DFG survey 2026 #19
Loading…
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
What should be done to prepare the survey application for the 2026 round?
Current state:
shacl-vue, including:update_shapesandproperty_shapesto group and order the fields in theContributorformdumpthingsbackend is still running an older versionWhat would be the ideal state?
The UI should be as intuitive, quick, and easy as possible for both new and existing users, i.e. users that completed the form before and those that haven't.
dumpthings:We likely want to upgrade to the latest dumpthings version. A categorization that mixes things up is whether users have a forgejo account or not. The ideal is for everyone to have one, so that they can use that for authentication and token retrieval for UI record submission (Q: does this also imply the need for a recent version of
dumpthings?). Changing tokens (or new tokens) will lead to a challenge for existing user data: their previous data will need to be migrated to a new inbox (on a newdumpthingsdeployment) for which we don't have or don't know a name yet.shacl-vue:When it comes to UI, we need to make use of
shacl-vue's wizard feature (and possibly custom plugin feature) to simplify survey submissions. The simplest submission will be of the kind"I already submitted a survey in 2025, nothing changed since then, let me resubmit with as few clicks as possible". For that we can implement a record-based wizard (a button that will show when a user edits their existingContributorrecord) that would e.g. just add an annotation to the effect of"Same as last year". And that is the only change that such a user would save and submit.If a user has not submitted anything before, or submitted something before that has now changed, a wizard+plugin approach could be used (instead of the current default
Contributor+ConsortiumParticipationforms) which would allow some streamlining of form-filling logic. The known improvements include:"(In the case of a change of status group or funding type since the last survey, please provide the relevant details here)"I think using a wizard+plugin would make such logic more intuitive, but the effort might not be worth it.
schema:I am not sure whether updating the schema is ideal or not. It will certainly allow more compatibility with the TRR379 research-information schema, and hence useful data transfer to avoid unnecessary duplication. E.g. if we already have most (all?)
TRR379Personrecords for researchers in the consortium, those could be used to prefillContributorsin the DFG survey app.It will also be useful to not have to maintain older schema versions purely for this deployment. However, updating the schema also implies a possible redesign of the relations between classes, the inclusion/exclusion of classes/properties, and also data migration. This is not trivial.
What should we do?
What we do next depends on our chosen approaches for the different components. Options include (not mutually exclusive, not all optional, not all required):
dfg-cp-surveytoken-users to known forgejo account namesdfg-cp-surveyContributorrecords toTRR379Personrecordsthings-v2and related schema componentsall still the sameannotation wizardWith regards to:
I made a local implementation of this, with the simple template:
and config:
Demo video:
In the process I also implemented the markdown rendering feature for https://hub.psychoinformatics.de/orinoco/shacl-vue/issues/386