running on a free cloud account, we are a team of a few people and we want to test joint collaboration to create a single data model. Each of us has his own endpoint. What would be the best way to collaborate without stepping into each other toes, and to contribute to a single data model? Does one of us has to share their endpoint, and everyone else should work on that shared 'team' endpoint? Does each of us have to work in his own Designer to create a Named Graph and then do we need to join all Named Graphs in the final Graph? We have also seen that the shared endpoint allows others to see the data in Explorer, but not the data model in Designer... is there a way around that?
Hi Antonio. Would love to learn more - why is your team collaborating but using individual endpoints? While possible with Stardog, as we follow open standards and allow exports and imports of data models and mappings, for a team that wants to collaborate having multiple endpoints would add overhead.
Designer supports importing data models from an endpoint and options to publish a new data model or replace an existing data model. You could also take advantage of Stardog's ability to support multiple data models, and stage changes in a second model in your Designer project and publish a combo model of the existing model and the second model, therefore leaving the existing data model produced by a collaborator unchanged.
Alternatively, each collaborator could use their own database on the same endpoint making sharing and access easier.
If you haven't already, I highly recommend our Designer trainings on Stardog Academy and docs which go into further detail.
Thanks Laura, ok, so recommendation is for everyone to use the same endpoint, got it. What we observed, in this way, is that the visual data model in Designer is only available to the user who created the model, and other users will be able to see the data in Explorer, but not to edit directly the data model in Designer, thus limiting the collaboration. Are we doing anything wrong or this is expected? and any way around that?
Designer projects are user owned and are working drafts of what can be published to Stardog. For other users of an endpoint to see/use your Designer work, you must publish the data model and mappings.
Working from the same data model and mappings -- Projects can be shared between users by exporting and importing the .stardogdesigner file. Each user owns their version of a project in Designer. As an alternative to having access to a .stardogdesigner file, data models from an endpoint you have access to can also be imported into Designer.
Publishing edits to a shared endpoint - When publishing from Designer, you can choose to replace existing data models and mappings.
Hello Laura, I work with Antonio, who asked the initial question. I have a follow up question now.
Indeed in Designer I can import a named graph that was created by someone else and publish to the same endpoint.
I have two named graphs and in both I have classes that have the same name and close to the same set of instances. I want the StarDog platform to understand that it is the same class and merge those two named graphs, how can I do that?
Please let me know if there is a way to treat those two classes from a different models as one?
Thanks,
Katarzyna