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

Using timezone: Eastern Standard Time
* ksclarke leaves01:07
* github-ff joins04:16
[fcrepo4-swordserver] claussni pushed 2 new commits to master: http://git.io/vsbEX
fcrepo4-swordserver/master faf3448 Aaron Coburn: clean up pom to correspond to upstream changes
fcrepo4-swordserver/master 545af74 Ralf Claussnitzer: Merge pull request #10 from acoburn/fcrepo-1684...
* github-ff leaves
* travis-ci joins04:21
fcrepo4-labs/fcrepo4-swordserver#42 (master - 545af74 : Ralf Claussnitzer): The build passed.
Change view : https://github.com/fcrepo4-labs/fcrepo4-swordserver/compare/dc38ea4bb9d4...545af741ecaa
Build details : https://travis-ci.org/fcrepo4-labs/fcrepo4-swordserver/builds/77284319
* travis-ci leaves
* travis-ci joins07:57
fcrepo4/fcrepo4#3985 (inbound-to-us - dec0838 : Esmé Cowles): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/commit/dec0838f4a1f
Build details : https://travis-ci.org/fcrepo4/fcrepo4/builds/77225914
* travis-ci leaves
* acoburn joins08:08
* MohamedAR joins08:12
* dhlamb joins08:30
* esm_ joins08:33
* osmandin joins08:37
* awead leaves08:44
* awead joins08:47
* acoburn leaves08:53
* acoburn joins09:01
* ajs6f joins09:15
* acoburn1 joins09:19
* acoburn leaves09:22
* awead leaves09:39
* whikloj joins09:54
* peichman joins09:56
* awead joins09:59
* ksclarke joins10:23
<whikloj>acoburn1/awoods/peichman/MohamedAR: is it IRC or Hangout today? I'm thinking a Hangout might be in order.10:33
<awoods>whikloj++
<peichman>@whikloj++
<acoburn1>whikloj: yes, hangout. thanks
<f4jenkins>Yippee, build fixed!10:53
Project fcrepo4-T2 build #336: FIXED in 5 min 33 sec: http://jenkins.fcrepo.org/job/fcrepo4-T2/336/
<awoods>whikloj??11:00
<whikloj>awoods: on my way
* dwilcox joins11:06
* awead leaves12:01
* pgwillia joins12:27
* awead joins13:00
* dwilcox leaves13:09
* dwilcox joins13:32
* dwilcox leaves13:35
<awoods>that was a doozie13:46
<whikloj>awoods++
* awead leaves
* awead joins14:04
* awead leaves14:06
* awead joins14:48
* awead leaves
* awead joins14:58
* github-ff joins15:03
[fcrepo4] escowles pushed 1 new commit to inbound-to-us: http://git.io/vspB9
fcrepo4/inbound-to-us cf0014f Esmé Cowles: Using uriFor instead of more verbose idTranslator invocation
* github-ff leaves
* dwilcox joins
<awoods>whikloj/acoburn1/MohamedAR/peichman: dwilcox notes: https://wiki.duraspace.org/display/FF/2015-08-24+to+2015-09-04+WebAccessControl+Meetings15:15
<dwilcox>Please update/correct since I missed the last 45 minutes or so
<acoburn1>dwilcox++15:16
<whikloj>agreed dwilcox++15:17
<dwilcox>You guys have more endurance than I do - that was a long meeting! :)
<whikloj>I think I may have passed out.15:18
awoods/acoburn/MohamedAR/peichman: will both our ACL container and the child Authorizations have a rdf:type "authorization"?15:19
<peichman>@whikloj: Only the children, the container should have a different type15:20
<acoburn1>I agree with peichman
<awoods>me too, me too
<peichman>possibly a locally defined one, unless the WebAC spec evolves to have one
<whikloj>peichman: that's what I was feeling, but we didn't say it. So rdf:type "Acl"?
<peichman>@whikloj: yeah15:21
<acoburn1>this is where an extension ontology can be put to use
<peichman>@whikloj: notionally in the same extension namespace as the hasAcl property
<acoburn1>peichman: exactly
<peichman>which would be handy, since the range of hasAcl should be an Acl
<whikloj>that begs the question, are we using "ex:Authorization" as our rdf: type or "acl:Authorization"?15:22
<acoburn1>acl:Authorization
<peichman>@acoburn:++
<whikloj>sounds good
<acoburn1>awoods: we need a github repo for this extension ontology (along with all of the ext modules that need to be pulled out of fcrepo4/ontology)15:23
awoods: should all of these ext-ontologies live in a single github repo or in separate ones?15:24
<whikloj>acoburn1: they should be tied to the extension they apply to, no?
I have a quick meeting, brb.15:25
<peichman>I would lean toward a separate repo, unless we were certain that an extension vocabulary wasn't going to outlive or outgrow a single extension module15:26
<acoburn1>awoods: the ontology should define how the code should be structured at a very high level, but it can (and some might say *should*) live separately
whikloj: ^^^15:27
<awoods>acoburn1: a new github repo (ontology-webac) in fcrepo4-labs (for now) makes sense.15:29
<acoburn1>awoods: ok, I'll add one15:30
<awoods>acoburn1: thinking longer term...
acoburn1: We will want ontologies and TCKs
<acoburn1>awoods: and for the others that are now in fcrepo4/ontology?
* jgpawletko leaves15:31
<ajs6f>The ontologies won't necessarily always be maintained in Github.
<awoods>acoburn1: Would it make sense to have the ontologies and TCKs together?
acoburn1: together, per extension.15:32
<acoburn1>awoods: that seems right
<awoods>acoburn1: how does that impact the name of the "ontology/TCK" github repo?15:33
<acoburn1>awoods: I'm terrible with names. Given that we don't have a TCK, we could call it ontology-webac for now and change it later15:36
* osmandin leaves
<peichman>newbie question: TCK?
<whikloj>^^ good question
<awoods>ajs6f: thoughts on the naming convention of such ontology/TCK github repos?15:37
<acoburn1>test compatability kit
<awoods>https://en.wikipedia.org/wiki/Technology_Compatibility_Kit
<ajs6f>Start with a name that describes the functionality, add -tck or -ontology as needed.
<peichman>ajs6f++
<ajs6f>We should consider deploying a tool to manage ontologies. Git is for source code.15:38
<whikloj>https://stanbol.apache.org/docs/trunk/components/ontologymanager/15:39
<ajs6f>whikloj: That is no longer under active development.
<whikloj>ajs6f: gah?!
<ajs6f>whikloj: Also, it's really meant more for managing the _use_ of set ontologies, not managing the evolution of ontologies, which is what we want to do,15:40
<whikloj>ajs6f: ahh
<awoods>ajs6f: tool suggestions?
<acoburn1>what we need is a durable repository that is able to track versions15:41
<awoods>punk
<whikloj>haha15:42
<ruebot>acoburn1++
<ajs6f>awoods: Something along the lines of the online version of Protege that Stanford runs.
<whikloj>http://protege.stanford.edu/15:43
<peichman>http://protegewiki.stanford.edu/wiki/WebProtegeAdminGuide
<awoods>ajs6f/whikloj: that is an editor. Does it also have version control?
<peichman>The desktop version does not, I don't know about the web version15:44
<whikloj>awoods: "Full change tracking and revision history"
<ajs6f>awoods: I don't know, and I don't think version control is actually very important for the ontologies. For source, yes, but not for the ontologies which change _much_ slower.
<whikloj>http://protegewiki.stanford.edu/wiki/WebProtege
<ajs6f>more slowly
whatever
<whikloj>....w...h...a...t...?15:45
<awoods>ajs6f: Tracking versions of the ontology seems critical.
<ajs6f>awoods: No. Tracking published versions of the ontologies is clritical
<awoods>ajs6f: Yes. agreed15:46
<whikloj>try it out, http://webprotege.stanford.edu/#List:coll=Home;
<awoods>GitHub is probably a reasonable place to start
<peichman>I think tracking the ontology versions under development is useful, even if they are not changing at the same rate as source code, because we get the same safety net for the ontology as for the code15:47
<awoods>peichman: That is my thinking as well.15:48
<peichman>GitHub + (something to manage published ontology versions)
* awead leaves
<ajs6f>Well, apprently Protege does that, so no problem. I have no particular brief for it— it's just a popular tool for the purpose, and Github is not a particularly useful tool for the purpose. None of the abilities that Github features (like kicking off CI builds) are useful with ontologies, whereas the specialized editors that products like protege offer _are_ useful.15:49
<whikloj>ajs6f: http://www.semantic-web-journal.net/content/webprot%C3%A9g%C3%A9-distributed-ontology-editor-and-knowledge-acquisition-tool-web
* dwilcox leaves
<awoods>I would say Git is for managing versions of text, not just source code.
* dwilcox joins15:50
<ajs6f>awoods: Sure. I'm saying we aren't just managing text. We are managing particular kinds of text with particular kinds of semnatics. Github offers services over text that help managing source code. Protege offers services over text that help managing ontology.15:51
Etc.
* github-ff joins
[fcrepo4] awoods pushed 1 new commit to master: http://git.io/vspD3
fcrepo4/master 99397f6 Aaron Coburn: Create parent POM module...
* github-ff leaves
* github-ff joins15:52
[fcrepo-module-auth-xacml] awoods pushed 2 new commits to master: http://git.io/vspD8
fcrepo-module-auth-xacml/master 40dfbf5 Aaron Coburn: update maven config based on upstream changes
fcrepo-module-auth-xacml/master 1ce59e9 Andrew Woods: Merge pull request #43 from acoburn/fcrepo-1684...
* github-ff leaves
* github-ff joins
[fcrepo-message-consumer] awoods pushed 2 new commits to master: http://git.io/vspD0
fcrepo-message-consumer/master 57fbe7e Aaron Coburn: refactor to use fcrepo-parent maven configuration
fcrepo-message-consumer/master cbacb08 Andrew Woods: Merge pull request #95 from acoburn/fcrepo-1684...
* github-ff leaves
<awoods>ajs6f: Are Protege "services" above and beyond what an IDE does for source code?15:53
<ajs6f>awoods: They are similar.
<awoods>ajs6f: no one is saying we should use Eclipse instead of Git?
<ajs6f>awoods: How on easrth are Git and Eclipse replacements for each other?15:54
<awoods>ajs6f: my point exactly
<ajs6f>awoods: Wrong again. We're comparing Github + IDE to WebProtege.
awoods: And Protege offers ontology-specific services that are _not_ available from any IDE (like refactoring in OWL).15:55
<awoods>ajs6f: Protege seems like an IDE for ontologies.
<acoburn1>awoods: the desktop tool is like an IDE, but I think ajs6f is talking about the Web version15:56
<ajs6f>awoods: Look, whatever. I have little faith at this point that we are ever going to get to the ontology-driven style of development that I hoped for a year ago. If we want to make this project about accreting great masses of source code, that's pretty much what we have the resources for.
* travis-ci joins
fcrepo4/fcrepo-module-auth-xacml#82 (master - 1ce59e9 : Andrew Woods): The build passed.
Change view : https://github.com/fcrepo4/fcrepo-module-auth-xacml/compare/49a33895f0b5...1ce59e9170c6
Build details : https://travis-ci.org/fcrepo4/fcrepo-module-auth-xacml/builds/77387304
* travis-ci leaves
<acoburn1>awoods: I've used the desktop version quite a bit (years ago) but I have no familiarity with the webprotege tool15:57
<awoods>acoburn1: nor I... nor anyone in this channel.15:58
* github-ff joins15:59
[fcrepo4] awoods closed pull request #874: Create parent POM module (master...fcrepo-1684) http://git.io/v3JnW
* github-ff leaves
* MohamedAR leaves16:00
* MohamedAR joins
<ajs6f>awoods: I've used it many times.16:02
aoods: It works very well.16:03
* jgpawletko joins16:04
* jgpawletko leaves
<awoods>awoods: does it allow for a team to collaborate on ontology editing?
ajs6f: does it allow for a team to collaborate on ontology editing?
<ajs6f>awoods Sure, why not?16:05
<awoods>acoburn1/whikloj/MohamedAR: In order to get the interfaces right, we probably need to sort out the class interactions. no?16:06
<peichman>@awoods: I think so
<acoburn1>awoods: yes16:07
<ajs6f>awoods: Google is your friend. http://protegewiki.stanford.edu/wiki/WebProtegeUsersGuide#Sharing_Project_with_Collaborators
<awoods>ajs6f: nice
<ajs6f>awoods: Eh. It's something.
<whikloj>awoods: yes, isn't that what we are doing in the Google doc?
* travis-ci joins16:10
fcrepo4/fcrepo4#3989 (master - 99397f6 : Aaron Coburn): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/compare/59defcbfffa0...99397f6e282a
Build details : https://travis-ci.org/fcrepo4/fcrepo4/builds/77387255
* travis-ci leaves
<acoburn1>whikloj/awoods: I can add a section to the google doc16:11
<whikloj>acoburn1: awoods did already
* dhlamb leaves
<acoburn1>whikloj: I mean really fleshing that out
<whikloj>acoburn1++16:12
<awoods>acoburn++
acoburn1++
<MohamedAR>acobourn1++
<awoods>acoburn2++
<MohamedAR>acoburn++
* acoburn joins16:15
<awoods>acoburn: I am reluctant to push any more to fcrepo4 before the following ticket gets reviewed/merged: https://jira.duraspace.org/browse/FCREPO-1640
acoburn: It has already be rebased a few times.16:16
<acoburn>awoods: I can review that now
* acoburn1 leaves
* awead joins16:26
* awead leaves16:31
* awead joins16:34
* dwilcox leaves16:36
* ajs6f leaves16:38
<acoburn>awoods: to confirm, the relevant PR is this: https://github.com/fcrepo4/fcrepo4/pull/864 correct?16:39
<awoods>acoburn: yes
acoburn: looks like it already needs rebasing.16:42
acoburn: shall I do that?
<acoburn>awoods: that would be great
<awoods>acoburn: rebased... building locally for sanity before pushing.16:43
<acoburn>awoods: thanks!16:44
* awead leaves16:51
* github-ff joins16:53
[fcrepo4] awoods force-pushed FCREPO-1640 from a289952 to 3a9d5ed: http://git.io/vOI6a
fcrepo4/FCREPO-1640 e5be921 ajs6f: Fix for FCREPO-1640
fcrepo4/FCREPO-1640 da7763e ajs6f: Correcting problematic hasVersions triple
fcrepo4/FCREPO-1640 2ce21bf ajs6f: Rebase-age
* github-ff leaves
<awoods>acoburn: rebased
<acoburn>awoods: I'm on it
<awoods>MohamedAR: Have you seen the comments on: https://github.com/fcrepo4-labs/fcrepo-module-auth-webac/pull/7/files#r3803146617:00
<MohamedAR>awoods: yes, just waiting for the google docs to be more solid before I start the refactoring.17:01
<awoods>MohamedAR: removing the ACCESS_TO_CLASS_ACL_LOCATION is probably safe.17:07
<MohamedAR>awoods: rename it to AuthorizationHandler as well?17:08
<awoods>MohamedAR: go for it17:09
<peichman>@awoods: we are going with a singe AuthorizationHandler instead of the AccessToClass and AccessTo handlers, correct?17:10
<awoods>peichman: that seems to make sense, given the ACL with child Authorization resources approach.17:11
* travis-ci joins
fcrepo4/fcrepo4#3990 (FCREPO-1640 - 3a9d5ed : Andrew Woods): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/compare/a28995205d4b...3a9d5ed3a7c9
Build details : https://travis-ci.org/fcrepo4/fcrepo4/builds/77397352
* travis-ci leaves
<whikloj>do we need a WebACAcl class to collect all the WebACAuthorizations now?17:12
* peichman leaves
<whikloj>^^ or interface to begin with
<awoods>whikloj: What would a WebACAcl class do that a Set<WebACAuthorization> would not?17:13
<whikloj>awoods: I guess nothing, this is handled in the AuthorizationHandler instead. Nevermind.17:14
<acoburn>whikloj: and the Set interface provides a lot of nice features for using filters and predicates
awoods: that PR looks good to me17:19
<awoods>acoburn: basically it just removes the confusing UI section you noted and punted out to three new versioning tickets.17:20
<acoburn>awoods: how would you like me to squash those commits?
awoods: re the versioning tickets, that makes a lot of sense. I'd like not to hold this up any longer17:21
<awoods>acoburn: maybe end up with two commits as follows:
UI:17:23
3a9d5ed3a7c9ace0bdd9f3a51f84029428f914c2
2101f30d83c3fd1a06817ee5c5b0d3e27e63f2fc
Main:
2ce21bf5262646ac50ccdb85dbf2d580d47b4a89
da7763e8985ef27a86e38270d7f97f7a8a5925d2
e5be921e05bef8ae58e9bfc1313b9e391d16f4cb
<acoburn>awoods: yep, that's what makes sense to me, too
* github-ff joins17:27
[fcrepo4] acoburn pushed 3 new commits to master: http://git.io/vshnG
fcrepo4/master 3b20c9e ajs6f: Fix for FCREPO-1640...
fcrepo4/master ac5a256 Andrew Woods: Update HTML UI logic to support single-subject Binary resources...
fcrepo4/master 3bee67c Aaron Coburn: Merge branch 'FCREPO-1640'...
* github-ff leaves
* github-ff joins17:28
[fcrepo4] acoburn closed pull request #864: Fcrepo 1640 (master...FCREPO-1640) http://git.io/vOIiG
* github-ff leaves
* esm_ leaves17:32
* terrellt leaves17:36
* terrellt joins17:39
* acoburn leaves17:40
* github-ff joins17:41
[fcrepo4] acoburn deleted FCREPO-1640 at 3a9d5ed: http://git.io/vsh4k
* github-ff leaves
* github-ff joins17:42
[fcrepo4] acoburn deleted FCREPo-1640 at 8da86e1: http://git.io/vsh4K
* github-ff leaves
* whikloj leaves17:45
* travis-ci joins17:46
fcrepo4/fcrepo4#3992 (master - 3bee67c : Aaron Coburn): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/compare/99397f6e282a...3bee67c3aed9
Build details : https://travis-ci.org/fcrepo4/fcrepo4/builds/77403058
* travis-ci leaves
* MohamedAR leaves17:53
* github-ff joins17:57
[fcrepo4] awoods pushed 1 new commit to master: http://git.io/vshzM
fcrepo4/master 731e92c Esmé Cowles: Filter references that aren't inbound to the node in question...
* github-ff leaves
* github-ff joins
[fcrepo4] awoods closed pull request #891: Filtering references that aren't inbound to the node in question (master...inbound-to-us) http://git.io/vsdTp
* github-ff leaves
* github-ff joins18:00
[fcrepo4] acoburn deleted inbound-to-us at cf0014f: http://git.io/vsh2n
* github-ff leaves
* travis-ci joins18:15
fcrepo4/fcrepo4#3993 (master - 731e92c : Esmé Cowles): The build passed.18:16
Change view : https://github.com/fcrepo4/fcrepo4/compare/3bee67c3aed9...731e92ca5ad3
Build details : https://travis-ci.org/fcrepo4/fcrepo4/builds/77407661
* travis-ci leaves
* ksclarke leaves18:23
* pgwillia leaves18:38
* the_mgt_ joins19:23
* the_mgt leaves19:27
* awead joins20:09
* awead leaves20:15
* ksclarke joins20:19
* ksclarke leaves20:25
* dhlamb joins20:35
* ksclarke joins20:42
* ksclarke leaves21:40
* ksclarke joins22:01
* dhlamb leaves23:05
* awoods leaves23:58
* awoods joins
* ksclarke leaves00:47
* f4jenkins leaves01:03
* f4jenkins joins