Report of the 26st TREE CG Meeting

Hi all,

Thanks for a very interesting TREE CG Meeting today!

*Decisions:*

  1. The next TREE CG meetings will not happen strictly on the first 
Wednesday of the month due to agenda constraints. The pace of once per 
month is kept. Check the agenda 
(https://calendar.google.com/calendar/u/0/embed?src=4b04fcfb0103a4f96d703b8b8255f15378d5f4729b9083bfc0129b676ae783d1@group.calendar.google.com&ctz=Europe/Brussels) 
- the *next one is the 14th of May*

**2. PR 123 solves multiple issues: 1) ambiguity in the spec on 
tree:view; and 2) making sure a collection IRI after redirects can also 
be used in the initialization. The biggest issue is that the discovery 
spec itself is not solved yet and it was argued that correcting the 
sentence in the overview to reflect this would make the spec consistent 
again, until the discovery spec is finalized in its own pace. Pieter C 
proposed a different approach, as we’re working with a living document 
at this moment for the discovery part, and proposed to, instead of 
changing the overview sentence, that we actually move selecting a 
rootnode based on the `tree:view` property as a first idea in the

  3. PR 116 on relation to SPARQL comparators (Bryan-Elliott) can be 
merged on the condition that the minor comments that were left are solved.

  4. PR  125 The TREE profile: we will first investigate whether and how 
we will allow adding relations at the end of the page before merging the 
PR. This can be discussed on the mailing list (and accepted for merging 
in case of consensus) or during the next TREE CG meeting, depending on 
how fast this progresses.

  5. For the next meeting on the 14th of May, we will prioritize Thomas 
Bergwinkl’s issue on SHACL relations as a discussion point as we already 
postponed this one twice.

Then we moved to *updates from the work items:*

  1. On the *discovery spec*, Julián started collecting use cases within 
the DiSHACLed project, where there’s a work package on discovery from 
DCAT-AP data portals using SHACL. The use cases are collected here: 
https://github.com/DiSHACLed/discovery-specification/issues - and you 
can add your own, even if you’re not participating in the project!

  2. On the *test suite: *tests were added since last time and comments 
from Ieben were processed. Once we have completion of a test suite of 
the full spec, we will notify the group again and request a review! 
https://github.com/TREEcg/specification/pull/120

We also then continued to the *core discussion* on the *Member 
Extraction Algorithm* (continued) and proposed to re-discuss the 
requirements and really scope it to TREE Web APIs, instead of trying to 
solve member extraction for any possible Web API out there. In that 
sense we came to the possibility of having 2 different ways: 1) the TREE 
MEA, which works specifically for TREE, is relatively easy to implement 
and contains what we want to promote for server providers; and 2) a 
Generalized MEA, which can work on any type of API. Rediscussing the 
requirements for the TREE ecosystem specifically is however a good next 
step for making further decisions.

Kind regards,

Pieter

Received on Wednesday, 2 April 2025 15:16:36 UTC