Creation Review

I strongly support the proposal to bring INCEpTION to the Eclipse Foundation. We used INCEpTION extensively in the GGPONC project, a German medical text corpus, which included the integration of custom ML models for annotation recommendations. The platform was very flexible and I was surprised how straightforward it was to adapt it to our project-specific needs.

It was also very easy to deploy and host on our university infrastructure, and the administration documentation was excellent. Whenever we ran into issues, bugs were addressed very quickly, and the frequent release cadence gave the project a strong sense of being actively maintained.

For projects that rely on annotation infrastructure over several years, that combination of flexibility, maintainability, and responsive development is extremely valuable. I would be very happy to see INCEpTION find a long-term home at Eclipse and think this would be a fantastic step for the project and its community.

I am a medical researcher at the University of Sydney. I run an Inception server to support my research. We use Inception to annotate clinical descriptions of rare diseases in the medical literature. The goal is to create a Gold Corpus of such annotations that can be used to train and evaluate LLMs to automate this task. Automated extraction of clinical descriptions from the medical literature would have many useful clinical applications. I evaluated many annotation tools for this task and only Inception fulfilled all my requirements. It has a great user interface, supports the most complex annotation schemes, handles large and complex medical ontologies, supports multiple annotators, has built-in recommenders, allows export in BioC format, the list goes on. In addition, Inception is well documented and Richard provides excellent support. With the rise of LLMs, a tool to manually curate text is more important than ever. AI is nothing without well annotated data. I wholeheartedly support this application to the Eclipse Foundation.

Hello, as an interested party, I strongly support this project. I have used INCEpTION extensively at my institution, and it has become an essential tool in my research and teaching.

• Institutional Use: I run an INCEpTION server at Leipzig University for my research group and for teaching purposes. I also regularly promote it in my communities, Digital Humanities and Classical Studies.
• Scientific Publications: I have used INCEpTION to prepare data for scientific publications, and I plan to do so again. See, for example, Berti, Monica. "Named Entity Annotation for Ancient Greek with INCEpTION." In Proceedings of CLARIN Annual Conference 2019, edited by Kiril Simov and Maria Eskevich, 1–4. Leipzig, Germany: CLARIN, 2019.
• Funding & Grants: INCEpTION is officially used in two of my funded projects: DFG project number 434173983 and the MSCA Doctoral Network MECANO project number 101120349.
• Teaching & Education: I use INCEpTION to teach BA/BSc and MA/MSc students in Digital Humanities and Classical Studies at Leipzig University. I have also used it in workshops for doctoral candidates in the MECANO Doctoral Network and presented it on various occasions in the Sunoikisis Digital Classics program (https://sunoikisisdc.github.io).
• Future Contributions: I have contributed to INCEpTION as a user and through scientific exchanges with its developers. As a digital humanist working with ancient languages, I am particularly keen on the current development directions: better support for structured HTML/XML-based documents, expanded functionality for the AI chat assistant, and broader support for ontologies along with better ways to work with them.

Hello, I am researcher at a university and our international team uses INCEpTION for tagging. It is a useful tool and I plan to submit a proposal for third party funding in which I will propose using INCEpTION because it is user-friendly and very useful at the same time. 

We strongly support the proposal to establish INCEpTION as a new project at the Eclipse Foundation to ensure its long-term maintenance and development.

As an institution offering INCEpTION as a service to our research community within CLARIN-IT, we have relied on INCEpTION as a robust platform that integrates exceptionally well with our infrastructure (such as our Vocabularies and Thesauri publishing services).

Currently, we maintain our service across two instances—our main instance and a dedicated project instance—totalling 58 projects and at least 39 registered users from multiple academic and research institutions.

The INCEpTION developers are always very reactive in their support and have greatly helped us make our service accessible via CLARIN Single Sign-On, which makes our instance easily usable by academic users. We are currently in a beta testing phase, but after the migration to a larger server, we plan to offer INCEpTION to a broader community. In particular, there is great demand for classroom use, where lecturers are less likely to set up a local instance.

We would like to be listed as an interested party representing CLARIN-IT / CNR-ILC.

I am writing on behalf of CLARIN:EL, which has offered an INCEpTION instance as a service to our research community for 7 years now. We fully support the proposal to move INCEpTION to the Eclipse Foundation, as we believe it will help secure the long-term maintenance and development of a tool that many of us rely on.

Our instance currently serves 134 users working in active annotation projects. They come from multiple countries across Europe and beyond, and they access the platform either through institutional accounts with federated login or through local accounts created in our repository. At present we host 117 active annotation projects containing 13,004 active annotations.

The platform has been used in a wide spectrum of activities from our part:

• Teaching: university classes and master's programs in lexicography.
• Research support: PhD candidates, whose work ranges from biomedical text annotation to hate speech, irony, stereotypes and intensity, sentiment analysis, and more.
• Outreach: public events such as the European Researchers’ Night, where we introduced novices of all ages to the world of NLP through hands-on annotation.

Given how broadly INCEpTION is used in our community, from education to advanced research, a sustainable, well-governed future for the project matters a great deal to us. We would be glad to be listed as an interested party in the proposal.

My team has been using INCEpTION for several years on in-house research projects to annotate medical documents. Over that time, we have observed the platform's rapid evolution.

We recently carried out a comparison of several annotation platforms. INCEpTION proved to be one of the most complete overall and, among fully open-source solutions, the richest in functionality.

Where most open-source annotation tools cover only part of the workflow, INCEpTION offers an exceptionally rich set of functionalities within a single application. It brings together, among other things:

• Flexible, user-defined multi-layer annotation schemes, configured directly in the browser.
• Entity linking against knowledge bases, with support for standard ontology formats and remote endpoints.
• Machine-learning recommenders that learn from the annotations made so far and take over the repetitive part of the work.
• Annotation of documents with their original layout preserved, notably PDF and, to a lesser extent, HTML, rather than only plain text.
  And the user keep the ability to switch between several annotation interfaces.
• Both annotation and curation within the same tool: several annotators can work independently on the same documents, and a curator can then review their annotations, resolve disagreements and merge them into a reliable gold standard.
• Fine-grained role management (e.g. project managers, annotators, curators) and integration with enterprise authentication, so the platform fits into an organization's existing identity and access infrastructure.

We agree with some improvement areas identified in the proposal: a better support for structured and layout-preserving formats, and richer interfaces to work with ontologies.