# Additional fields to capture from alto2txt for the first run at newspaper ingest pipeline

A ticked todo means we have agreed on a decision, [y] indicates we will add field. [n] means we skip for now

## Certain

There a few fields which we almost certainly want, which are to do with the provenance of the newspapers:

### Source note

<mods:relatedItem type="original">
    <mods:note type="source note">British Library Heritage Made Digital Newspapers</mods:note>
         <mods:location>
             <mods:physicalLocation authority="marcorg">Uk</mods:physicalLocation>
             <mods:physicalLocation>British Library</mods:physicalLocation>
       </mods:location>
 </mods:relatedItem>

Including this information in the metadata will help establish the provenance of newspaper in DB but also means that derived/sampled datasets also include sufficient metadata about provenance to remove copyright worries for external users of the data.

- [x] Capture decision [y] just the `<mods:note type="source note">British Library Heritage Made Digital Newspapers</mods:note>` text contents and see below for BLN data type.

(for later)

In BL Newspaper format the relevant provenance is probably:

<dc:Rights>Copyright &#x00A9; The British Library Board</dc:Rights>

## Maybe 🤷‍♂️

### Processing information

At the moment the following is captured:

From the METS

<mets:name>CCS docWorks/METAe Version 7.0-1</mets:name>

Some additional processing info is available in the ALTO:

<processingSoftware>
    <softwareCreator>Nuance Communications, Inc.</softwareCreator>
    <softwareName>OmniPage</softwareName>
    <softwareVersion>20</softwareVersion>
</processingSoftware>

- [x] Capture decision [n]

### Language

<mods:languageTerm type="code" authority="rfc3066">en</mods:languageTerm>

- [x] Capture decision [y]

### Page Number
Page number of the article in the physical newsaper. To extract via alto file path.

For the BNA from FMP this probably means adding the `fileloc2` variable (i.e. ALTO filename) to the alto2txt output:


Note that this requires changes to each of the `extract_text_[...].xslt` files (for different mets versions).

See: https://github.com/alan-turing-institute/Living-with-Machines/issues/2423

## Do you want to get into this stuff in this doc?
https://github.com/alan-turing-institute/Living-with-Machines/issues/2336#issuecomment-740566981

### Source collection, where we can be explicit, probably passed as parameter

e.g. JISC, FMP, BNA, LwM Sponsored ([needs name](https://github.com/alan-turing-institute/Living-with-Machines/issues/2416))

- [] Capture decision [y/n]