Myths on Maps

Category: "Activity log"

June 20, 2013

a puzzling thing about the map browser buttons

Posted by on 20 Jun 2013 in Activity log
In the "reader" screen, it makes sense to have the eyeball, info, details and little open book icons, because they each do something in that screen. In the "browse" screen, it makes sense to only have the "details" icon, because that's all you're using. But in the "map" screen all four little icons show up, but if you click on any of them except the eyeball, all that happens is that everything else (except the empty map) vanishes, leaving the user with no idea what to do or where to go.

It would make sense to remove all of the icons from the "map" screen except the little eyeball, since it's the only one that works.

"character" button doesn't work on map? no "place" is put on map

Posted by on 20 Jun 2013 in Activity log
Steps:

1. go to the map.

2. search by character.

3. choose "c".

4. Choose "Calesius".

5. Click on "Calesius".

6. Map will inform you that it has put on the map one place associated with Calesius.

7. However, no place has appeared on the map.

8. Info has opened up under "Calesius". One heading is "places". Click on "Places".

9. One place is associated with "Calesius": Arisbe.

10. Click on "Arisbe." One place will now show up on the map. But why didn't it show up before, when the map announced that it had added that place to the map?

Other observations:

1. In fact 2 places ought to be associated with Calesius, one of them being "Troy", where Diomedes killed him. Now I'm wondering if we didn't bother to associate any of the characters in the catalog of ships with Troy. This would be disturbing.

2. When you are on the map, searching under "character", and you click on a character, and info shows up, and now you click on 'citations', a list of citations shows up, but when you click on one of the list of citations - any one, it doesn't matter - the map says "(something or other) null". The texts do, then, eventually show up in the sidebar, interestingly. But the map has already told you (something or other) "null" in the green popup, so probably you didn't wait around for the texts because you figured they weren't going to show up.

breadcrumbs and a "go back" button

Posted by on 20 Jun 2013 in Activity log
Desperately needs a "go back to previous" button somehow. Also a "clear the map" button. I have needed both of these several times in the half hour I've been working so far today, for example; it's the most critical function that people will need that we don't have, that I can find so far.

a thing we need on the map: something should happen when you click on a spot

Posted by on 20 Jun 2013 in Activity log
When I go to the map, and choose "text", and ask it to (for example) map all the places mentioned in Book 3 of Homer's Iliad, it puts 11 places on the map. But when I click on one of the places, nothing happens; it doesn't tell me what the place is or anything else. It needs to tell me (at the least) what place it is.

differentiate worked-up texts from non-worked-up texts?

Posted by on 20 Jun 2013 in Activity log
It would be good if texts that we haven't marked up muchyet (like Pausanias and Homer) were indicated in the list of texts by being in a lighter colour, or being in italics, or in italics, or in some other way differentiated from the texts that we have actually worked on – so far really only Apollodorus, though we do have some done in Homer. (Although the Metamorphoses are listed, none of the places for people in the Metamorphoses are showing up on our map, even the ones that do appear elsewhere in our data. For Pausanias and Homer, places that are showing up elsewhere in our database are showing up on the map – I'm not sure what the difference is between how it's working, and how Pausanias and Homer are working.) Since the Metamorphoses aren't actually marked up at all, and places in the Metamorphoses aren't showing up on the map at all, should we remove that from the list to the side for the time being? Pausanias and Homer seem to be working in so far as there is any markup from Perseus for them.
June 7, 2013

Character disambiguation notes (for later reference)

Posted by on 07 Jun 2013 in Activity log
It is not clear to me why Actor (1), the son of Deion and Diomede, is also listed as the father of Eurytion. That may be a separate Actor. Someone may have had a reason for not separating these two. (There are already 4 Actors.)
June 6, 2013

What we're working on now

Posted by on 06 Jun 2013 in Activity log
After discussion, we have agreed that we should be able to publish Apollodorus this month. I will spend the intervening time doing enough proofreading and spot checks to find any systemic issues and alert either Lauren or Greg, who will fix/flag them.

Greg will spend the time he has available for the project working on the map; first correcting bugs, and second fiddling with what happens when one clicks on a location dot. A priority is being able to remove a dot from the map; another is being able to change colours; a third is what gets displayed when one clicks on a dot.

Greg will also look into finding a way to make a "Journey" appear to have been generated (i.e. something that looks like a journey), so that I can put it on the poster and on the video, and present that as a "future development" at the Bristol conference. "Display family on the map" would also be a great "future development". Or indeed a "present development".

I will write up the front matter (what the project is for, brief history of the project, acknowledgements, how to use the site, and my "nihil obstat, imprimatur"). I will also produce a screenshot-video showing how to use the site. I will also think about possibly applying for money from somewhere. I will, finally, make a poster.

Lauren will ...

Saving bandwidth: auto-titling with CSS and so forth

Posted by on 06 Jun 2013 in Activity log

The XHTML version of Apollodorus Library Book 1 includes a huge number of title attributes and class attributes. The titles are, not surprisingly, hover-induced info like 'Show character details'. Many of the classes aren't really used for CSS, but were jQuery selectors. The net result of all of this is that Apollodorus Library Book 1 weighs in at 221K. After a bit of research and fiddling I've come up with a way to reduce this significantly. By significant, I mean HUGE - the same page is now 144K.

I've removed all of the jQuery selectors classes and replaced them with html5 data attributes that I use anyway. So, instead of a selector like $(".classname") I now use $("[data-type]"), which removes 1649 instances of 'class="retrieve"' from that page.

I've also swapped title attributes out for an all-CSS approach, so an anchor like <a data-id="ouranus" data-type="character">Sky</a> can be given a pseudo-title with the following CSS:

a[data-type]:hover {text-decoration: underline;position:relative;}
a[data-type]:hover:after {
content: "Show " attr(data-type) " details for " attr(data-id);
padding: 4px 8px;
text-transform:capitalize;
color: rgb(255,255,255);
position: absolute;
left: 50%;
top: 100%;
white-space: nowrap;
z-index: 2;
border-radius: 5px ;
background-color: rgba(0,0,0,0.8);
}

NOTE: the initial position:relative seems to be crucial here. I haven't investigated why.

An example page can be viewed here.

May 23, 2013

XSLT to enable faster searches

Posted by on 23 May 2013 in Activity log

Used XSLT to add an @n attribute to all paragraphs holding the @xml:id of the preceding or first-child milestone element, to enable faster searches on p tags instead of ranges between milestones, while still returning the correct target milestone.