Jack Reed

Instead of pending, should we `skip` tests we know aren't actively going to change?

There are also some parallelization approaches outlined here: https://docs.travis-ci.com/user/speeding-up-the-build/

I'm not sure if this is related, but we dug into this a bit in ArcLight and found out some oddities w/ Blacklight, rails-ujs, and jquery. Depending on the location...

Based off of the IIIF spec: > A manifest must have exactly one id, and it must be the http(s) URI at which it is published. https://iiif.io/api/presentation/2.1/#id So we should...

cc @anarchivist cc @dbrunton

Some more info about this: http://www.loc.gov/aba/pcc/bibframe/TaskGroups/URI%20FAQs.pdf and https://www.w3.org/blog/2016/05/https-and-the-semantic-weblinked-data/ Changing the `@id`'s does have implications.

Some comments about IE compat: http://iiif.io/api/auth/0.9/implementation/#cors-and-xdomainrequest-in-ie-9-and-below

I've also noticed that some tile sources work as expected (no seam) while others present the seam:  ### No seam https://dzkimgs.l.u-tokyo.ac.jp/iiifimgs/zuzoubu/12b02/0001-0025.tif/info.json ### Seam http://codh.rois.ac.jp/pmjt/iiif/200014778/200014778_00023.tif/info.json

Thanks for the comment. I see the seam in your video, but still cannot find it locally. However, on `https://dzkimgs.l.u-tokyo.ac.jp/iiifimgs/zuzoubu/02/0002.tif/info.json` the seam is present for me.

I've created an example we can use to test: https://bl.ocks.org/mejackreed/bec44e64a18dfcb5bd89d691a2f86a55