962125 - Supersearch shouldn't use pretty names for search fields

Status: RESOLVED WONTFIX

(Socorro :: Webapp, task, P3)

Description

[Benjamin Smedberg]

Supersearch currently has pretty English names for search fields. I think this is actually hurting people who might use the fields, and the UI should instead just expose the fieldnames exactly as they appear in the raw data. So "EMCheckCompatibility", "AsyncShutdownTimeout", and so on.

The search page should also add a link to https://developer.mozilla.org/en-US/docs/Crash_Data_Analysis which provides more info about the metadata.

Comment 1

[Will Kahn-Greene [:willkg] ET needinfo? me]

As a data ingestion pipeline, I think it's important to separate "the raw data coming in" from the "normalized, sanitized, cleaned up data" that comes out of the processor that is searchable by Supersearch. I'm moving towards a "Supersearch only searches processed crash data" model where the processed crash will always have pretty names.

Underlying this request is concern about confusion of documentation of fields and where the data comes from. I think that'll get solved as I build out schemas and improve that side of the pipeline using similar systems that we use for Telemetry now.

I'm going to mark this as WONTFIX.