Time
S
Nick
Message
07:05 jri joined #dataverse
07:58 axfelix joined #dataverse
08:07 axfelix joined #dataverse
08:10 axfelix joined #dataverse
10:22 axfelix joined #dataverse
13:17 djbrooke joined #dataverse
13:46 djbrooke_ joined #dataverse
14:03 pameyer joined #dataverse
14:04 djbrooke joined #dataverse
14:30 djbrooke joined #dataverse
14:56 djbrooke_ joined #dataverse
15:58 djbrooke joined #dataverse
16:15 g_ joined #dataverse
16:38 djbrooke joined #dataverse
16:41 djbrooke joined #dataverse
16:42 djbrooke_ joined #dataverse
16:43 djbrooke joined #dataverse
16:44 djbrooke_ joined #dataverse
16:45 djbrook__ joined #dataverse
16:46 djbroo___ joined #dataverse
16:49 djbrooke joined #dataverse
17:28 djbrooke joined #dataverse
17:56 djbrooke joined #dataverse
17:58 djbrooke_ joined #dataverse
17:59 djbrook__ joined #dataverse
18:04 donsizemore joined #dataverse
18:04 donsizemore
@pdurbin quick Q?
18:07 pdurbin
donsizemore: hi! I've been trying to take notes at https://github.com/IQSS/dataverse/issues/2937 so I don't lose any of the great tips you're sending me!
18:08 donsizemore
@pdurbin: i shouldn't say this (knock on wood) but digging into Shibboleth has been fun
18:08 djbrooke joined #dataverse
18:08 donsizemore
@pdurbin: your note about Harvard's IdP folks includes a footnote with an e-mail address
18:09 donsizemore
@pdurbin: does this mean that they won't release the R&S bundle to to the entire federation, but they might to an individual SP like dataverse-test.irss.unc.edu?
18:10 pdurbin
They have been willing to release the equivalent to the R&S bundle to a number of Dataverse servers at Harvard (test boxes, etc.).
18:10 pdurbin
donsizemore: how do you plan to handle the Emory IdP/
18:10 pdurbin
?
18:15 pdurbin
it's the same problem
18:15 pdurbin
a trust problem
18:16 donsizemore
@pdurbin oh, i wrote to the emory people separately
18:17 donsizemore
@pdurbin: the longer term would look at what we could do without the R&S bundle
18:18 pdurbin
donsizemore: so you're going to have an XML file with Emory and Harvard in it? And also a feed for the R&S institutions? Multiple `MetadataProvider` elements.
18:19 donsizemore
@pdurbin: i just asked emory if they'd release the R&S bundle -- they're discussing the best way to handle it. yowch, so asking any entity to release that bundle to our SP will require a local conf file?
18:20 pdurbin
donsizemore: I hear you on the R&S category. It's too restrictive to require it. But what should the Dataverse software do with IdPs (like Emory and UT Austin) that are part of InCommon but don't even provide an eppn?
18:21 donsizemore
@pdurbin: do they not release it because of phishing concerns?
18:22 pdurbin
I'm really not sure. It would make my life easier if *all* the InCommon IdPs were part of the R&S category.
18:23 donsizemore
@pdurbin: would enough user requests make that happen at the community level, or would there be some better way of going about it politically?
18:24 pdurbin
I am no expert on the politics of InCommon.
18:25 pdurbin
From my perspective I need at least an eppn from all the IdPs. Otherwise, how can I know the user is the same user logging in?
18:29 donsizemore
@pdurbin: agreed
18:32 pdurbin
I was hoping that all the 430 IdPs in InCommon would send the required attributes Dataverse needs. Last week we learned that some don't even send eppn: https://docs.google.com/spreadsheets/d/1fWHGamXetTQw3cpf6Tc4efmTvoC97lpEeTLvN7Mxz3g/edit?usp=sharing
18:33 pdurbin
It would be nice if we could somehow filter out the ones that don't even send eppn.
18:33 pdurbin
I have no idea how to identify them though.
18:33 sivoais joined #dataverse
18:33 pdurbin
donsizemore: but yes, the ones that don't even send eppn could be added on the side through a static XML file using the trick you mentioned.
18:34 pdurbin
After an agreement has been reached with the IdP owner to release eppn and the other required attributes.
18:35 bricas joined #dataverse
18:36 donsizemore
@pdurbin: i can look into identifying which group in the federation doesn't even send eppn
18:36 bricas
pdurbin: 'ello once again.
18:36 pdurbin
bricas: hey there
18:36 bricas
bricas: looks like we're going ahead with our dataverse install (finally!)
18:37 bricas
oops
18:37 pdurbin
NICE
18:37 bricas
pdurbin: --^ :)
18:37 bricas
pdurbin: going to wait until 4.4 comes out, though.
18:38 pdurbin
ok
18:38 bricas
pdurbin: has dataverse been tested against solr 5, at all?
18:40 pdurbin
I haven't tested it but the guy who made this pull request may have: https://github.com/IQSS/dataverse/pull/2985
18:43 bricas
oh nice. there's a solr 5 schema there. perfect.
18:43 djbrooke joined #dataverse
18:43 pdurbin
bricas: I feel bad that no one here has looked at that pull request. If you want to try it out that branch, please be my guest.
18:50 bricas
it would be lovely if i could just plunk down the 5.x schema and be on my way :)
18:51 pdurbin
We're still using Solr 4.6.0.
18:53 bricas
we dropped solr 4 (which we *were* using for some drupal sites) for 5 so we could get rid of tomcat
18:54 pdurbin
I hear Solr 5 is more like a service.
18:56 bricas
we had 4.x deployed from a war to tomcat, but 5.x is stand-alone. it's been far more reliable honestly.
18:56 bricas
well it's a fresh install so i'll just give it a try and see what breaks. :P
19:28 djbrooke_ joined #dataverse
19:46 djbrooke joined #dataverse
20:35 djbrooke_ joined #dataverse
21:13 djbrooke joined #dataverse
21:19 djbrooke joined #dataverse
21:20 djbrooke_ joined #dataverse
22:00 djbrooke joined #dataverse
