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

Using timezone: Eastern Standard Time
* mohamedar joins06:35
* github-ff joins07:02
[fcrepo-module-auth-webac] mohideen opened pull request #34: Delegate Update and Unit Test (delegate_impl...delegate_impl) http://git.io/vZmvB
* github-ff leaves
* github-ff joins07:24
[fcrepo-module-auth-webac] acoburn closed pull request #34: Delegate Update and Unit Test (delegate_impl...delegate_impl) http://git.io/vZmvB
* github-ff leaves
* github-ff joins
[fcrepo-module-auth-webac] acoburn deleted delegate_impl at ad7ca8d: http://git.io/vZmI0
* github-ff leaves
* dwilcox joins07:33
* dwilcox leaves07:53
* dhlamb joins08:00
* mohamedar leaves08:14
* ksclarke leaves08:29
* awead joins08:34
* ksclarke joins09:02
* ajs6f joins09:10
* esm_ joins09:12
* acoburn joins09:18
* ajs6f leaves09:24
* ajs6f joins09:30
awoods: We host the ontologies on some Duraspace server, right? Wouldn't it make more sense for them to dereference to the live repo wherein predicates from them were discovered?09:31
* jgpawletko leaves09:34
* ajs6f leaves
* ajs6f joins
* jgpawletko joins
<awoods>ajs6f: are you suggesting serving the ontologies from GitHub?09:50
<ajs6f>awoods: No, from the repos.
<awoods>ajs6f: what "repos"?09:51
<ajs6f>awoods; THE FEDORA REPOs.
<awoods>ajs6f: can you give me an example of such a FEDORA REPO?
<ajs6f>awoods: Do you think that no one is going to run a Fedora repo, anywhere in the world?09:52
<acoburn>ajs6f: but other projects may also want to reference those ontologies, no? I mean independent of a locally running fedora repo.
<ajs6f>acoburn: Yeah, the interop thing is more challenging. I'm worried about the fact that thousands of repos all over the Internet are going to depend on a single srver at Durapsace for their semantics.09:54
<awoods>ajs6f: is there another appropriate hosting site for the ontologies?09:55
<ajs6f>awoods: For the last time, I am suggesting that each repo should host them.09:57
<awoods>ajs6f: got it.09:58
<ajs6f>awoods: acoburn is right about the cross-reference thing. I'd have to think about that a lot more. But my main point is that for preservation purposes, we have introduced a SPOF at the Duraspace server.09:59
<awoods>ajs6f: I wonder if there is another appropriate hosting site for the ontologies?10:00
<ajs6f>awoods:acoburn: I actually don't much care about other projects referencing them. I would be astonished if some other project cared about Fedora's ontologies. I'd be more troubled about other on-site systems. (E.g. triplestores or indexes)
awoods: The whole point here is to avoid a single site.10:01
<acoburn> awoods/ajs6f: why not put them on S3?
<ajs6f>acoburn: That just shifts the SPOF from Duraspace to Amazon. We're not talking about technical machinations here. It's an organizational question.10:02
<awoods>acoburn: and S3 does not support 303 redirects
<acoburn>awoods: true
<ajs6f>awoods:acoburn: I'm thinking about this kind of thing:10:03
What happened with VOID.10:04
(Which I think we actually use, funny enough.)
<acoburn>ajs6f: what about using the local repo as a type of "cache", capable of returning the ontology, but the underlying source of the ontology is still where it currently resides
<ajs6f>acoburn: Yeah, that's what we want. Just… how to do that.
* dhlamb leaves10:09
* github-ff joins10:11
[fcrepo4] awoods pushed 2 new commits to master: http://git.io/vZmp1
fcrepo4/master f11a3de Aaron Coburn: made inaccessibleResource individual lowercase to align with ontology
fcrepo4/master af47018 Andrew Woods: Merge pull request #906 from acoburn/fcrepo-1722-followon...
* github-ff leaves
* github-ff joins10:12
[ontology] awoods pushed 2 new commits to master: http://git.io/vZmpy
ontology/master 2c0985d Aaron Coburn: added inaccessibleResource individual
ontology/master da509a3 Andrew Woods: Merge pull request #29 from acoburn/fcrepo-1722...
* github-ff leaves
<awoods>acoburn: Wasn't the "Applied Linked Data" working group largely formed as a result of needing to cache externally hosted ontologies?10:16
<acoburn>awoods: sort of — it was mostly caching data, not ontologies10:17
awoods: e.g. the author record for someone from LOC, as opposed to the ontology itself10:18
awoods: I realize that there is a fine line b/t "data" and "ontologies"
awoods: if there is a line at all
<awoods>acoburn: and the issue was performance? or what to do if LOC goes down?10:19
<acoburn>awoods: both
* github-ff joins
[fcrepo-module-auth-webac] awoods pushed 2 new commits to master: http://git.io/vZYew
fcrepo-module-auth-webac/master 50b80a3 Aaron Coburn: remove interface/impl for WebACAuthorization in favor of a POJO
fcrepo-module-auth-webac/master be01ce3 Andrew Woods: Merge pull request #33 from acoburn/fcrepo-1723...
* github-ff leaves
<acoburn>awoods: the issue was this: folks wanted to store something like: <> dc:creator <some-opaque-id>
awoods: but then retrieve the label for that opaque URI
awoods: in order to display it to the user in the web front end or in a solr search10:20
<awoods>acoburn: did the WG come up with an approach?
<acoburn>awoods: they were going down the route of "linked data fragments" when I got too busy and dropped off the group10:21
awoods: apparently, there is a specification (or recipe) for a type of caching server, called a "linked data fragments server"
awoods: terrelt was part of that group. he'd know more10:22
awoods: it's basically just a caching reverse proxy (think: squid or varnish)
awoods: but there are some special features for linked data (I think)
<ajs6f>acoburn: awoods: There are some considerably special features.10:23
acoburn:awoods: Limited kinds of subsetting or searhcing, designed to balance computation between client and server.
<awoods>acoburn: thanks. Following the evolving conversations of that WG may be informative for the issue ajs6f raises.10:24
* github-ff joins
[fcrepo-module-auth-webac] awoods deleted readme_travis_status at 77100af: http://git.io/vZYfg
* github-ff leaves
<acoburn>ajs6f: that would make sense. so some type of rdfs entailment?
* travis-ci joins
fcrepo4-labs/fcrepo-module-auth-webac#21 (master - be01ce3 : Andrew Woods): The build passed.
Change view : https://github.com/fcrepo4-labs/fcrepo-module-auth-webac/compare/f7a1120293d0...be01ce3d3abf
Build details : https://travis-ci.org/fcrepo4-labs/fcrepo-module-auth-webac/builds/79295068
* travis-ci leaves
<ajs6f>acoburn: No, simpler than that. More like just selecting triples from a larger set.
<acoburn>ajs6f: oh right. It's been a while since I looked at that
<f4jenkins>Project fcrepo4-release-tests build #350: UNSTABLE in 1 min 21 sec: http://jenkins.fcrepo.org/job/fcrepo4-release-tests/350/
<ajs6f>acoburn: I don't know if it's changed, but it seemed to me to do a good job explaining.
<acoburn>ajs6f: that would cut down a lot on network I/O, too10:27
<ajs6f>acoburn: That's the idea, yeah
acoburn: It's not obvious to me how it solves the namespace caching prob, but whatever. It's been almost an hour, so I've already lost interest in the whole question.10:28
* travis-ci joins10:29
fcrepo4-labs/fcrepo-module-auth-webac#22 (master - 1e44642 : Andrew Woods): The build passed.
Change view : https://github.com/fcrepo4-labs/fcrepo-module-auth-webac/compare/be01ce3d3abf...1e44642ba9e5
Build details : https://travis-ci.org/fcrepo4-labs/fcrepo-module-auth-webac/builds/79295936
* travis-ci leaves
* whikloj joins10:31
* ajs6f leaves10:34
* jmignault joins10:42
* ajs6f joins10:45
* github-ff joins10:46
[fcrepo-module-auth-webac] acoburn opened pull request #35: Clean up POM (master...fcrepo-1724) http://git.io/vZYqi
* github-ff leaves
* jmignault leaves10:47
* jmignault joins10:48
* jmignault leaves
* jmignault joins
<f4jenkins>Yippee, build fixed!10:52
Project fcrepo4-T2 build #351: FIXED in 4 min 20 sec: http://jenkins.fcrepo.org/job/fcrepo4-T2/351/
Aaron Coburn: made inaccessibleResource individual lowercase to align with ontology
* github-ff joins10:56
[fcrepo4] awoods pushed 2 new commits to master: http://git.io/vZY31
fcrepo4/master e0dde06 Yinlin Chen: Could use Metaspace to replace PermGen...
fcrepo4/master b061f33 Andrew Woods: Merge pull request #902 from yinlinchen/Metaspace...
* github-ff leaves
<f4jenkins>Project fcrepo4-client build #797: FAILURE in 34 sec: http://jenkins.fcrepo.org/job/fcrepo4-client/797/11:09
* f4jenkins leaves11:27
* ksclarke leaves11:29
* f4jenkins joins11:31
Yippee, build fixed!11:34
Project fcrepo4-client build #798: FIXED in 1 min 32 sec: http://jenkins.fcrepo.org/job/fcrepo4-client/798/
* github-ff joins11:37
[fcrepo-module-auth-webac] acoburn opened pull request #36: sonar-related cleanup (master...fcrepo-1726) http://git.io/vZYg1
* github-ff leaves
* whikloj leaves11:41
<ruebot>awoods: any luck on the cert issue with m2?11:46
<awoods>ruebot: I was just chatting with Tim Donohue about it...
<awoods>ruebot: I'm not sure there is anything to be done on the server-side. This appears to be a Godaddy issue: http://stackoverflow.com/questions/18746565/godaddy-ssl-cert-not-working-with-java11:47
* ksclarke joins
* ksclarke leaves
* ksclarke joins11:48
* ruebot snorts11:51
* ruebot sighes
<awoods>ruebot: what is the frequency of your builds? Have they been running regularly for months? years?11:52
ruebot: I am curious why this issue did not pop up earlier.11:53
<ruebot>awoods: very infrequent. drupal filter isn't touched that often.
<awoods>ruebot: like every two years?
* ksclarke leaves11:54
<awoods>ruebot: can you pass me a link to the fail islandora git repo?11:55
<ruebot>awoods: https://travis-ci.org/Islandora/islandora_drupal_filter/jobs/79058323#L120511:57
<ruebot>awoods: basically i was setting it up to use Java8, and added fcrepo 3.8.1 to the TravisCI config
awoods: and it looks like the exact error from the stackoverflow link
* dhlamb joins12:03
* ksclarke joins12:05
* github-ff joins
[fcrepo-module-auth-webac] acoburn created fix_readme_again from master (+0 new commits): http://git.io/vZYMJ
* github-ff leaves
* github-ff joins12:06
[fcrepo-module-auth-webac] acoburn pushed 1 new commit to fix_readme_again: http://git.io/vZYM3
fcrepo-module-auth-webac/fix_readme_again d8d921c Aaron Coburn: change travis-ci link to point to fcrepo4 rather than fcrepo4-labs
* github-ff leaves
* github-ff joins
[fcrepo-module-auth-webac] acoburn opened pull request #37: change travis-ci link to point to fcrepo4 rather than fcrepo4-labs (master...fix_readme_again) http://git.io/vZYMz
* github-ff leaves
<acoburn>awoods: would you like a jira ticket for this ^^^12:07
<awoods>acoburn: no thanks. Readme updates can fly below the radar.12:08
<acoburn>awoods: especially one as trivial as that
* github-ff joins12:09
[fcrepo-module-auth-webac] awoods pushed 1 new commit to master: http://git.io/vZYD4
fcrepo-module-auth-webac/master 5d97976 Andrew Woods: Merge pull request #37 from fcrepo4/fix_readme_again...
* github-ff leaves
<awoods>acoburn: Are your other two tickets ready ready for merge... or are they in flux?
<acoburn>awoods: they are now both ready for you
<awoods>acoburn: thanks12:11
<acoburn>awoods: and I think that does it for now (I've got to move on to other projects)
<awoods>acoburn: what is the approach for detailing the "<osgi.import.packages>" section of the pom.xml? Is it even needed now?12:12
acoburn: It sounded like those imports help in generating features.xml?
<acoburn>awoods: they do12:13
awoods: but it can stay as "*" for now
<awoods>acoburn: At what point will we want the details there?
<acoburn>awoods: when I get to this ticket: FCREPO-169712:14
awoods: I'll want to add a feature for -auth-webac, and when I do that I'll add the explicit dependency list
<awoods>acoburn: ok... although it may be healthy if someone else took fcrepo-169712:15
<acoburn>awoods: I completely agree12:16
* jmignault leaves
* github-ff joins12:24
[fcrepo-module-auth-webac] awoods pushed 1 new commit to master: http://git.io/vZY7z
fcrepo-module-auth-webac/master efd6ddd Aaron Coburn: Clean up POM...
* github-ff leaves
* github-ff joins
[fcrepo-module-auth-webac] awoods closed pull request #35: Clean up POM (master...fcrepo-1724) http://git.io/vZYqi
* github-ff leaves
* github-ff joins12:27
[fcrepo-module-auth-webac] awoods pushed 1 new commit to master: http://git.io/vZY5y
fcrepo-module-auth-webac/master 46f25b4 Aaron Coburn: Sonar-related cleanup...
* github-ff leaves
* github-ff joins12:28
[fcrepo-module-auth-webac] awoods closed pull request #36: sonar-related cleanup (master...fcrepo-1726) http://git.io/vZYg1
* github-ff leaves
* awead leaves12:34
* awead joins12:35
* escowles joins13:02
* piyapong joins13:57
* piyapong leaves
* awead leaves14:22
* ksclarke leaves14:28
* acoburn1 joins14:32
* acoburn leaves14:33
* acoburn joins14:34
* acoburn1 leaves14:37
* ksclarke joins14:52
* awead joins14:53
* acoburn leaves15:36
* dhlamb leaves15:40
* ajs6f leaves16:04
* ajs6f joins16:08
* ajs6f leaves16:39
* jgpawletko leaves17:30
* esm_ leaves17:45
* jgpawletko joins17:54
* jgpawletko leaves
* ksclarke leaves18:29
* ksclarke joins18:30
* dwilcox joins18:46
* ksclarke leaves18:51
* dwilcox leaves18:54
* the_mgt_ joins19:10
* the_mgt leaves19:13
* ksclarke joins20:16
* ksclarke leaves20:22
* ksclarke joins20:26
* terrellt leaves20:47
* terrellt joins20:55
* dhlamb joins21:28
* awead_ joins21:55
* awead leaves
* ksclarke leaves22:58
* github-ff joins23:03
[fcrepo4] awoods opened pull request #907: Add logic to handle patch requests that reference the root... (master...fcrepo-1725) http://git.io/vZsP5
* github-ff leaves
* github-ff joins
[fcrepo-module-auth-webac] awoods opened pull request #38: Add integration tests for: (master...fcrepo-1725) http://git.io/vZsPx
* github-ff leaves
* dhlamb leaves23:45
* awead leaves00:33
* awead joins00:42
* awead leaves01:34