Problem Statement
When users copy content from Microsoft Office applications (such as Excel cells or PowerPoint objects), these applications often place multiple formats onto the clipboard simultaneously: the plain text/values and a graphical capture (image) of the selected area.
Currently, Glean Assistant automatically accepts both the text and the image when pasted into the chat. This leads to the following issues:
- Increased Latency: The system triggers image-reading/multimodal processing for the unintended attachment, which noticeably slows down the overall response time.
- Operational Friction: Frequent users feel forced to manually click the "X" button to delete the image every time they paste, just to ensure a faster and text-centric response. This repetitive manual intervention creates a significant burden for power users.
- Limitations of Workarounds: While we are aware that using the "Paste as plain text" shortcut (Ctrl + Shift + V) can bypass this, it is not a fundamental solution. Relying on a specific keyboard shortcut for every paste action adds mental overhead and is not an intuitive experience for all users. We believe the system should handle this intelligently by default.
Proposed Solution
While this is partly due to the default behavior of Office products, I request a system-level control within Glean to manage simultaneous pasting. Specifically:
- Smart Paste Logic: Implement a mechanism to prioritize text when both text and an image are detected in the clipboard from the same source.
- Toggle Setting: Provide a user preference setting such as "Ignore images when text is present in clipboard" or a "Paste as Text Only" shortcut.
- Prompt for Selection: A brief UI hint asking if the user wants to "Paste text only" or "Paste with image" when multiple formats are detected.
Expected Benefits
- Seamless Workflow: Users can quickly move data from Excel/PowerPoint to Glean without the constant need for manual cleanup.
- Performance Optimization: By avoiding unnecessary image processing, the system can provide faster, text-based answers, reducing server-side load and improving user satisfaction.
- Enhanced Accessibility: A cleaner UI experience that caters to users who primarily interact with structured data and text.
