🚨 Deprecation Notice: Orbeon Dropped eXist Support — Should MerMEId Explore a Replacement?

[imranasif87]

While investigating a bug reported here, we discovered that resolving the issue would require upgrading Orbeon Forms.

However, during our investigation, we came across this diagnosis here, which shows that the latest versions of Orbeon no longer support the eXist database (support deprecated in 2019 and officially ended in October 2021).

Given this, we would like to open a discussion:
Should we consider replacing Orbeon with an alternative solution that is actively maintained and compatible with eXist or another supported backend?

See: https://doc.orbeon.com/form-runner/persistence/db-support#older-database-support

[peterstadler]

I think it's fair to say that XForms is the most arcane technology in the whole MerMEId stack – and I realize that XML and XQuery is considered an outdated technology by some people. So, in terms of maintenance and sustainability it is highly welcome to replace XForms with something else.
The major problem is that those XForms are at the heart of MerMEId and more or less all its software logic is expressed in XForms.

If there were some (human or AI) resources to transform the current XForms what would be the target? Fore or some other Javascript framework?

That said, there's the MerMEIding project and there are people already working on something to update/replace MerMEId but I do not know its status.

[imranasif87]

Thanks for your feedback.

Yes, we’ve also been discussing Fore, a model-driven language inspired by the XForms 2.0 standard but adapted to work with modern HTML5 Web Components. One of the advantages is that we can easily Dockerize the two core services in our stack—eXist-db and the Fore web application—independently.

The main challenge we anticipate is the need to rewrite our existing forms using Fore’s Web Component syntax. However, once this migration is complete, we believe the resulting tech stack will be future-proof, maintain compatibility with eXist-db, be more lightweight and flexible, and benefit from active development aligned with modern web standards.

That said, there are potential downsides, such as a less mature ecosystem compared to Orbeon Forms, and limited documentation and community support at this stage.

Given our reliance on eXist-db, Fore seems to be the more suitable long-term solution. However, before making a full transition, we plan to develop a prototype application, including a CRUD interface for complex XML files, to identify and resolve any issues early on.

We’re seriously considering this direction and will continue the discussion in more detail.