Myths on Maps

I am including a short, descriptive name for each event as I come across it. This helps me keep my place in the text, and figure out when the same event is being mentioned in different places.

For now, I have also stopped recording births and marriages when the location isn't specified, unless they are otherwise interesting. This information is included in the genealogy, but if anyone thinks it would be good to have it in the events table, too, I can do that.

I'm recording as "nowhere" both events that genuinely don't have a place and events for which the place is not given.

I met with LB and SA to discuss how the association of people with places should be organized. This is how I see my tasks now:

For now, I will be leaving the genealogy spreadsheet as it is, and use the list of people in the first column of that spreadsheet as the list of people to whom I will refer when I make the other spreadsheets. In this list, people with the same name are distinguished by a number which follows their name. If I need to add to this list, I will add to the bottom of the list. If I need to take people off the list, I will leave a blank row, so as not to mess with their ID numbers.

I will take Carly's list of places, finish eliminating the duplicates (so that each place name will be unique), and create a list of places to which I will refer when creating the other spreadsheets.

Next, I will create an events spreadsheet. The first column will have an event ID number. The second column will contain the name of the place in which the event took place (the name will be the same as one from the places spreadsheet). The third column will contain the names of the people present at the event (from the list of people), separated by commas. The fourth column will have the citation for the presence of the people in the third column at the place in the second column. More columns of people and citations will be filled in as needed, in the manner of the first two. There will be one citation per citation column, but in the people column there will be as many people as were explicitly mentioned, separated by commas. I am not to infer the presence of people not specifically mentioned by name in the text cited, unless the evidence is overwhelmingly obvious (e.g. a mother's presence at the birth of children may be assumed, even if she is not mentioned by name).

Once the events spreadsheet is complete, the journey spreadsheet can be made. In the first column of that spreadsheet will be a journey ID number. In the second column will be an ordered list of the ID numbers of the events (from the events spreadsheet) which make up the journey. For a series of events to be considered a journey, there must be a description in the text (or a very strong implication) of the group (the members of which can change, but which must include at least one person) traveling from one place to the next.

Places and people will be entered with their names, rather than their ID numbers, for ease of proofreading and of making up the journey spreadsheet. Events and journeys will be entered using their ID numbers, because of the difficulty of coming up with a short but unique name for each event. Though it wasn't discussed in the meeting, I imagine that eventually we might want to be able to name the journeys, so that they can be searched for.

Met with LB and LM because the one, big spreadsheet containing all the information was getting unwieldy.

LM will create three spreadsheets:
1) characters {essentially the data in the genealogy spreadsheet}
2) places {list from Carly expanded as needed}
3) events [id#; 0 or 1 places; citation_1; 1 to N comma delimited character(s) mentioned by citation 1; citation_2; 1 to N comma delimited character(s) mentioned by citation_2; etc.] {derived by LM from current places spreadsheet}
4) journeys [id#, 2 or more comma delimited event_id#s] {derived by LM from current places spreadsheet]

These spreadsheets will have to be reworked into proper db structures later.

Rough first draft of database tables we'll eventually need:

characters table {identifies a person or god}:
- id [integer]
- name [string]
- group(s) [comma delimited string(s)] {to be replaced with list of pointers to groups table}
- generation [null or integer]
- generation_confidence [boolean] {boolean now, could eventually have more values}
- place_of_birth [string] {to be replaced with pointer to place list}
- place_of_birth_citation [string]
- place_of_death [string {to be replaced with pointer to place list}
- place_of_death_citation [string]
- notes [string]

places table {identifies a place}:
- id [integer]
- name [string]
- notes [string]

events table {relates a place and characters}:
- id [integer]
- place [null or pointer to place table]
- participants [comma-delimited list of character ids] {or derive from event_characters table}
- event_citation [string]
- notes [string]

character_character table {relates a character and a character}:
- id [integer]
- primary_character [pointer to characters table]
- related_character [pointer to characters table]
- relationship [mortal father, mortal mother, god father, god mother, other parent, married-to]
- relationship_citation [string]

groups table {identifies a group}:
- id [integer]
- group_name [string]
- notes [string]

group_character table {relates a character and a group - a character can belong to 0,1,>1 group}:
- id [integer]
- character [pointer to characters table]
- group [pointer to groups table]

journeys table {identifies a journey}:
- id [integer]
- name [string]
- notes [string]

journey_event table {relates an event and a journey - the order of event records for each journey matters}:
- id [integer]
- journey [pointer to journeys table]
- event [pointer to events table]

event_character table {relates a character and an event}:
- id [integer]
- event [pointer to events table]
- character [pointer to characters table]

I have more or less finished by genealogy table, though does need some fine-tuning and disambiguation.

I'm planing to start work on the places table, as described in the posts below. I will also be creating two supplementary tables for expeditions. Both tables will have an identical list of expedition names in the first column (e.g. "Voyage of the Argonauts"). The first table will record, in chronological order, the places visited by the expedition. The second table will record the people who went on the expedition, noting their first and last destination within the expedition, if noted in Apollodorus and if different from the expedition as a whole.

I'm still interested in somehow being able to record the people in the same city at the same time, but I'm unsure how best to do this.

As discussed at the meeting, I will make the changes to my people genealogy table as SA outlined below (the existing parent column needs to be separated into the new columns). I will then create a new places table with the list of names from the genealogy table, noting place of birth and place of death, if known, followed other places known to be visited by each person in chronological order. (Though not discussed, I assume that if the chronological order differs from the narrative order, I will attempt to preserve the chronological order.)

I'll email LB as questions or problems arise with interpreting Apollodorus, or how best to translate his narrative into the table.

In the future, co-ordinates will need to be added to the places I've added to Carly's place names table, and someone will need to make sure that the text of Apollodorus (the Epitome in particular) can be retrieved using the citations in the tables. At the moment, on Perseus, particular passages in the Epitome can't be retrieved with their citations, though the Library seems to be working fine.

Met with LB and LM to discuss how to structure data in excel for eventual use on interactive map
People table, parent field : to be separated into parent-male-mortal, parent-female-mortal, parent-male-god, parent-female-god, parent-male-step, parent-female-step, parent-other; each field can have zero or more values separated by commas;
People table, parent-reference field : to be separated as described above for parent field

Places table
name field is key, identical to the name field in the people table
place-of-birth and place-of-birth-reference fields : zero or one text value
place-of-death and place-of-death-reference fields : zero or one text value
place
and place-reference
fields : one text value each, the places appear in chronological order

Eventually we'll need to create a person-names table so that we can have one canonical spelling and n variant spellings of a person-name. That will be populated manually.

Eventually these will be processed into xml structures.

I'm still working on the genealogy, going through Apollodorus and, for each person, noting the number of their generation (I'm up to 19), their parents, their place of birth, if given, and also writing down the citation from Apollodorus. This has also involved trying to work out when different people have the same name, and when the same people have different names. In the former case, I've been assigning each different person a number based on the order they appear in the text (eg Eurydice (1), Eurydice (2), etc.), and in the latter case I've just been writing down both names in the same cell in my spreadsheet.

For the numbers of the generations, I've been assigning a child a number one greater than the number of their parent with the highest number (so a child with a parent from generation 5 and from generation 7 gets a generation number of 8). Where a generation number is uncertain, I've marked this with a question mark. For example, if the generation number can't be determined through parentage, I've been giving a person the same generation number as their spouse, with a question mark. If that method isn't available, I've been assigning the same generation number as others who took part in the same adventure and seem to be contemporaries. Sometimes I haven't been able to determine a generation at all, but I will go back over it and give it a try.

I've also been recording the Greek version of the people's name, because I thought maybe at some point we might want it to be possible for people to search by different transliterations. I don't know how possible that will be, though.

After I've finished the genealogy (and it's taking longer than I thought, right now I'm in book 2) I think my next task is to go over Carly's spreadsheets and make sure that all the info Dr. Bowman wants is noted. I think it will be particularly important to add citations from the text.

It would be great to know if it sounds like I'm on track, or if there's something I should be doing differently.

I worked 20.5 hours on this last week, and hope to be able to devote more time to it this week.

Lauren has two main tasks this month:

a) go through Carly's Apollodorus and make sure that all of the information I have asked for is noted for later coding somehow. Make sure that when Carly made a list of heroes and places she did not simply make separate lists, but has attached the names and places to each other. A lot of the categories of information I've asked for in the last post is not stuff Carly will have been noting; please go back and put that stuff in.

b) Make a timeline of generations, based on Apollodorus, who thinks in generations. Feel free to use Carly's timeline, which was based on the wall chart, and the timeline in the book you've got that I've also got, because those people will have done a lot of work; but make sure that it agrees with Apollodorus, and fix it where it doesn't, because we have to start somewhere, and Apollodorus will have to do. Apollodorus works down from one family to the next, so the tricky bit will be matching generations across families, but do your best. Also, there's more than one marriage sometimes, and octogenarians marrying 10 year olds, and so on, but, as I say, do your best ...

c) If you get through all that and are looking for more to do, start on Pausanias and do the same thing.

I'm about to set up the money, I hope, so that you can submit timesheets to Stewart Arneil, who will do administration. I will talk to Stewart.

I'm writing to confirm what Greg and I discussed this morning.

First, again, don't worry about writing summaries of stuff; this site exists to allow people to visually see which myths are attached to which sites, according to which authors; it is for the user to sum up this information as they see fit, not for us to do for them.

My image of the site is one that permits filtered searches. So, someone could just look for "Herakles", and every site that Herakles is said to have visited will light up on the map. Click on one of those sites, and a list of all the authors that mention Herakles in connection with that site will show up. Click on one of the authors, and the passages in which Herakles is mentioned in connection with that site and that author show up. Click on one of the passages and the paragraph shows up, but one can scroll back and forth. Passages will be cited in traditional format (that is, work, book, line number for poetry; work, book, chapter, section number for prose; keyed to the original, NOT to the translation). Texts will show up in English, not in the originals; people who want to can use the line numbers to find the Greek or Latin somewhere else.

But the user can filter the search, and look for "Herakles in Apollodorus", and only those sites that Apollodorus mentions Herakles as having visited will light up. Click on a site after doing that search, and only the passages of Apollodorus in which Herakles is mentioned would show up.

Other searches I would like people to be able to do - and this matters, because Lauren will have to go through and make sure that all of this information is encoded somewhere:

• a) by character (hero)
• b) by site (geography)
• c) by generation of character (by the timeline Lauren is establishing now)
• c1) by the order in which the character visited the sites (did he visit Thebes first in Apollodorus but last in Pausanias and sometime in the middle of his career in Vergil? we want to see the route of his travels)
• d) by family of character (blood and marriage - perhaps separate these two)
• e) by general area hero is associated with (Argolid, Peloponnese, Crete, Asia Minor, Attica, Northwest Greek, Aetolia, etc - you can rely on Big Powell to tell you what the main areas are, Lauren; he's got the myths divided up into the general geographic areas I want)
• f) author
• g) era of author
• h) geographic area of author, where known (good luck with Homer, there. Also with Apollodorus)
• i) ethnicity of character (especially if a character is identified as non-Greek or non-Roman, as say Pelops is, or all those Latin characters in Virgil) j) metamorphoses (does the character turn into an animal or a star?)

If you can think of any more categories we need to know about, Lauren, please add them in and let me know now - better we start getting all the information now, than have to go back and revisit all the texts later and redo it!

By the end of the summer, I would like to see a working map with the myths from Apollodorus on it, and filtered searches available for it. Even if this is all we ever get, because I run out of money and can't get any more, a searchable map of Apollodorus would still be a useful thing.

So: does this sound doable?