pleiades matching/editing workflow (priority: hestia critical)

[atomrab]

It now seems to have been decided that we will follow a manual workflow for the correction of bad or missing Pleiades references. Eric's version of gapvis has a built-in error-reporting pop-up for places. Once an error is reported, an asterisk appears next to the place name, linking to a page where the Pleiades URI can be updated.

There are three cases in which this will be necessary. All can be seen at 1.148: http://gap2.alexandriaarchive.org/gapvis/index.html#book/1/read/148/614.

1. Samos is recognized as a place, but has been connected to a bad URI -- should be connected to http://pleiades.stoa.org/places/599926. I've tried to make this change using the regular system, but so far it hasn't been updated in the interface. In this chapter, I'm not even sure that Samos is being recognized as a place.

2. Mycale has not been recognized as a place, apparently, or at least it doesn't appear in the timeline or on the map, although it is in bold in the text and I think is one of the [Unknown reference]/0 mention sites at the bottom of our book summary. There is a proper Pleiades URI for this place: http://pleiades.stoa.org/places/599805. I tried to update this one, too, but so far that change is also not reflected in the interface.

3. This was a good chapter, because it made me notice another problem. There is a reference to Poseidon of Helicon in this chapter, in the context of the sanctuary of the Panionion. The Helicon in his title (which refers ultimately to Mount Helicon in Boeotia) has been associated with the site of Helike in Achaea (http://pleiades.stoa.org/places/570281). I am not sure that an epithet should receive an actual place reference on the map in any case (even if it's correct), since this isn't a reference to a place in the text in the same way that a mention of a site in the narrative is (though I would be happy to have a conversation about this -- it certainly could be interesting to see where gods from particular places show up). If everyone agrees with me, we need a way to strip the place reference from this token altogether.

So, to summarize, here are the three things I need to be sorted ASAP to be able to get the text in shape for my class, which begins in the second week of January: I need to be able to correct bad URIs easily (current workflow is fine, but it has to actually work, and changes need to cascade across the dataset); I need to be able to add URIs to places that are showing up as "zero references" (still don't know what this means -- id'd as place but no match found? not id'd as place?); and I need to be able to strip URIs from things that look like place names but aren't really (like Poseidon of Helicon away from Helicon).

[ekansa]

OK:

As discussed in our call, we should change the "Report an Issue" hover - over tool tip thing in GapVis to link to a page that will have a form that will be used to report issues of the type noted by Adam:

-- correct associations between a place in the text and a Pleiades identifier (lots of places with wrong identifiers)
-- strip Pleiades identifiers from words that aren't places
-- add Pleiades identifiers to words that are places

The Report an Issue hover over tool should point to the following URL:

http://gap2.alexandriaarchive.org/report/token-issues/ {token-id}

The token-id is available as the "data-token-id" attribute of the elements that encapsulate place-tokens (see example: http://gap2.alexandriaarchive.org/books/1/page/209.json)

[atomrab]

This workflow is fully functional -- I've been able to deal with a large number of problem places this way. I'm considering this issue closed, as well as #19.