket |  |
| Votes |  |
| List |  |
| Roadmap |  |
| Relevance |  |

## OPEN POINTS / FEEDBACK

_No work packages_

## Features and Chunks

_No work packages_

## Bugs and other work packages

_No work packages_

## Technical Details

| Attribute | Value |
| --- | --- |
| Module | Modules and project features / Project list |
| Feature Flag |  |
| Requires doc change | No |
| Enterprise plan | Community (COM) |

## Details

| Attribute | Value |
| --- | --- |
| Pull preview |  |
| Mockups |  |
| Accessibility | No |

## People

| Attribute | Value |
| --- | --- |
| Assignee |  |
| Accountable |  |
| Product manager |  |
| Feature Lead |  |
| Developers | Jan Sandbrink |
| Designer |  |
| Ordered by |  |
| Signed off by |  |
| Interest by |  |

## Estimates and progress

| Attribute | Value |
| --- | --- |
| PD (min) |  |
| PD (max) |  |
| Work |  |
| Labor costs |  |

## Description

**As** a user  
**I want to** load fewer projects into the project selector and load them asynchronously,  
**so that** the project selector doesn't feel sluggish when dealing with a large number of projects.

**Acceptance criteria**

  - Only load 300 projects at once

  - Entering a search term, will search across all projects and show up to 300 of them

  - When opening the project selector while in a project, the selector opens with the given project selected
    
      - we ensure that up to 300 descendants are visible as well (depth-first)
    
      - we also ensure that up to 300 siblings of the current project are visible as well

**Background**

The current behaviour of the project selector is as follows:

1.  Show a skeleton loader, as long as loading projects has not finished

2.  Load **all** projects (even if it's ten thousands and multiple network requests are necessary)

3.  Perform all filtering locally
    
      - which can feel sluggish for many projects as well
