Kick off meeting notes

[nealhaddaway]

Use cases:
USE CASE 1 - remove duplicates, identify unique info from each db
USE CASE 2 - input final set and find out where they were found - back to front, input original RIS files AND final one, retain title abstract info and find out where they were excluded, still need the deduplication. Could either base it on CiteSource labelled RIS files OR new files
USE CASE 3- compare multiple strings in the same (or different) databases. So refer to ‘sources’ because could be comparing strings

Inputs:

1. Multiple RIS files
2. Option outputs (RIS lists) from screening at multiple levels
3. Optional search history data as a JSON that can be embedded inside each RIS file (see https://estech.shinyapps.io/searchrecorder/)

Outputs:

1. Visualisations:

• Record level search summary table - different levels (search results, abstract screening, full text screening)
• Summary Search summary table
• Co-occurrence matrix
• Circle plot
• Screening level source contribution flow diagram
• Data for PRISMA flow diagram (or additional file)
• Textual summary report

2. RIS outputs

• Full RIS file compiling all individual files
• Unique records only plus deduplication/source data
• Option to remove some fields (e.g. abstracts)

House rules

1. Document cleanly with roxygenisable content
2. Issue-level new branch in GitHub - name branch according to the issue, merge back when the issue is completed
3. Say WHY the function is doing what it’s doing
4. Use tags ESMARConf2022
5. Add yourself as the issues you’re working on
6. If things are not clearly issues, then comment in Discussions first
7. License - GPLv3? Will poll everyone