Erato

[gfish]

I was inspired to add group/subgroup support to Erato, my image indexing system. (Originally described here.) Now I can create a group, create subgroups within the group, assign images to the subgroups, and have all of them exported to the external web space, with an index file and slideshow automatically generated for them -- all from the comfort and saftey of the internal network. The output is all flat, static files. No file-accessing scripts exposed to the big, dark, scary internet.

The DC trip images, for example, are now all here. The group is 'trips' and the subgroup is 'dc2003'. The interface isn't very pretty, but it is quite functional.

The only big thing I'd still like to do with Erato at this point is play more with the description metadata. I already have a view that lists sets of description words by their frequency. (Not surprisingly, the porn terms from xiadyn's collection largely dominate the list.) I'd really like to steal one of those concept mapping systems to generate a linked graph of related words based on how many images they are found together in. Not a particularly important addition, obviously. But it would be pretty cool.


b&w: 166
nude: 105
fetish: 59
goth: 48
woman: 45
fishy: 36
mir: 30
lunarbot: 30
comic: 28
robot: 25
camera: 23
house: 23
animated: 22
mars: 20
bondage: 20
xia: 20
logo: 20
ball: 20
magic: 20
eight: 20
trip: 20
vix: 19
mom: 18
steve: 18
diet: 18
goedde: 18
space: 18
view: 17
pathfinder: 17
neuro: 17
tattoo: 16
tony: 16
ryan: 16
fishdom: 15
picture: 15
gyre: 15
old: 14
red: 14
standoff: 14
no: 13
background: 13
dc: 13
number: 12
map: 12
bruce's: 12
room: 12
detail: 12
to: 12
shoot: 11
loom: 10
front: 10
sts: 10
onion: 10
cover: 10
image: 10
corvi: 10
clothing: 10
republic: 10
intersections: 10

Comments

[xiadyn.livejournal.com]

I think there's still a bunch of porn I never quite got around to indexing, too. Hmm.

[gfish.livejournal.com]

Yes, another 407 images in the first temp directory.