How do you get the judge information for something you see on CL?

[mlissner]

When I look at this page:

https://www.courtlistener.com/opinion/27/uphoff-figueroa-v-alejandro/

It shows that the Judges were:

Judges: Lynch, Chief Judge, Boudin and Stahl, Circuit Judges

and that the Panel was:

Panel: Michael Boudin, Sandra Lea Lynch, Norman H. Stahl

How do I get this information via the API or bulk data?

[mlissner]

Our data model for judges is tricky because the way judges take cases is tricky.

First, many fields have both a text version and a normalized version. When this is the case, the field name for the text version usually has a suffix of _str and the normalized one does not. We take this approach because often we get the text of the judge's name, but can't easily look up that judge in our DB. When that happens, we'll have the text version but not the normalized one. For example, if a docket said that the judge was "Michael Jordan", we'd put that value in our _str field, and then we'd try to look it up in our judge DB. Since we don't have a judge with that name, we'd only populate the _str field.

Beyond that, there are a number of fields to consider. For federal cases in the RECAP collection, we have these four fields:

• Docket.assigned_to(_str): This is the judge the case was assigned to. Often though, Article III judges (those appointed by the president and confirmed by the Senate) will refer their work out to magistrate judges. When that happens, we have...

• Docket.referred_to(_str): These fields are where we store the referral information.

For items in our opinion DB, it's a bit trickier. Here are the fields to consider:

• `Docket.panel(_str): Not yet used, but is intended to hold the initials of the judges on the panel that heard this case. This field is similar to the 'judges' field on the cluster (see below), but contains initials instead of full judge names, and applies to the case on the whole instead of only to a specific decision.
• Cluster.panel: A m2m field listing numerous panel members that heard a case.
• Cluster.non_participating_judges: The judges that heard the case, but did not participate in the opinion. I don't think this field is used yet.
• Cluster.judges: A text field containing a list of judges involved in an opinion.
• Opinion.author(_str): The author of the opinion
• Opinion.per_curiam: A boolean indicating the case is per curiam.
• Opinion.joined_by(_str): Judges that joined the lead other of the case.

Altogether, this could, perhaps, be simpler, but follows our concept that the data goes at the highest possible level. Unfortunately, that splits judge information across opinions, clusters, and dockets, but as we complete these fields, we're hopeful they'll slowly come to paint a very detailed picture.