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

Using timezone: Eastern Standard Time
* jonathangee joins07:54
* daniellamb joins08:13
* ajs6f joins09:16
* osmandin joins09:19
* ksclarke joins09:32
* awoods joins09:40
<pivotal-bot>Gregory Jansen added comment: "There is a ManifestMonitorIT that add a new bag to the running connector. It verifies receipt of a node a..." https://www.pivotaltracker.com/story/show/4738061909:52
Gregory Jansen added comment: "The discrepancy here could be that the bag must contain a manifest before it is added by the connector. A..." https://www.pivotaltracker.com/story/show/4738061909:55
Gregory Jansen started "Watcher for federated filesystem that triggers events" https://www.pivotaltracker.com/story/show/4738061909:56
Andrew Woods edited "Protect the /authorize endpoint in the OAuth way" https://www.pivotaltracker.com/story/show/5156988709:59
Andrew Woods edited "Three-quarter-assed story for policy-driven storage" https://www.pivotaltracker.com/story/show/4825347710:01
Andrew Woods edited "Three-quarter-assed story for policy-driven storage" https://www.pivotaltracker.com/story/show/48253477
Andrew Woods added "Allow users to authenticate via LDAP" https://www.pivotaltracker.com/story/show/5419667410:03
Andrew Woods added "Allow users to authenticate via Shibboleth " https://www.pivotaltracker.com/story/show/54196700
Andrew Woods added "Allow users to authenticate via CAS" https://www.pivotaltracker.com/story/show/54196724
Andrew Woods added "Allow resources to be protected at the jcr node level" https://www.pivotaltracker.com/story/show/5419673810:04
Andrew Woods added "Ensure nodes inherit access policy from ancestors in jcr tree" https://www.pivotaltracker.com/story/show/54196760
Andrew Woods added "Ensure nodes overwrite access policy of ancestors in jcr tree" https://www.pivotaltracker.com/story/show/54196776
Andrew Woods added "Access should be controlled when resource is requested directly" https://www.pivotaltracker.com/story/show/54196796
Andrew Woods added "Access should be controlled when resource is searched" https://www.pivotaltracker.com/story/show/54196804
Andrew Woods added "Content owner should be able to update access policy on his/her items" https://www.pivotaltracker.com/story/show/5419681010:05
Andrew Woods added "Repository admin should be able to update access policy on his/her items" https://www.pivotaltracker.com/story/show/54196816
Andrew Woods added "Access rights should include: read and write" https://www.pivotaltracker.com/story/show/54196834
Andrew Woods added "Access policies should be stored in jcr" https://www.pivotaltracker.com/story/show/54196848
Andrew Woods added this label to 5 stories https://www.pivotaltracker.com/story/show/5419684854196834541968165419681054196804
Andrew Woods added "Create policies with REST-API" https://www.pivotaltracker.com/story/show/5419741210:07
Andrew Woods added "Create policies based on properties" https://www.pivotaltracker.com/story/show/5419744810:08
Andrew Woods added "Create multiple policies that are all applied" https://www.pivotaltracker.com/story/show/54197458
Andrew Woods added "Policies should allow for storing the same item in multiple stores" https://www.pivotaltracker.com/story/show/54197470
Andrew Woods added this label to 4 stories https://www.pivotaltracker.com/story/show/54197470541974585419744854197412
Andrew Woods removed this label from 4 stories https://www.pivotaltracker.com/story/show/54197470541974585419744854197412
Andrew Woods added this label to 4 stories https://www.pivotaltracker.com/story/show/54197470541974585419744854197412
Andrew Woods unstarted "Create new, specific performance enhancement tickets based on the profiling investigation" https://www.pivotaltracker.com/story/show/5156786110:12
Andrew Woods edited "Create new, specific performance enhancement tickets based on the profiling investigation" https://www.pivotaltracker.com/story/show/51567861
Andrew Woods unstarted "Run profiling against current fcrepo4 master and look for any easy bottlenecks to clear off" https://www.pivotaltracker.com/story/show/52325659
Andrew Woods unstarted "Run benchmarks on SCC cluster at FIZ Karlsruhe with fcrepo3 and fcrepo4" https://www.pivotaltracker.com/story/show/5228954510:13
Andrew Woods unstarted "Large scale ingests fail on the SCC cluster" https://www.pivotaltracker.com/story/show/52981805
Andrew Woods unstarted "Published events are missing a resolvable location for the resources" https://www.pivotaltracker.com/story/show/5286074910:14
Andrew Woods edited "Published events are missing a resolvable location for the resources" https://www.pivotaltracker.com/story/show/52860749
* ajs6f leaves10:20
* ajs6f joins10:33
* jongibson joins10:43
<cbeer>ajs6f: i'm hammering away at the rdf:type expressions for mixins.10:46
if i assert <some-object> a mixinType:ThatDoesntExist
should we helpfully create that mixin? or declare it a problem? or throw an error?10:47
* barmintor joins
<ajs6f>M. Good one.10:48
I can see different people wanting different things.
<cbeer>it gets even weirder when i say..10:49
<ajs6f>Maybe (I know it would be annoying) this is a configurable thing? Validate/no-validate? Or maybe you can throw an HTTP header for "force"?
<cbeer><some-object> a <http://something/somewhere/that/is/a/real/rdf/Class>
<ajs6f>Oh, fooey. I hadn't thought about that.
<cbeer>so we expect fcrepo4 to just nicely persist that for us
<ajs6f>We don't want to represent the whole SemwWeb in our mixin registry.
And we don't want magic URIs for "internal" types.10:50
Maybe we take the "disengaged" way for now:10:51
We expose the mixins via HTTP.
And people can use those URLs if they want.
And we check any incoming rdf:type assertion,10:52
and if it matches a mixin URL, create a mixin on that node.
Does that make sense for the second problem?
<cbeer>and if it doesn't, just store it as a property?
We're not doing coreference.10:53
AAt least, not as a core functionality.
<cbeer>not unreasonable.. there's something funny when you store it as a property and then come along later and make it a JCR mixin
but i'm happy to ignore that
<ajs6f>I would want that to happen during the synchronous API call.
Oh, wait!
I see what you mean.
If I haven't declared the mixin yet.10:54
But then I do, after I already used it.
Used it in the sense of rdf:type.
<cbeer>exactly. i'm happy to ignore that for now, though
<cbeer>as you describe makes the initial impl easier, at least.
<ajs6f>Time for the INFERENCING SEQUENCER!
It's a start.10:55
<cbeer>i don't need to worry about things that are primary types, or things that we don't have control over
<ajs6f>I think we should be pushing people away from p-types towards mixins.
p-types feel like the internal administrative type system, with mixins more like the trad content modeling that we've done for over a decade.10:56
<ajs6f>Semantic MediaWiki's logging contains terms like "mContainsOldMagic". That's positively Tolkienish.
<barmintor>"we've attempted to do for over a decade" maybe.
<ajs6f>Speak for Columbia. We model the living daylights out of our content.
We model those bits to within an inch of their lives.10:59
* ajs6f1 joins
* ajs6f leaves
<ajs6f1>We hold them down and model them until the blood runs freely.11:00
* daniellamb leaves11:11
* daniellamb joins11:17
* ajs6f1 leaves11:23
* ajs6f joins11:29
<pivotal-bot>Andrew Woods edited "Access policies should be able to be stored in jcr" https://www.pivotaltracker.com/story/show/5419684811:38
<ksclarke>@karma jcr
<cbeer>ajs6f: i've found an.. interesting.. jcr feature12:14
so.. in our CND, we had something like:
[fedora:resource] > fedora:relations, mix:created, mix:lastModified, mix:lockable, mix:versionable mixin
- * (STRING) multiple COPY
and mix:created defines:
[mix:created] mixin
- jcr:created (date) protected
- jcr:createdBy (string) protected
turns out our "* (String) COPY" overrode the protected attribute on jcr:createdBy12:15
which we were trying to explicitly set in FedoraResourceā€¦ i guess i'll stop doing that?
<ajs6f>So we're not supposed to be able to mutate it, but we accidentally did?
<cbeer>exactly. and I went in and changed (STRING) to (undefined).. which does not override the mixin types12:16
i don't think we need to set createdBy explicitly any more anyway.. at least, if we have authn wired into the injected session stuff
<ajs6f>Sounds good to me. Is this a bug in MODE, or expected behavior?12:17
<cbeer>i'm not sure. probably expected behavior, though
<cbeer>it's at least convenient behavior for us :)12:18
* gregjansen1 joins12:30
* edInCo joins12:35
<pivotal-bot>Andrew Woods edited "Allow multiple decisions to be applied" https://www.pivotaltracker.com/story/show/5419745812:40
<cbeer>ajs6f: 09:37 ] cbeer> am I missing something, or should PropertyType.URI be in JcrValue#equals? https://github.com/ModeShape/modeshape/blob/master/modeshape-jcr/src/main/java/org/modeshape/jcr/JcrValue.java#L22912:42
<ajs6f>As long as there is a def't of equals for URI. WHich ther eis.12:44
<pivotal-bot>Andrew Woods edited "Update policy driven storage to move configuration from Spring into the JCR tree." https://www.pivotaltracker.com/story/show/4905979912:53
* gregjansen1 leaves13:03
<pivotal-bot>Chris Beer accepted "MODE-1997" https://www.pivotaltracker.com/story/show/5389466713:16
<cbeer>and 3.4 is due out tomorrow, so i won't bother with a snapshot, i guess
as long as mode-1998 is fixed
<ajs6f>You ure they'll fix that before tomorrow?13:17
<cbeer>seems like a pretty big bug.. or we've misunderstood something13:19
<ajs6f>Well, why not.
* github-ff joins13:30
[fcrepo4] cbeer pushed 8 new commits to master: http://git.io/mIfCyg
fcrepo4/master 9514f4d Chris Beer: check for a last modified date before using it in caching headers
fcrepo4/master 252e754 Chris Beer: pre-define the RDF namespace.
fcrepo4/master 1b7c317 Chris Beer: handle rdf:type with special mixin logic
* github-ff leaves
* travis-ci joins13:47
[travis-ci] futures/fcrepo4#936 (master - dc01984 : Chris Beer): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/cdf0dc01ea42...dc01984e414e
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/9613637
* travis-ci leaves
<bljenkins>Yippie, build fixed!13:48
Project fcrepo4 build #1071: FIXED in 18 min: http://ci.fcrepo.org/jenkins/job/fcrepo4/1071/
* Chris Beer: check for a last modified date before using it in caching headers
* Chris Beer: pre-define the RDF namespace.
* Chris Beer: handle rdf:type with special mixin logic
* Chris Beer: add Link[rel=meta] header to content requests
* Chris Beer: jcr:createdBy is a protected property, and will be auto-populated from the Session attributes
* Chris Beer: allow fedora:resource to have any type of properties, not just strings
* Chris Beer: pass in an explicit type when creating jcr properties
* Chris Beer: pass in a namespace prefix map to the rdf => jcr property translator
<cbeer>MODe-1998 is a blocker for 3.4
so i guess that's good
<bljenkins>Project fcrepo-fixity-corrupter build #207: SUCCESS in 56 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-fixity-corrupter/207/
* ajs6f leaves13:52
<bljenkins>Project fcrepo-kitchen-sink build #474: SUCCESS in 5 min 54 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-kitchen-sink/474/13:55
* ajs6f joins13:58
* ajs6f leaves
<pivotal-bot>Chris Beer added "mode-1998" https://www.pivotaltracker.com/story/show/5421944414:02
Chris Beer started "mode-1998" https://www.pivotaltracker.com/story/show/54219444
* gregjansen joins14:28
* ajs6f joins14:47
* bljenkins leaves15:30
* ajs6f leaves16:04
* ajs6f joins16:10
<pivotal-bot>Chris Beer added "Use ETags on ALL THE THINGS" https://www.pivotaltracker.com/story/show/5423195816:12
Chris Beer estimated "Use ETags on ALL THE THINGS" as 1 point https://www.pivotaltracker.com/story/show/54231958
Chris Beer started "Use ETags on ALL THE THINGS" https://www.pivotaltracker.com/story/show/54231958
Chris Beer added "The root node should have an etag value" https://www.pivotaltracker.com/story/show/54231974
Chris Beer accepted "mode-1998" https://www.pivotaltracker.com/story/show/54219444
Chris Beer added comment: "https://github.com/ModeShape/modeshape/pull/895" https://www.pivotaltracker.com/story/show/5347902916:22
Chris Beer finished "Help MODE track down bug in initial content import" https://www.pivotaltracker.com/story/show/53479029
<ksclarke>so pivotaltracker is the ticket system?16:36
used by ff16:37
just a note, the link to the Fedora Futures wiki from within pivotaltracker (https://www.pivotaltracker.com/projects/684825/overview) is bad16:40
* github-ff joins
[fcrepo4] cbeer pushed 1 new commit to master: http://git.io/DJ_gsg
fcrepo4/master 8b3be13 Chris Beer: mint etags for all fedora resources and use them for conditional requests.
* github-ff leaves
* ajs6f leaves16:55
* travis-ci joins17:02
[travis-ci] futures/fcrepo4#937 (master - 8b3be13 : Chris Beer): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/dc01984e414e...8b3be1340b95
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/9621290
* travis-ci leaves
<cbeer>i think we need to revisit the semantics of POST /some/path/to/an/objecg17:13
<barmintor>intermediate objects, or just updates?17:14
<cbeer>i'm not sure we're actually in conflict with https://dvcs.w3.org/hg/ldpwg/raw-file/default/ldp.html#ldpc-HTTP_POST
but we aren't nicely aligned to it either.
barmintor: as I read LDP, a POST to a Container should create a new object (like POST /some/path/to/an/object/fcr:new now)17:15
if you want updates, go use PUT
<barmintor>right, right
<cbeer>and LDP suggests creating an object as a named path should be a PUT anyway
<cbeer>4.5.6 LDPR servers may choose to allow the creation of new resources using HTTP PUT.
i don't think there's a problem dropping POST-based updates
(in that.. i don't think we had that anyway. or, if we did, it is with SPARQL-Update and we can keep both)17:17
but how important are POST-based creates?
(or, I think it'd be weird to have POST-based creates and POST-based creates-child-object17:18
<bljenkins>Project fcrepo-fixity-corrupter build #208: SUCCESS in 3 min 18 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-fixity-corrupter/208/17:21
Project fcrepo-kitchen-sink build #475: SUCCESS in 7 min 53 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-kitchen-sink/475/17:29
<cbeer>huh. i didn't realize you could POST binary payloads to /just/the/node/path and it'd create nodes for you17:34
s/create/nodes/create fcr:content nodes/
<awoods>cbeer: were you expecting an error?17:35
<cbeer>awoods: i thought that's what the fcr:content endpoint was for17:36
<awoods>...making the fcr:content endpoint redundant?17:37
<cbeer>i thought fcr:content was the one we liked, though. because it's symmetric and all that.17:38
i can POST/PUT content to a URL, and GET it from that same URL
<awoods>so maybe POSTing a binary to /some/path should be more restrictive...17:39
there is also the question you raise about changing POST to PUT, as well.
* ksclarke leaves18:42
<pivotal-bot>Chris Beer added comment: "https://github.com/futures/fcrepo4/commit/8b3be1340b95edc861da58a73a54916a3dad06e9" https://www.pivotaltracker.com/story/show/5423195819:01
Chris Beer delivered "Use ETags on ALL THE THINGS" https://www.pivotaltracker.com/story/show/54231958
* github-ff joins19:13
[fcrepo4] cbeer created remove-fcr-new (+1 new commit): http://git.io/pu2G4Q
fcrepo4/remove-fcr-new 0aba466 Chris Beer: replace .../fcr:new with a POST to a Node
* github-ff leaves
* edInCo leaves19:22
* barmintor_ joins19:25
* barmintor leaves19:27
<bljenkins>Project fcrepo-fixity-corrupter build #209: SUCCESS in 50 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-fixity-corrupter/209/19:30
* github-ff joins19:33
[fcrepo4] cbeer opened pull request #104: replace .../fcr:new with a POST to a Node (master...remove-fcr-new) http://git.io/b2afTw
* github-ff leaves
<bljenkins>Project fcrepo-kitchen-sink build #476: SUCCESS in 7 min 58 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-kitchen-sink/476/19:38
* edInCo joins20:04
* jongibson leaves20:14
<pivotal-bot>Andrew Woods delivered "Help MODE track down bug in initial content import" https://www.pivotaltracker.com/story/show/5347902920:35
Andrew Woods accepted "Help MODE track down bug in initial content import" https://www.pivotaltracker.com/story/show/5347902920:36
Andrew Woods edited "Modify a storage policy" https://www.pivotaltracker.com/story/show/4825607320:42
Andrew Woods edited "Modify a storage policy at runtime" https://www.pivotaltracker.com/story/show/48256073
Andrew Woods estimated "Create policies with REST-API" as 3 points https://www.pivotaltracker.com/story/show/54197412
Andrew Woods edited "Create policies with REST-API" https://www.pivotaltracker.com/story/show/54197412
Andrew Woods deleted "Create a storage policy in the repository" https://www.pivotaltracker.com/story/show/4843741920:44
Andrew Woods added "Ensure invalid storage policies are rejected" https://www.pivotaltracker.com/story/show/5425051820:45
Andrew Woods edited "Ensure invalid storage policies are rejected" https://www.pivotaltracker.com/story/show/54250518
Andrew Woods added "Define and Document Storage Policy Description Language" https://www.pivotaltracker.com/story/show/5425054820:46
Andrew Woods edited "Define and Document Storage Policy Description Language" https://www.pivotaltracker.com/story/show/54250548
Andrew Woods added "Create iRODS Connector" https://www.pivotaltracker.com/story/show/5425057420:47
Andrew Woods added "Create Strawman AuthN/Z Design" https://www.pivotaltracker.com/story/show/5425063620:49
Andrew Woods edited "Create Strawman AuthN/Z Design" https://www.pivotaltracker.com/story/show/54250636
* github-ff joins20:51
[fcrepo4] ajs6f pushed 1 new commit to master: http://git.io/Acl_Qg
fcrepo4/master 88bfddc ajs6f: Merge pull request #104 from futures/remove-fcr-new...
* github-ff leaves
* ksclarke joins21:00
* travis-ci joins21:06
[travis-ci] futures/fcrepo4#939 (master - 88bfddc : ajs6f): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/8b3be1340b95...88bfddc8c7a4
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/9628982
* travis-ci leaves
<bljenkins>Project fcrepo-fixity-corrupter build #210: SUCCESS in 2 min 23 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-fixity-corrupter/210/21:23
Project fcrepo-kitchen-sink build #477: UNSTABLE in 7 min 47 sec: http://ci.fcrepo.org/jenkins/job/fcrepo-kitchen-sink/477/21:31
* kaarefc joins21:54
* edInCo leaves23:06
* edInCo joins23:26

Generated by Sualtam