Manual Publishing

During intake, if data wasn't suitable for automation, DataSF staff will instruct the publisher on how to load data manually to the portal.

Notice to publish

DataSF staff will make sure the publisher has sufficient permissions to create new datasets on the open data portal and will create an account for them if they don't.

We then send a templated email that's a notice to proceed with manual publishing instructions. By this point they've already received an email acknowledging someone would follow up with next steps and thanking them for their submission.

The template looks something like this:

  1. I've created an account for [xxxxxxxxxxxxxxxxxx], you should get a separate email and complete the user account registration for SF Open Data

  2. Go ahead and upload your datasets per those instructions, but do not make the dataset public

  3. Finally, on completion, respond to this email with the url of the private dataset (don't worry, we can access the private datasets without you doing anything else)

If you get stuck at any point, please let me know and I can help you.

Variations on the notice

We may vary the follow up in a handful of ways:

  1. If we already have data in hand through the intake form (public data only), we will preload the data, share with the publisher and ask them to review

  2. If a user has an account already or has published before, we instruct them to use their existing account, but follow the same process

  3. If we want to load data manually as a precursor to automation, we'll set those expectations in the follow up

Next steps

After we hear back from the publisher, we'll enter into Review and Approval.