My apologies for not sharing earlier. Card's portrait mode designs are available in this new version of the Card exploration file in Figma.

While working on this version, I realized that, rather than thumbnails, portrait cards should probably use a different component: Image. Thumbnails should be reserved to provide small content previews, and they don't need to be adjusted to specific aspect ratios (such as 16:9, in this case). I would like to collect opinions and concerns, technical or otherwise, related to this approach: Should we create a new Image component? Should Thumbnail remain separate or exist as a (smaller) variant of Image?

Status update: Initial specs are ready, but there are some questions that should be discussed with the team in order to proceed, like the need to create an image component and the relationship of that element with Thumbnail.

In the context of using the machine translation form MinT to support Wikipedia readers (T359072), we are considering to use the portrait orientation of the card component.

The context is the following, the user has selected an article to read an automatic translation (e.g., translating the Moon from English to Korean). Although the user selection is the main option to proceed, we want to surface articles in the target Wikipedia that may already exist on the selected topic (to enable the user to access the human-created content instead). In this context, we want to represent the selected article in a visually prominent way with enough text translated from the article (for which the portrait version of the card seems a great fit), while the alternative can be presented in a less prominent way (using a card in landscape version).

This scenario is illustrated below in a mockup:

Thanks for sharing this use case @Pginer-WMF! What is the timeline for the MVP (T359072)? Design-System-Team is currently focused on a number of different work streams and it may take a bit of time to get to this. Would the Language team be open to upstreaming this change into Codex? I believe the engineers are already comfortable with Vue, so there might not be as much of a learning curve required.

In T310632#9601432, @CCiufo-WMF wrote:

Thanks for sharing this use case @Pginer-WMF! What is the timeline for the MVP (T359072)? Design-System-Team is currently focused on a number of different work streams and it may take a bit of time to get to this. Would the Language team be open to upstreaming this change into Codex? I believe the engineers are already comfortable with Vue, so there might not be as much of a learning curve required.

We want to make progress on the MinT for Wikipedia Readers MVP (T359072) during this fiscal year. I think it can be a good opportunity for the Language team to upstream a change to Codex. I guess we can use the landscape version as a temporary solution while the upstream version is built by the Language team. Some guidance and review may still be needed from the Design Systems team, but having a temporary solution may provide some more flexibility.

Queuing for refinement given the recent need in real use cases.