OpenAtlas

Thanks for the quick response!

The issues with this issue ;-) are two (+ subissues):

A)
1. that our "classes" like place, feature, find etc. are defined by a system type.
2. The type field is a unique one for each of these entities. E.g. "Place" or subtype of "Place" for what we defined as places.
3. If the user does not enter a type, it is still a place but has no "type" that defines him as a place even though in theory it is one.
4. If you query for entities that are exactly linked to the "Place" type, you get no results.

B) As project manager with multiple users entering data it would be very beneficial if not even necessary to "force" users to add data to certain project relevant fields

So, in my opinion there are technical and practical reasons for that.

About the first issue: query. Not sure if I understand what you mean by "it is still a place but has no "type" that defines him as a place". A place has the code E53 (paired with a physical object E18) if you are interested if this is a place, feature or strategraphic unit you can look at the system type. This sounds more like a how to query question.

About the second issue, here another example:

• Ignore it and leave it as it is
• Guess types
• Track down the user who created the entry and discuss it with that user

None of these options support an environment where users care about each others data.

If certain types are really that important this sounds more like an issue for project management. You need project specific standards per project for data entry like e.g. which language to use which can't be enforced either and has to be done by convention.

However, this may be a topic we should discuss in the team. I'm meeting Christoph and Berni tomorrow, Thursday, and you are welcomed to join.

I now remember, we had this discussion some time ago. Please notice that there is the system type (place, feature, ...) as database field in the entity table. This is not the same as the standard type which is used in forms and can be empty.

We changed naming for all the types sometimes so that can be confusing ;)

Discussed this feature again with Stefan today and we decided to put in on the agenda for next meeting.

Update:
like already said we will discuss setting required fields at a meeting with the whole team. But I like to separate this topic from the "how to know which entity is what" question. I realized it's not that easy and not documented (except in the code itself) so I created a wiki page for that: OpenAtlas_and_CIDOC_CRM_class_mapping and hope that it helps to clear things up a little. If there are still issues with the CIDOC - OpenAtlas class mapping please open another issue for that so that we can tackle these two complex questions separately.

Thanks for considering this. I opt for:
"Make it required only when inserting an entry so that former entries can be updated even if the required type is missing"
+1

When everything works as planned it would be great if required types would be on top of the types list (underneath name and alias) to make them easier to spot and fill in.

It wasn't possible for me to rearrange types according to their required status like requested from Nina. If this has a higher importance I would like to ask you to make an own new issue for this. Maybe Andi can solve it.

I was able to find and fix the issue with administrative units.
So showing untyped and entities and setting the place types as required now works too. I already pushed to the THANADOS online version but be advised: checking untyped and/or making administrative units (or historical places) takes a long time on that instance.