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

Using timezone: Eastern Standard Time
* thomz joins02:47
* cmmills leaves05:17
* jackhill leaves
* anarchivist leaves
* ajwagner leaves
* barmintor leaves05:18
* f4jenkins leaves
* awoods leaves05:19
* cbeer leaves
* awoods joins06:28
* f4jenkins joins
* barmintor joins
* ajwagner joins
* jackhill joins
* anarchivist joins
* cmmills joins
* 7GHAAEYAP joins
* cbeer joins
* dwilcox joins07:46
* 7GHAAEYAP leaves08:09
* dhlamb joins08:35
* mohamedar joins09:01
* bseeger joins09:06
* ajs6f joins09:13
* youn joins09:14
* coblej joins09:17
* dwilcox leaves09:21
* github-ff joins
[fcrepo4] ajs6f closed pull request #1119: Upgrade to guava 19.0 (master...fcrepo-1959) https://git.io/vP2cF
* github-ff leaves
* dwilcox joins09:23
* acoburn joins09:24
* mikeAtUVa joins09:27
<youn>I have been following the discussion on Fedora releases and am wondering if it would make sense to coordinate releases (schedules, features, testing) with front-ends, representing different target audiences. This could help make migration easier and perhaps allow Fedora developers to organize their work to focus on certain feature sets.09:29
<awoods>youn: Yes, that goal was part of the motivation for the thread... although the voices on the thread have not echoed that point.09:32
* peichman joins09:33
<ajs6f>youn:awoods: That privileges the voices of those who happen to use recognizable front-ends in a way with which I am not at all comfortable.09:34
* travis-ci joins09:40
fcrepo4/fcrepo4#4771 (master - 9a22d76 : Aaron Coburn): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/compare/1288372228d1...9a22d764e467
Build details : https://travis-ci.org/fcrepo4/fcrepo4/builds/167355244
* travis-ci leaves
* osmandin joins09:42
* dwilcox leaves09:43
* coblej leaves09:45
* coblej joins09:51
* dwilcox joins
* coblej leaves09:55
* github-ff joins09:59
[fcrepo4] ajs6f pushed 1 new commit to master: https://git.io/vPVXC
fcrepo4/master a11360d Aaron Coburn: Add core types to be used to replace the use of jcr.Session in the kernel API (#1107)...
* github-ff leaves
* coblej joins10:02
* acoburn leaves10:05
* coblej leaves10:07
* coblej joins10:17
* travis-ci joins
fcrepo4/fcrepo4#4772 (master - a11360d : Aaron Coburn): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/compare/9a22d764e467...a11360d0413a
Build details : https://travis-ci.org/fcrepo4/fcrepo4/builds/167365041
* travis-ci leaves
* github-ff joins10:18
[fcrepo4] awoods opened pull request #1122: Ensure rest api returns 507 response on No Space Left (master...fcrepo-1931) https://git.io/vPVDE
* github-ff leaves
* github-ff joins
[fcrepo4] awoods closed pull request #1115: Ensures that the rest api will return a 507 response with a message body (master...fcrepo-1931) https://git.io/vP0tf
* github-ff leaves
* coblej leaves10:22
* osmandin leaves10:24
* coblej joins10:27
* acoburn joins10:39
* apb18 joins10:46
* osmandin joins10:48
* dhlamb leaves10:51
* youn leaves10:53
* dhlamb joins10:56
* bseeger leaves10:59
* bseeger joins
<ajs6f>On my way! to call11:00
<awoods>https://wiki.duraspace.org/display/FF/2016-10-13+-+Fedora+Tech+Meeting
* dhlamb is here
* coblej is here11:01
<bseeger>*is here*
<acoburn>*is here*
* barmintor is here11:02
* ajs6f is here11:05
* thomz leaves
<ajs6f>I can prolly do that, bt not for a few days.11:08
* ajs6f is same as barmintor11:09
Yep.
If you need a third, you can ping me.
<osmandin>Same here, but not until next week.
<barmintor>acoburn: thank you so much for the work on this release, that's some good stuff
<ajs6f>what barmintor says11:10
acoburn++
Freedome from JCR is freedom to thrive.
<acoburn>barmintor: just wait until we have the JCR dependencies removed :-)
<barmintor>awoods: let's ping esme11:13
* github-ff joins11:18
[fcrepo-camel] ajs6f pushed 2 new commits to master: https://git.io/vPVx8
fcrepo-camel/master dbb2701 Aaron Coburn: Use an in-memory broker configuration for test suite...
fcrepo-camel/master f5c625a A. Soroka: Merge pull request #127 from acoburn/fcrepo-2264...
* github-ff leaves
<ajs6f>We should do a point release for every commit.11:20
4.7.1, 4.7.2, 4.7.3 … 4.7.9308259408503948534689011:21
<barmintor>awoods: do we have a countdown latch test that demonstrates this bug?11:22
or something similar
<ajs6f>And we don't have crowds of volunteers to test this stufff.
Throw lots of 409s.11:23
<barmintor>1. Verification test11:24
* travis-ci joins11:25
fcrepo4-exts/fcrepo-camel#302 (master - f5c625a : A. Soroka): The build passed.
Change view : https://github.com/fcrepo4-exts/fcrepo-camel/compare/397ad6fb92e3...f5c625ab665c
Build details : https://travis-ci.org/fcrepo4-exts/fcrepo-camel/builds/167388204
* travis-ci leaves
<barmintor>2. Heavyweight change sufficient
<ajs6f>Just firing a lot of threads joiing each other.
* apb18 joins the call late
<barmintor>3. Backport may go out first as an incident of release process
<ajs6f>We need a proper semaphore or the like, or use Awaitility with AtomicBooleans.
<barmintor>4. After the fix is demonstrated, we can re-evaluate synchronization strategy to be more focused11:26
in a subsequent patch release
ajs6f++ // yeah, there are some strategies for testing race conditions we should be able to address if the test env is multiprocessor11:27
<dhlamb>echo11:28
barriers?11:29
<barmintor>mikeAtUVa++
<ajs6f>mikeAtUVa: Good stuff here: https://github.com/awaitility/awaitility/wiki/Usage11:30
<mikeAtUVa>I think the backport should wait on the improved integration tests.11:31
<ajs6f>mikeAtUVa: Simple example: https://github.com/fcrepo4/fcrepo4/blob/master/fcrepo-jms/src/test/java/org/fcrepo/integration/jms/observer/AbstractJmsIT.java
<mikeAtUVa>We knew it was an issue for the last few releases... we just didn't have anyone willing to fix it.11:32
<barmintor>mikeAtUVa: and thank you for that branch!
<ajs6f>mikeAtUVa: example I did with Atomics for Jena : https://github.com/apache/jena/blob/master/jena-arq/src/test/java/org/apache/jena/sparql/core/mem/TestDatasetGraphInMemoryThreading.java11:33
NO ONE has any confidence in INSP.
<barmintor>c'est vrai
acoburn suggests that ISPN is as much a data integrity threat as the pairtree synchronization, and separably patchable11:34
<ajs6f>^^^11:35
So we have two vicious dogs biting our rear, and the question is which do we kick first?
<barmintor>which is hard to dispute, but also like acoburn I would wait on 4.7.1
<mikeAtUVa>It seems like the choice is definitely have two critical releases this fall, or poissibly have only one.11:36
<ajs6f>mikeAtUVa speaks for UVa, but I don't think we're going anywhere until we leave INP behind.
"This time, we really really really promise it'll be stable."11:37
<barmintor>eek @ "consider the pairtree algorithm before multithreading"11:38
<ajs6f>I'm in favor of minimizing the number of times we have to loudly say ZOMG don't use that repo it will poison your cat!
* barmintor cries a single tear
<ajs6f>Fedora 4: Embarrassing but necessary.11:40
<mikeAtUVa>If someone can write the better integration tests quickly.11:41
<ajs6f>I can look at that next week.11:42
But not before.
* mikeAtUVa would sign off on either approach.
<ajs6f>mikeAtUVa: Haven't we done that?11:45
barmintor: I have a solution for that, named Cavendish.11:46
<mikeAtUVa>ajs6f: we run backup from modeshape 4 and restore to modeshape 5 (on a different server) every single night.
<ajs6f>mikeAtUVa: Damn, we're hot!
<mikeAtUVa>ajs6f: we have to use esme's fix, but that's incorporated into the script.
<ajs6f>mikeAtUVa: But our repo is really small, right.
<mikeAtUVa>ajs6f: Yes... 2GB... but they're PRICELESS theses and dissertations!11:47
<barmintor>you could also hypothetically dodge this by skipping the binary export except that is specifically the the thing we are trying to do
* mohamedar leaves11:48
<ajs6f>mikeAtUVa: {bites tongue}
* mohamedar joins
* jenncolt joins
<ajs6f>-1
barmintor: So you are talking about a "Fedora Query" spec?11:50
ancillary to the main body of specs?11:51
Seems more like a "closure for retrieval" thing, not a "query" thing.
<apb18>Is there a document that describes this?11:52
<barmintor>https://docs.google.com/document/d/1Jv0X30SNBO8foT_c-5Hi0uz1V46UavuxJTyfeqW1LkI/edit#11:53
<ajs6f>This still carries a bunch of assumptions about the whole system (assumptions about global behavior).
<jenncolt>permissions on that
<ajs6f>barmintor: That doesn't appear to be public? Or maybe it's just me.
<barmintor>https://docs.google.com/document/d/1Jv0X30SNBO8foT_c-5Hi0uz1V46UavuxJTyfeqW1LkI/edit?usp=sharing
<jenncolt>ty11:54
<ajs6f>Hopefully "subresources by type, including subresources of root node" is in that category of stuff you went aay from.
Beause that is general query territory.
<dhlamb>ajs6f: would just be a convienence at that point, not performance boost if i understand you correctly11:56
<ajs6f>right
<apb18>It would be interesting to use API-X to sketch out an adapter to do that, and see how it goes11:58
<ajs6f>I think barmintor has informationalized this thing like a jackhammer.12:01
barmintor++
<barmintor>lol
<ajs6f>Inspiring words.12:02
* jenncolt_ joins
* osmandin leaves
* ajs6f leaves12:03
* jenncolt leaves
<barmintor>acoburn: I think, fwiw, that ldpath would be a backend for an adapter here- it would be able to satisfy everything we've discussed right now, but has a stronger affinity to particular backends than we envision of Fedora Impls (right?)
<acoburn>barmintor: I've worked with a number of different LDPath backends, and, really, the best one is marmotta's ldcache12:04
barmintor: I have an implementation of this that sits independently of Fedora, but which is set up to interact directly with a Fedora repo12:05
barmintor: …and the entire linked data cloud
barmintor: effectively, it exposes an endpoint that is proxied to a Fedora resource and indicates an LDPath program (either via query param or via POST body)12:06
barmintor: the result of the HTTP operation is a JSON document with the LDPath program applied to the Fedora resource (and any other resources that were followed)12:07
<dhlamb>apb18: api-x meeting not on same line as fcrepo-tech call?
<acoburn>barmintor: the current implementation uses a filesystem backend for caching data, but that is entirely swap-out-able
<barmintor>acoburn: that's all very interesting- I was not sure how repurposable Marmotta's impl was outside MArmotta12:08
<apb18>dlamb: It's the same conference line/number/credentials. Did I put a typo in the agenda page?
<acoburn>barmintor: yeah, I'm just using the marmotta libraries and embedding them into my own code12:09
* bseeger leaves
<dhlamb>apb18: no, i think maybe i'm forgetting timezones
apb18: looking at your email, that's 1pm EST
apb18: and i'm thinking AST....
* dhlamb facepalms
apb18: see you in an hour
<apb18>dhlamb: No worries, you just gained an hour!12:10
* dhlamb feels the glass is half full
* jenncolt_ leaves12:11
* coblej leaves12:13
* coblej joins12:19
* coblej leaves12:24
* github-ff joins12:27
[fcrepo4] acoburn opened pull request #1123: Adieu jcr.Session, it's been nice knowing you (master...fcrepo-1868.2) https://git.io/vPwtn
* github-ff leaves
<dhlamb>awoods: notes in agenda page12:28
<awoods>thanks, dhlamb12:29
* coblej joins12:31
* coblej leaves12:35
* acoburn leaves12:36
* coblej joins12:37
* acoburn joins12:38
* coblej leaves12:42
* github-ff joins
[fcrepo4] awoods pushed 1 new commit to master: https://git.io/vPwmr
fcrepo4/master 2b41b3d Aaron Coburn: Remove sonatype's oss-parent pom (#1118)...
* github-ff leaves
* coblej joins12:44
* coblej leaves12:49
* coblej joins
* ddavis joins12:51
* travis-ci joins13:00
fcrepo4/fcrepo4#4777 (master - 2b41b3d : Aaron Coburn): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/compare/a11360d0413a...2b41b3d3838c
Build details : https://travis-ci.org/fcrepo4/fcrepo4/builds/167412803
* travis-ci leaves
* github-ff joins13:01
[fcrepo-message-consumer] jgreben opened pull request #97: Update pom.xml (master...patch-1) https://git.io/vPw3S
* github-ff leaves
<awoods>apb18: are people having trouble connecting?13:03
<apb18>awoods: not that I am aware of, are you?
<awoods>apb18: yes
* mjgiarlo leaves13:05
<apb18>https://docs.google.com/document/d/10Zv_IpCwItMEuomB6Xpoy5d3x2914c6KzP9s5IcLzHk/edit
* coblej leaves13:08
* mohamedar leaves13:09
* ylchen joins13:10
* mjgiarlo joins
<dhlamb>great job apb18, can't wait to give it a whirl13:11
* bseeger joins13:13
<dhlamb>apb18: i can try that for you
<apb18>https://github.com/fcrepo4-labs/fcrepo-api-x/pull/4213:14
<awoods>[INFO] ------------------------------------------------------------------------13:19
[INFO] Reactor Summary:
[INFO]
[INFO] fcrepo-api-x ....................................... SUCCESS [ 1.538 s]
[INFO] fcrepo-api-x-model ................................. SUCCESS [ 5.417 s]
[INFO] fcrepo-api-x-registry .............................. SUCCESS [ 6.885 s]
[INFO] fcrepo-api-x-jena .................................. SUCCESS [ 7.251 s]
[INFO] fcrepo-api-x-routing ............................... SUCCESS [ 6.540 s]
[INFO] fcrepo-api-x-binding ............................... SUCCESS [ 2.864 s]
[INFO] fcrepo-api-x-execution ............................. SUCCESS [ 0.686 s]
[INFO] fcrepo-api-x-karaf ................................. SUCCESS [ 1.128 s]
[INFO] fcrepo-api-x-test .................................. SUCCESS [ 0.649 s]
[INFO] fcrepo-api-x-integration ........................... SUCCESS [02:04 min]
[INFO] fcrepo-api-x-ontology .............................. SUCCESS [ 0.431 s]13:20
[INFO] ------------------------------------------------------------------------
[INFO] BUILD SUCCESS
[INFO] ------------------------------------------------------------------------
[INFO] Total time: 02:53 min
[INFO] Finished at: 2016-10-13T13:18:44-04:00
[INFO] Final Memory: 64M/479M
[INFO] ------------------------------------------------------------------------
git status
On branch pull/42
Your branch is up-to-date with 'origin/pull/42'.
apb18 ^^
* jcolt joins13:31
<dhlamb>lost audio :(13:32
re-joining call
acoburn++13:37
kar++
wrap your features up as an archive++
* ylchen leaves13:42
<dhlamb>apb18: posted my build results on your PR
* mohamedar joins13:50
* youn joins14:00
* github-ff joins14:04
[fcrepo-message-consumer] acoburn opened pull request #98: Add deprecation message (master...update_readme) https://git.io/vPw08
* github-ff leaves
* jcolt leaves
<acoburn>awoods: that's for you ^^^14:05
<awoods>on a call
* acoburn leaves
* acoburn joins
* ddavis leaves14:10
* osmandin joins14:36
* youn leaves14:40
* ajs6f joins14:55
* mohamedar leaves15:03
* mohamedar joins15:09
* mohamedar leaves15:17
* mohamedar joins15:31
* dhlamb leaves16:06
* bseeger leaves
* bseeger joins16:10
* mohamedar leaves16:15
* apb18 leaves16:22
* osmandin leaves16:24
* anarchivist leaves16:28
* anarchivist joins16:32
* youn joins16:37
Is the query service intended to be for SPARQL queries? Will there be constraints on the queries that may be used?16:39
<ajs6f>I don't think the people who intend to suggest it actually have any real idea of what they are going to suggest. That's not being snarky— it's just really, really early.
barmintor made the point that he was only bringing it up to avoid surprise later.16:40
<youn>ajs6f: One way to counter the tendency to privilege users of recognizable front-ends might be to have workshops or hackathons with focus groups at the end to bring code in the wild into release pipelines.16:44
<ajs6f>youn: Focus groups? Not sure what you mean. Showing people code that does random things and asking them how they feel about it?16:45
* apb18 joins
<youn>ajs6f: Focus groups in the morning; code revisions after lunch. Focus groups may be the wrong term -- someone to point out something is or is not working.16:46
<ajs6f>youn: Are you talking about try to get voices from workshops into the planning for the long-term future of the product, or into the short-term decision making for "what's in the next release"?16:47
<youn>ajs6f: I have heard that innovation happens around the edges.16:48
<ajs6f>youn: Okay. Are we trying to encourage innovation generically? I'm not that interested in so doing, myself. I want to move Fedora down a tree of paths that has a pretty strong central "branch", which is a set of ideas about how to make information durable that date back to the beginning of the project.16:50
youn: But I like the idea of hearing as much as possible from workshop participants, because they demonstrate commitment just by coming to the workshops.
youn: Got to run, but maybe we can talk more about this next week. Maybe you will put this on the agenda for the next tech meeting?16:52
<youn>ajs6f: happy to16:53
<ajs6f>youn++
o-&-o16:54
* ajs6f leaves
* dwilcox leaves16:55
* dwilcox joins
* dwilcox leaves
* dwilcox joins16:56
* dwilcox leaves
* dwilcox joins16:57
* dwilcox leaves
<bseeger>awoods: I'm seeing the random unit test failure as well (failed first time, succeeded the second): https://gist.github.com/bseeger/64d295bdc546fed2a77ee51675f28b4417:00
awoods: I was looking at acoburn's latest PR on my local machine17:01
* acoburn leaves17:06
* bseeger leaves
* github-ff joins17:09
[ontology] acoburn created fcrepo-1936 (+1 new commit): https://git.io/vPrer
ontology/fcrepo-1936 45a3aa2 Aaron Coburn: Remove JCR-sourced and other unused properties from ontology...
* github-ff leaves
* github-ff joins
[ontology] acoburn opened pull request #41: Remove JCR-sourced and other unused properties from ontology (master...fcrepo-1936) https://git.io/vPreP
* github-ff leaves
* mikeAtUVa leaves17:13
* youn leaves17:18
* peichman leaves
* apb18 leaves18:12
* github-ff joins19:49
[fcrepo-message-consumer] awoods pushed 2 new commits to master: https://git.io/vPrCO
fcrepo-message-consumer/master ccc8073 Aaron Coburn: Add deprecation message
fcrepo-message-consumer/master c303f10 Andrew Woods: Merge pull request #98 from acoburn/update_readme...
* github-ff leaves
* mikeAtUVa joins20:56
* mikeAtUVa leaves21:31

Generated by Sualtam