Early Access Priorities Poll

[cubap]

What priorities would help your project succeed? Vote below and add replies describing your test case or (even better) linking to real resources you would like accommodated. If you are a developer who would like to contribute, please reach out to coordinate efforts—we are very friendly.

Candidate Focus Areas

• Group Management expands the options for putting together your team and assigning roles. Possible enhancements would include granular controls over public access, organizing classroom groups, or moderated crowd-sourcing. Currently, these types of interfaces either do not exist or user specifications must be crammed into existing vendor offerings.
• Manifest Tools enable more complicated tools for building, combining, and manipulating custom manifests that can be imported into a project for transcription. There are several good options for building IIIF Manifests loose in the world, but none are well-integrated with transcription or permanently available on any open repositories by default.
• Image Analysis tools provide options for users who are attempting line or feature detection, especially tools that are fully automated. It may also provide for test cases where a page layout if complex but repeatable, so semi-automated processes like identifying type-A and type-B layouts makes line detection more accurate. This may include integration of existing third-party offerings (suggest yours favorite) and the creation of our own as standalone tools following Web Annotation standards.
• Annotation Tools improve the manual indication of lines and columns, identifying text flow, entity recognition within transcriptions, or feature tagging. This would be very driven by the test cases that the community can provide. We've designed our system to accept standard annotations, so there is a lot of room here for integrations.
• Specialized Transcription Interfaces supports new types of materials for transcription. Within our new ecosystem, anyone can create an interface that supports right-to-left text, specific languages or scripts, continuous scrolls, or even resources with no images. Our core application seeks to accommodate several high-value types of transcribable resources, based on the needs of our users.
• AI Integration encourages us to develop agent-readable instructions and APIs for use with external LLMs. As a free service, we have no plans to implement automation internally, but it seems irresponsive to not help our community connect their own services by providing tools and describing useful interactions.