Share your thoughts on using notebooks with your Power BI semantic models (preview)

[emlisa]

Hit Reply to tell us what you think about using notebooks with Power BI semantic models so we can continue to improve the experience.

For example:

• Are there any specific notebook samples you'd like to see included in the product?
• Do you have suggestions for new capabilities?

Thanks,

-Power BI team

To read more about this capability, please check out our documentation.

[aprayer]

@emlisa, I recently started using semantic-link-labs and am very excited by all that it does. It makes it so much easier to work with Fabric APIs. The ability to easily clone models and reports alone will prove to be a game-changer for us when it comes to managing reports and models at scale.

That said, I work in an industry that deals with highly sensitive data and data compliance requirements, so security is a top-of-mind consideration around using semantic-link and semantic-link-labs with models that contain sensitive data. As such, I wanted to ask:

1. Are the semantic-link and semantic-link-labs libraries vetted and backed by Microsoft from a security standpoint?
2. In other words, can we as customers feel comfortable using them in their current state?
3. Or does the "labs" nature of semantic-link-labs mean we should avoid adoption for anything other than experimentation?

Appreciate your time!

[aprayer]

Any update here?

[emlisa]

Hello, my apologies for the delayed response. Thank you for reaching out and for your enthusiasm around semantic-link-labs, we’re excited to see how it’s helping streamline your work with Fabric APIs.

To clarify Microsoft’s support with these:

• semantic-link is the officially supported Microsoft library, designed for production use and enterprise-grade reliability.  If security, compliance, and supportability are top priorities then semantic-link is the recommended choice.
• semantic-link-labs, on the other hand, is an open-source, community-driven extension hosted on Microsoft’s GitHub. It offers early access to experimental features and advanced capabilities. While it’s a powerful tool for prototyping and internal testing, it is not officially supported by Microsoft. Its purpose is to serve as a “public preview” space where new ideas can be explored and validated before being promoted into the supported semantic-link package.

[mkulikowski]

Hello - I think this is a fantastic work on your end! Where can I check definitions of the columns?

Again - thanks a lot for this - really helpful!

[emlisa]

Thank you for your kind words! I'm glad you find the work helpful!

You can use the Best practice analyzer sample notebook to help esnure your semantic model, including column definitions, are following best practices. Additionally, you can use the Memory analyzer sample notebook to provide insights into memory and storage statistics for your semantic model, including its columns.

If you'd like to see the properties of your column you can open the semantic model in the web and see the column properties in the Properties pane. You also can use Fabric notebooks and semantic link to help understand your model definition better. Running functions like list_semantic_model_objects from Semantic Link Labs will list of all semantic model objects (including columns) that are used in the report and where the objects were used.

I hope this helps! Please let us know if you have any additional feedback or suggestions regarding this capability.

[Anonymous · Accepted solution]

Hi @emlisa,

As of now, there are no widely reported issues from the community regarding the use of notebooks with Power BI semantic models in preview. However, since this feature is still in its early stages, One user faced unexpected challenges and requested the general availability of this feature.

https://community.fabric.microsoft.com/t5/Service/Preview-Feature-editing-semantic-model-from-web-an... 

Regards,
Vinay Pabbu

[emlisa]

Thank you for linking the other thread. I will keep an eye on it to see if the user has any further questions or concerns.

Just to clarify, this feedback forum was specifically created for feedback related to using Fabric notebooks with Power BI semantic models. We are using a separate feedback forum to track feedback regarding preview for editing semantic models on the web.

[Anonymous]

Hi @emlisa,

Thank you for the clarification. I appreciate you pointing that out and will take note of it for future reference.

Regards,

Vinay Pabbu