Log of the #duraspace-ff channel on chat.freenode.net

Using timezone: Eastern Standard Time
* awead leaves03:10
* awead joins
* jgpawletko joins06:44
* ajs6f joins06:54
* dwilcox joins07:23
* github-ff joins07:40
[ontology] escowles created tech-metadata-cleanup (+1 new commit): http://git.io/vUbZI
ontology/tech-metadata-cleanup 8113976 Esmé Cowles: Removing digest and mimeType from repository ontology (in favor of premis:hasMessageDigest and ebucore:mimeType)
* github-ff leaves
* github-ff joins07:41
[ontology] escowles opened pull request #26: Removing digest and mimeType from repository ontology (master...tech-metadata-cleanup) http://git.io/vUbZg
* github-ff leaves
* whikloj leaves08:17
* whikloj joins08:18
* dhlamb joins08:19
* whikloj leaves08:22
* whikloj joins08:53
* github-ff joins09:00
[migration-utils] mikedurbin force-pushed local-fedora-server from a82d31f to b8a2d62: http://git.io/vUb5J
migration-utils/local-fedora-server f685e4c Mike Durbin: Added support for migrating R and X 'local.fedora.server' urls.
migration-utils/local-fedora-server b8a2d62 Mike Durbin: Added support for migrating R and E 'local.fedora.server' URLs.
* github-ff leaves
* travis-ci joins09:03
fcrepo4-labs/migration-utils#35 (local-fedora-server - b8a2d62 : Mike Durbin): The build passed.
Change view : https://github.com/fcrepo4-labs/migration-utils/compare/a82d31f9e964...b8a2d62c5df2
Build details : http://travis-ci.org/fcrepo4-labs/migration-utils/builds/62698186
* travis-ci leaves
* github-ff joins09:15
[ontology] awoods pushed 1 new commit to master: http://git.io/vUbxZ
ontology/master ec9e8a6 Andrew Woods: Merge pull request #26 from fcrepo4/tech-metadata-cleanup...
* github-ff leaves
<ajs6f>awoods: I don't think rebasing this Java8 thing is worth it. It's a month out of date. It would be faster to start over, and we would end up with better code.09:16
<awoods>ajs6f: ouch09:17
<ajs6f>awoods: This was not well planned.
<awoods>ajs6f: It was sounding like your time is limited. Is this something you have time for?
<ajs6f>awoods: No. I suggest we leave the branch as a source of examples, at least for a while. Then we shift the compiler version. Then we can start making incremental changes, instead of try to do too much.09:18
<ajs6f>I will send a PR to change the compiler version.09:19
* mikeAtUVa joins09:22
* github-ff joins09:25
[migration-utils] awoods pushed 1 new commit to master: http://git.io/vUNvr
migration-utils/master 3deb4d6 Mike Durbin: Add support for migrating R and X 'local.fedora.server' urls....
* github-ff leaves
* github-ff joins09:26
[migration-utils] awoods closed pull request #11: Added support for migrating R and X 'local.fedora.server' urls. (master...local-fedora-server) http://git.io/vUa0H
* github-ff leaves
* github-ff joins
[migration-utils] awoods deleted local-fedora-server at b8a2d62: http://git.io/vUNvx
* github-ff leaves
* travis-ci joins09:27
fcrepo4-labs/migration-utils#37 (master - 3deb4d6 : Mike Durbin): The build passed.
Change view : https://github.com/fcrepo4-labs/migration-utils/compare/b053cfeaec7f...3deb4d63fc66
Build details : http://travis-ci.org/fcrepo4-labs/migration-utils/builds/62701709
* travis-ci leaves
<f4jenkins>Project migration-utils build #61: FAILURE in 2 min 24 sec: http://jenkins.fcrepo.org/job/migration-utils/61/09:29
* acoburn joins09:31
awoods: ping
<acoburn>awoods: for the URLs pointing to the mode source, I assume that pointing to the 4.2.0-tagged tree would be preferable to using master09:32
* github-ff joins
[fcrepo4] ajs6f created BumpCompilerVersion (+1 new commit): http://git.io/vUNU8
fcrepo4/BumpCompilerVersion 9f3e18d ajs6f: Shift compilation to Java 8
* github-ff leaves
<awoods>acoburn: agreed... we want to point to a specific version, not a moving target.09:33
<acoburn>awoods: I just sent a commit with URLs pointing at master, but I think pointing to the 4.2 version makes more sense
awoods: I'll another commit to the PR
* github-ff joins
[fcrepo4] ajs6f opened pull request #794: Shift compilation to Java 8 (master...BumpCompilerVersion) http://git.io/vUNUM
* github-ff leaves
<awoods>acoburn: I'll wait
* travis-ci joins09:37
fcrepo4/fcrepo4#3649 (BumpCompilerVersion - 9f3e18d : ajs6f): The build has errored.
Change view : https://github.com/fcrepo4/fcrepo4/commit/9f3e18d355c3
Build details : http://travis-ci.org/fcrepo4/fcrepo4/builds/62702728
* travis-ci leaves
<ajs6f>awoods: [ERROR] Failed to execute goal org.apache.maven.plugins:maven-compiler-plugin:3.2:compile (default-compile) on project fcrepo-kernel: Fatal error compiling: invalid target release: 1.8 -> [Help 1]09:38
DOes Travis not have Java 8 installed?
<awoods>ajs6f: likely not... I will send a note to the U-Maryland folks.09:39
<acoburn>ajs6f: you need to update the .travis.yml file
<ajs6f>acoburn: Right, now I remeber you doing that months ago. I'll do that now.
<acoburn>ajs6f: change oraclejdk7 to oraclejdk8
* github-ff joins09:41
[fcrepo4] ajs6f pushed 1 new commit to BumpCompilerVersion: http://git.io/vUNIh
fcrepo4/BumpCompilerVersion 98510e4 ajs6f: Updating Travis CI configuration for Java 8
* github-ff leaves
<f4jenkins>Project fcrepo4 build #2866: FAILURE in 9 min 3 sec: http://jenkins.fcrepo.org/job/fcrepo4/2866/
<acoburn>awoods: ok, I pushed a commit with the updated (version-tagged) links to the MODE source09:45
<awoods>acoburn: thanks
* acoburn leaves10:08
* acoburn joins
<dhlamb>is this correct? http://www.jcp.org/jcr/nt/1.0hierarchyNode looks like we're missing the trailing / for mix and nt namespaces that get added to rdf:type by fedora10:16
also looks like mix:lastModified and mix:created are being added to rdf:type by default, which don't seem like types to me10:19
<acoburn>dhlamb: these are correct10:20
dhlamb: the mix and nt namespaces don't contain a trailing /, which is just bad namespace design, but we can't do anything about that
<dhlamb>acoburn: wow, ok10:21
acoburn: and lastModified and created belong as rdf:type's as well?
<ajs6f>dhlamb: That's not going to get fixed until the next version of the JCR spec, for which no one is holding their breath.
<dhlamb>acoburn: ok. i will stop questioning and carry on.10:22
acoburn: thx. helpful as always.
<acoburn>dhlamb: as for the mix:created and mix:lastModified types, I don't know the answer. They don't seem appropriate for rdf:type, but maybe ajs6f knows10:23
<ajs6f>acoburn : dhlamb: They are an artifact of JCR, and the fact that RDF type are persisted a JCR mixins. We should probably be hiding them.10:24
dhlamb: If you find them confusing, file a ticket to excise them from the response.10:25
<dhlamb>ajs6f: i'm no rdf master. that's for others to decide. i have to sync rdf:type between fedora resources and drupal nodes and just stumbled across that stuff. i'm ambivalent.10:27
<ajs6f>dhlamb: Those are not really RDF types. You shouldn't translate them into other systems.
<acoburn>dhlamb/ajs6f: well I find them confusing
<dhlamb>ajs6f: acoburn: i'll make a ticket. let's cut 'em then.10:28
<ajs6f>Kill 'em all, let G-d sort 'em out.
<dhlamb>k, done: https://jira.duraspace.org/browse/FCREPO-154110:34
and now i see a typo and don't have privileges to edit the ticket :(10:35
<ajs6f>ty pes -> types
awoods: I fixed that, but can you fix dhlamb's permissions?10:36
* scossu joins10:43
<awoods>ajs6f/dhlamb: one moment10:44
dhlamb: you should be able to edit the issue now.10:48
acoburn: do you have suggestions on the squashing approach of: https://github.com/fcrepo4/fcrepo4/pull/79210:49
<acoburn>awoods: is there any reason not to squash them into a single commit?10:50
<ruebot>awoods: Mark Leggott offered up the UPEI conference line for the Islandora Fedora4 IG call. So, we're good there now :-)
<awoods>acoburn: only if the file history comes into play for some reason.10:51
<dhlamb>awoods: thx10:52
<acoburn>awoods: but given the current PR, the history isn't going to track the change of ExternalJsonSidecarExtraPropertyStore
<awoods>acoburn: just checking
<acoburn>awoods: in order to track the history, I'd need to restart the PR, make the changes to that class in place (in the org.modeshape package) for one commit10:53
<f4jenkins>Project fcrepo4-T2 build #236: UNSTABLE in 5 min 27 sec: http://jenkins.fcrepo.org/job/fcrepo4-T2/236/
<acoburn>awoods: and then have another commit track the change in file location (and change of package name)
<awoods>acoburn: let's go with what you have already.10:54
* dwilcox leaves11:04
<ajs6f>acoburn: https://mail-archives.apache.org/mod_mbox/jena-dev/201505.mbox/%3CF828312D-463C-416F-8E8E-64F9E0CB8A90@email.virginia.edu%3E11:10
* pgwillia joins
<ajs6f>I would like to see Jena move that way.
<acoburn>The commons-rdf project looks really promising11:13
* dwilcox joins
<acoburn>I saw some list traffic about this a few months ago on the clerezza-dev list11:14
ajs6f: and a streaming api would fit nicely with some of the java8 idioms11:19
<ajs6f>acoburn: Yes, and even better if _we don't have to code it_.
<acoburn>ajs6f: isn't that always the goal ;-)11:20
<ajs6f>acoburn: My goal is to eliminate _all_ the code.
<acoburn>ajs6f: did you see Rick Hickey's talk about "transducers"? It's a clojure thing but it's very much along these lines11:21
ajs6f: https://www.youtube.com/watch?v=6mTbuzafcII11:22
* mikeAtUVa leaves
<ajs6f>acoburn: No, I hadn't. I'll listen to anything he has to say, and not just because of his magnificent Weird Al haircut.11:23
* github-ff joins11:27
[fcrepo4] awoods pushed 1 new commit to master: http://git.io/vUAf0
fcrepo4/master dff2d5f Aaron Coburn: Remove org.modeshape exports from fcrepo-connector-file bundle...
* github-ff leaves
* github-ff joins11:28
[fcrepo4] awoods closed pull request #792: remove org.modeshape exports from fcrepo-connector-file bundle (master...fcrepo-1534) http://git.io/vUi1D
* github-ff leaves
* osmandin joins11:31
* travis-ci joins11:40
fcrepo4/fcrepo4#3654 (master - dff2d5f : Aaron Coburn): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/compare/31e068b35fee...dff2d5f0be46
Build details : http://travis-ci.org/fcrepo4/fcrepo4/builds/62719703
* travis-ci leaves
* osmandin leaves12:24
* escowles joins12:46
* osmandin joins13:07
<awoods>I'm out for the day, team.13:13
* awoods leaves13:20
<ruebot>acoburn: in rdfs, is there a way to say the value of a predicate to should be a given type of thing. ex: Boolean or integer14:31
<acoburn>ruebot: are you thinking of rdfs:range?14:32
<ruebot>acoburn: no. that doesn't look like it.14:33
acoburn: maybe a better example, so for this one https://github.com/Islandora-Labs/islandora_ontology/blob/master/relsext.rdf#L84-L87
acoburn: it should be true or false. is that something that would be/should be stated in the ontology?14:34
acoburn: or, would that just be stated in an rdfs:comment?
* whikloj_ joins14:35
<acoburn>ruebot: it seems you want to use a typed literal14:36
<acoburn>ruebot: some examples are here: http://www.w3.org/TR/swbp-xsch-datatypes/#sec-rdf-dt
<ajs6f>ruebot: I really think acoburn is right. You do want rdfs:range.14:37
http://islandora.ca/ontology/relsext#generate_ocr rdfs:range xsd:boolean
* whikloj leaves
<ruebot>acoburn++ ajs6f++
<acoburn>ruebot: what ajs6f said ^^^14:38
<ruebot>Cool. So something like this then: https://github.com/fcrepo4/ontology/blob/master/ldp-fedora-mapping.rdf#L4214:39
<acoburn>ruebot: yes, where the resource value is the class of suitable objects for your property14:44
<ruebot>acoburn++ ajs6f++ #thanks again!
<ajs6f>ruebot: Enjoy the pain of RDF.
<ruebot>ajs6f: enjoying the learning experience now. i'll probably come loathe it one day.14:45
* acoburn leaves15:44
* osmandin leaves15:47
* dwilcox leaves16:15
* dhlamb leaves16:21
* ajs6f leaves16:44
* awead leaves16:47
* escowles leaves17:07
* whikloj leaves17:40
* pgwillia leaves18:39
* scossu leaves18:48
* ksclarke leaves19:26
* awead joins20:54
* ksclarke joins21:17
* dhlamb joins21:20
* awead leaves21:43
* scossu joins22:20
* scossu leaves22:26
* scossu joins22:31
* scossu leaves22:39
* awead joins23:17
* jgpawletko leaves23:19
* Guest95947 leaves23:51