Top Five Things We’ve Learned About SQL LRS

SQL LRS

The Apache 2.0 learning record store. Now available on Iron Bank.

Top 5

We asked, and this is what you had to say. So without further ado, here are the top five things we learned from all of you about SQL LRS.

  1. You like the flexibility

    People like the flexibility that SQL provides in quickly connecting to and using whatever BI tool or dashboarding platform you like. We are the first to admit that we’ll never create BI and visualization tools as awesome as what is already available from Superset, Tableau, Miscrosoft Power BI, or Google Data Studio. So rather than force you into dashboards that show you “HOW MANY VERBS” or that provide myriad ways to create false correlations, we’d rather give you a foolproof and automated way to push your xAPI data directly into the standard enterprise-grade reporting platforms that organizations trust. And because it’s SQL, you can just use the SQL connectors that all of the big BI platforms already provide.

  2. You like the strict conformance

    Stakeholders responsible for data like how strict SQL LRS is in ensuring complete conformance with the xAPI specification and IEEE P9274.1.1. From what we’ve been told, there are apparently some commercial LRSs out there that allow users to take shortcuts in their data compliance — which sounds great in the moment, but which causes big headaches when LRS administrators change roles or when people in the BI department are trying to create reports for data that isn’t in the format that it is supposed to be in. We had one distraught LMS customer come to us because they couldn’t access the data in the LMS’s on-board LRS. When we finally managed to extract the data, we found that the LMS’s LRS was out-of-conformance and that there were errors throughout their data set. This non-conformance meant that none of the data worked with the reporting dashboards they’d spent the last year designing according to the xAPI specification.

  3. You like the deployment options

    People like the ability to deploy SQL LRS anywhere. So far, we know about AWS, Azure, and Google Cloud deployments. We heard that someone was looking to deploy on Oracle Cloud. And we’ve had numerous organizations reach out for help deploying on-premise both in traditional data centers and on mobile devices. We like to say that we built SQL LRS so that it could be run on a smart toaster. So, we’re really happy to see people leveraging the DockerHub install and finding all kinds of places to deploy.

  4. You want more flavors

    Folks want more SQL flavors. SQL LRS comes with Postgres for enterprise deployments as well as SQLite and H2. We’ve had asks to build out capability for MySQL as well as MSFT SQL Server. We’d like to provide these builds, but it will take time and resources to get it done. If your organization is interested in funding this as an open source project, get in touch and let’s discuss.

  5. You wanted a secure Apache 2.0 LRS — and that’s what you got

    SQL LRS was hardened to meet the most advanced DevSecOps requirements. The US Department of Defense just released SQL LRS in Iron Bank — Iron Bank is Platform One's authorized, hardened, and approved container repository that supports the end to end lifecycle needed for modern software development. You can check out the public repository available via Repo One. The significance of this is that there is now a free, Apache 2.0 open source, enterprise scale LRS available in Iron Bank for use across the DoD and by stakeholders across the entire US Federal Government. And the team here at Yet is happy to provide support, services, and training for these implementations… just ask.

What’s Next?

Companies and corporations are telling us that they appreciate the Apache 2.0 license — as earlier open source LRSs were distributed under copyleft licenses that hampered their ability to fully use them. And one of the things we’re expecting to see is that — in concert with the maturation of the data standards supporting the Total Learning Architecture — organizations are going to realize the power in creating a network of federated LRSs. And that kind of federated architecture depends on a flexible open source license both to keep costs down and to ensure a handle on technical IP concerns. We think SQL LRS is a good option for these federated enterprise deployments.

Previous
Previous

Open Source xAPI Capabilities for the Total Learning Architecture

Next
Next

xAPI Profile Validation: Open Source Resources from Yet