Log of the #fcrepo channel on chat.freenode.net

Using timezone: Eastern Standard Time
* acoburn joins08:15
* dhlamb joins08:23
* mhwood joins08:29
* acoburn leaves08:52
* acoburn joins
* dwilcox joins08:57
* jgpawletko joins09:17
* awoods_ leaves09:30
* awoods joins09:38
* mhwood leaves09:48
* dwilcox leaves09:53
* ksclarke joins09:58
* ajs6f joins10:45
* github-ff joins10:50
[fcrepo4] acoburn opened pull request #849: remove dependency on guava (master...fcrepo-1646) http://git.io/vYIXi
* github-ff leaves
<awoods>ajs6f: fyi, Hydra has the use case of allowing some users to GET metadata/description but not GET the content/binary... as an example of when it may be useful to have different security policies on the nonRdfSource and its description. https://wiki.duraspace.org/display/FF/University+of+York+Use+Case+-+Access+Control
<ajs6f>awoods: We already said we can't conflate the URIs, didn't we?10:51
<awoods>ajs6f: can you be more verbose?
<ajs6f>awoods: The URIs of the two resources, data and description.10:52
<awoods>ajs6f: are you saying the use case is doable or not doable?
<ajs6f>awoods: I'm saying we already did it.10:53
<f4jenkins>Yippee, build fixed!
Project fcrepo4-T2 build #304: FIXED in 5 min 44 sec: http://jenkins.fcrepo.org/job/fcrepo4-T2/304/
* awoods: Upgrade jetty-maven-plugin
* awoods: Reintroduce javadoc plugin into build process
* A. Soroka: Respond with 400 status on request with empty path segments
<awoods>ajs6f: I am pondering the mechanics of associating one ACL with a description (that actually gets stored on the binary) and another ACL with the binary (which also gets stored on the binary).10:55
<ajs6f>awoods: I thought you were trying to restrict access to URIs, not JCR crap.10:56
<awoods>ajs6f: indeed
<ajs6f>awoods: So what do JCR nodes have to do with it?
<awoods>ajs6f: JCR is where the implementation happens... but hopefully there is no issue.10:58
<ajs6f>awoods: If we are restricting access to URIs by using JCR mechanics, we are almost certainly wrong and there will eventually be trouble.
<awoods>ajs6f: Modeshape AuthZ relies on access to specific nodes.11:00
<ajs6f>awoods: We should not be using MODE to do it. I get that it's convenient and gets most of the job done quickly. it will eventually bite us, because it couples URIs and nodes too directly.
<awoods>ajs6f: we should probably not be using MODE to do transactions or versioning either.11:03
<ajs6f>awoods: It's debatable, certainly. I would feel much more comfortable if we were working some, I don't know, some kind of standard or specification.11:04
<awoods>ajs6f: the standards apply at the REST API level... below is implementation detail.11:05
<ajs6f>awoods: Right. Having standards available would prevent us from leaking impl into the HTTP, which I'm confident we are doing.
* github-ff joins11:16
[fcrepo4] acoburn opened pull request #850: remove unnecessary VisibleForTesting annotation (master...fcrepo-1647) http://git.io/vYINE
* github-ff leaves
* dhlamb leaves
<ajs6f>awoods/acoburn/escowles: http://aksw.org/Projects/RDFUnit.html11:18
<awoods>ajs6f: sparql-query... it could be used against an external triplestore after a series of repo actions.11:19
<ajs6f>awoods: Or, with some jiggering, with fcr:transform.11:20
<awoods>ajs6f: always
<ajs6f>awoods: This https://github.com/AKSW/RDFUnit/wiki/Patterns-Generators seems like the cool part. Ontology -> test case.11:24
* dhlamb joins11:43
* dhlamb leaves11:47
* dwilcox joins12:08
* dwilcox leaves12:20
* dhlamb joins12:22
* dwilcox joins12:23
* ajs6f leaves12:24
* chadmills leaves13:34
* github-ff joins13:38
[fcrepo4] acoburn closed pull request #850: remove unnecessary VisibleForTesting annotation (master...fcrepo-1647) http://git.io/vYINE
* github-ff leaves
* ajs6f joins13:59
<acoburn>ajs6f/awoods: just got fcrepo-kernel-modeshape bundle to deploy properly in karaf14:23
<ajs6f>acoburn++ # PERSISTENCE PAYS OFF.14:24
<awoods>acoburn: historic
<acoburn>… and it uses guava 18.0
<ajs6f>acoburn: So the Guava issue goes away?
<acoburn>ajs6f: yes
<ajs6f>acoburn: Rocktacular!14:25
* ajs6f leaves14:59
anybody else on the webacl call? still hold music?15:03
<ruebot>i'm ok with letter A15:21
* jgpawletko leaves15:27
* jgpawletko joins15:48
* dwilcox leaves16:05
* acoburn leaves16:51
* jgpawletko leaves16:57
* cmmills joins17:09
* github-ff joins17:21
[fcrepo4] awoods pushed 2 new commits to master: http://git.io/vYq1P
fcrepo4/master 0ff8913 Aaron Coburn: remove dependency on guava
fcrepo4/master 2386ffb Andrew Woods: Merge pull request #849 from acoburn/fcrepo-1646...
* github-ff leaves
* dhlamb leaves17:32
* travis-ci joins17:38
fcrepo4/fcrepo4#3871 (master - 2386ffb : Andrew Woods): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/compare/59af39c2ced7...2386ffb3eacf
Build details : https://travis-ci.org/fcrepo4/fcrepo4/builds/72196367
* travis-ci leaves
* the_mgt_ joins19:11
* the_mgt leaves19:14
* ksclarke leaves19:19
* dhlamb joins21:05
* jgpawletko joins21:28
* awoods leaves22:31
* ff_logger_ joins23:33
* ff_logger__ joins23:39
* dhlamb leaves23:42

Generated by Sualtam