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

Using timezone: Eastern Standard Time
* mcritchlow leaves00:42
* rbenji joins02:17
* rbenji leaves04:04
<dojobo> hi, i'm running the jar version just to play around, and i tried to import an xml file whose root element is <resource> (and it is defined in the xsd)06:07
but when i try to import, i get "Unrecognized element "{http://datacite.org/schema/kernel-3}resource", in import XML."06:08
any ideas?06:09
(oh and this is the user-friendly interface of the webapp)06:10
* jrgriffiniii leaves06:17
* jrgriffiniii joins06:18
* escowles_ leaves06:24
* jrgriffiniii leaves06:42
* jrgriffiniii joins
* dwilcox joins07:30
* jrgriffiniii leaves08:29
* bseeger joins08:45
* jrgriffiniii joins08:53
* rbenji joins09:01
* whikloj joins
* ajs6f joins09:08
* bseeger leaves09:33
* bseeger joins09:34
* dwilcox leaves09:35
* bseeger leaves09:51
<awoods>dojobo: If you would not mind posting your question and associated XML document to fedora-tech@googlegroups.com, I would be happy to take a look.09:56
* dwilcox joins
* jrgriffiniii leaves09:57
* bseeger joins09:58
* jrgriffiniii joins10:01
* bseeger leaves10:07
<dojobo>thanks awoods :)10:09
i ended up manually writing in some sparql queries to try it out10:10
i was very much in an exploratory phase and hoping to use it as a metadata store for this new project my dept is starting10:11
unfortunately the sales pitch to my boss failed :(
so, i don't think i'll waste anyone's time with the question
* jrgriffiniii leaves10:14
* bseeger joins10:15
<awoods>sorry to hear that, dojobo. What solution does your boss prefer?10:16
* jrgriffiniii joins10:17
* bseeger leaves10:19
* bseeger joins10:20
* ajs6f leaves10:22
* bseeger leaves10:26
<dojobo>well, we're tasked with fetching metadata about some of our org's publications and data sets10:44
transforming it to datacite xml, assigning a doi, and registering it with datacite
(and of course storing said doi's and metadata, and serving nice pages)
my pitch was to use fedora as the data store, whether as a repo for the pdf's or not10:45
(the pdf's are currently provisioned by another dept)
but my boss would just rather some relational tables and php10:46
* ajs6f joins10:49
<awoods>dojobo: thanks for the feedback and interest. Good luck.10:50
<dojobo>she said she wants something simple, and didn't seem interested in durability or linked data. i was also a bit surprised that she didn't consider an IR being part of the library's mission10:51
(she didn't want us to be responsible providing pdf's, for example)
<ajs6f>dojobo: Priorities…10:52
awoods: Do yoiu have the URL for meeting10:53
<awoods>ajs6f: https://wiki.duraspace.org/display/FF/2016-04-21+-+Fedora+Tech+Meeting
<ajs6f>awoods++ # you are the greatest humanitarian this generation has yet produced.
* bseeger joins10:59
* ajs6f is here11:00
<dwilcox>dwilcox is here11:01
<bseeger>can someone post the meeting link…
<bseeger>:) thanks!
* acoburn joins
* whikloj is here
<acoburn>*is here*
<bseeger>* is here *11:02
* escowles joins
* jrgriffiniii leaves
<whikloj>awoods: hi11:03
* jrgriffiniii joins11:04
It's a nice number.11:07
<whikloj>awoods: sure11:08
* bscoleman joins
* barmintor joins11:10
hello sorry I'm late11:12
<escowles>+1 the HTTP headers can be different from the triples when we're talking about the representation changing
<ajs6f>escowles: Sure can. But not the other way, at least not in this case. I suppose, however, there could be a case where the resource has genuinely changed but the representation dosn't capture that.11:14
* jcoyne joins
<ajs6f>escowles: hard to think of an example.
<barmintor>I'm reading this PR now
<escowles>ajs6f: that's kind of the case with inbound links (the resource has changed, but the triples you get without special Prefer headers wouldn't change)
<ajs6f>escowles: Yeah, that gets in to the whole thing about "I could get several different representations at a given URI, depending on headers. Does ETag change when _any_ of them change?"11:15
escowles: I hate that thing.11:16
<escowles>ajs6f: yep, very hard to know what's even the right thing to do, much less implement it
<ajs6f>escowles: This is why I try not to get emotionally involved.
<barmintor>awoods: the thing you just described sounds reasonable to me, since some of the properties will change when the binary is updated.11:17
* bscoleman leaves11:25
<barmintor>I suddenly realize that I am the Asger of Fedora 411:26
<ajs6f>barmintor: What do you mean? You're an OWL fan/.11:27
barmintor: You talk quickly moving your shoulders?
<barmintor>We had a long running talk about Datastreams as first-class objects that was in some part about understanding the stored binary and its versioned properties as distinct entities
<ajs6f>barmintor: There is a powerful earnestness in your facial expressions?
* peichman joins11:28
<ajs6f>barmintor: Oh, I remember part of this. This came out of the introduction o RELS-INT, right?
<barmintor>there is also a powerful earnestness in my facial expression
<ajs6f>barmintor: Don't forget the rugged good looks.
<barmintor>ajs6f: yes, but also in part do interrogating of the model vs the constraints of the AIP serialization11:29
<ajs6f>barmintor: Right. Urg. Let's not get _any_ kind of serialization caught up in this.
barmintor: Serialization requirement, I should say.11:30
* github-ff joins
[fcrepo4] acoburn closed pull request #1022: Tie binary description ETags/lastModified dates to the binaries they describe (DEV...fcrepo-1983) https://git.io/vwLmE
* github-ff leaves
<barmintor>ajs6f: right, but- you have to reckon with how it bleeds out. Much of FCR3 activity was dependent on FOXML's implications, ya know11:31
<ajs6f>barmintor: Oh, yes, the history is clear. I just only want to bring certain things forward from that.11:32
barmintor: In an ideal world, the spec could maintain agnosticism. It remains to be seen if we can be that clever.11:33
<barmintor>"For users which don’t want to store the repository data in a RDBMS and who also aren’t interested in clustering, this should be the default go-to store in ModeShape 5."11:38
* barmintor looks for the one that is clusterable
* barmintor looking
* barmintor looking11:39
<ajs6f>barmintor: I eagerly await the release of your new clusterable Fedora impl.11:43
<barmintor>ajs6f: we kind of had one!
kind of
* barmintor sniffles11:44
<ajs6f>barmintor: We did not make the choices that made this change. We do need to respond to them. I
* bscoleman joins
<ajs6f>barmintor: I don't think we disagree that standardization and reimpling is the real lasting solution.
<barmintor>ajs6f: yeah, but the interim is undefined
<ajs6f>barmintor: And that is different… how?11:45
Fedora 4: The Interim is Undefined.
We _will_ have a policy.
<barmintor>I am kind of curious about how much jboss would charge to have rhauch just give us a "clustered with large blobs" configuration11:46
<ajs6f>barmintor: And can we pay it in beer.
<ajs6f>Got to run. I think acoburn is on the right track here, details to be worlked out.11:48
* ajs6f leaves
<barmintor>of curse sufia pcdm is going to blow up into a thousand sprockets of json-ld so I'm hosed.11:58
afk a sec11:59
thanks awoods !12:02
* acoburn leaves12:03
* jcoyne leaves12:14
* dwilcox leaves12:15
* jcoyne joins
* jcoyne leaves
* jcoyne joins12:16
* jcoyne leaves12:18
* jcoyne joins12:19
* bseeger leaves12:26
* jcoyne leaves12:39
* jcoyne joins12:40
* bscoleman leaves12:47
* dwilcox joins12:51
* bseeger joins
* acoburn joins12:53
* awoods_ joins13:02
* awoods leaves13:04
* jcoyne leaves13:07
* jcoyne joins13:08
* acoburn leaves13:14
* bseeger leaves13:15
* anarchivist leaves13:19
* anarchivist joins13:23
* github-ff joins13:37
[fcrepo4] awoods opened pull request #1026: Remove JCR details from Tombstone (DEV...fcrepo-1997) https://git.io/vwlkd
* github-ff leaves
* ajs6f joins14:07
* barmintor leaves14:22
* osmandin joins14:37
Is there a public demo version of fcrepo running somewhere?14:40
<awoods>osmandin: http://demo.fcrepo.org:8080/fcrepo/
<osmandin>awoods: perfect. Thanks.14:41
* bseeger joins14:42
* github-ff joins14:52
[fcrepo4-vagrant] awoods opened pull request #41: Enable ldp:contains triples for indexing (4.5.1-RC...fcrepo-1849) https://git.io/vwllj
* github-ff leaves
* jackhill leaves14:53
* bseeger leaves14:54
* peichman leaves14:58
* bseeger joins15:00
* ajs6f leaves15:03
* jcoyne leaves
* jackhill joins15:08
* dwilcox leaves15:49
* bseeger leaves16:06
* osmandin leaves16:29
* rbenji leaves16:52
<awoods>reubot: ping17:12
ruebot: ping
<ruebot>awoods: firing one up now.17:29
<awoods>ruebot: thanks
<ruebot>awoods: sorry, submitting festival applications for zorton and the cannibals :-)17:30
<awoods>ruebot: good luck
<ruebot>awoods: which branch do you want me to build? this your pull request, or one of the existing branches?17:31
<awoods>ruebot: 4.5.1-rc or master
ruebot: or my PR...
ruebot: since you are planning on testing that anyways
<ruebot>awoods: fcrepo-1849 it is17:34
awoods: ok, i'm up; virtualbox says 9080 is open17:37
$ curl -i http://localhost:908017:38
curl: (56) Recv failure: Connection reset by peer
<awoods>ruebot: I see the same17:43
ruebot: but if you ssh in... the curl command works.
p.s. Apparently Prince died today.
<ruebot>awoods: yeah. i'm gonna put 1999 on here in a few minutes :-)17:44
awoods: yeah. it does work inside. that's weird.17:45
<awoods>ruebot: I pulled down an older release... and it worked as expected.17:46
<ruebot>awoods: is there some sort of ip restriction on 9080? like localhost, or something?
<awoods>ruebot: I don't think so, because older releases of fcrepo4-vagrant work as expected.
<ruebot>awoods: the reindexing service is camel-toolbox? i forget. acoburn just addressed a few things recently with it, right?17:48
<awoods>ruebot: yes, it is camel-toolbox
ruebot: that is a good point, that service starts up differently now...17:49
<ruebot>awoods: https://github.com/fcrepo4-exts/fcrepo-camel-toolbox/commit/a1a92c87e33696e5a532c2506b9d78bff0592df0#diff-93a25f8ccbf43284222d15ca5637414dR25 -- maybe that?17:50
<awoods>ruebot: https://jira.duraspace.org/browse/FCREPO-1980
<ruebot>awoods: i bet it doesn't see the host of the vm as localhost, that's why
<awoods>ruebot: I think you are on to something17:51
* whikloj leaves17:58
<ruebot>awoods: how should i test your PR? i see lots of stuff in fuseki, but i don't see any ldp:contains18:01
<awoods>ruebot: You should be able to create a container in Fedora named: collection18:03
<ruebot>awoods: nvm. i forget to make something that contains something :-)
<ruebot>awoods: i see <http://www.w3.org/ns/ldp#contains> in fuseki now
awoods: if that's it, shall i merge?
<awoods>ruebot: yes, please.18:04
<ruebot>awoods: done!
* github-ff joins
[fcrepo4-vagrant] ruebot closed pull request #41: Enable ldp:contains triples for indexing (4.5.1-RC...fcrepo-1849) https://git.io/vwllj
* github-ff leaves
<awoods>ruebot: and add a link to the related ticket
<ruebot>awoods: done!18:06
<awoods>ruebot: thanks
<ruebot>awoods: complete aside, all the docker stuff came over yesterday https://hub.docker.com/u/islandora/ -- specifically https://hub.docker.com/r/islandora/claw-fedora/18:07
awoods: complete horizontal scaling put to the test via docker. every component has its own container.18:08
<awoods>ruebot: I should give it a spin. What is CLAW's use of Docker vs. Vagrant?18:09
<ruebot>awoods: i imagine we'll support both. we also have an ansible setup now.18:10
<awoods>ruebot: are you shooting for "the more tooling the better"? or do you have reasons for both docker and vagrant?18:11
<ruebot>awoods: it was a bunch of contract work nigel did, and i signed off on all of it yesterday. the foundation had some onetime money to spend on it, so i took advantage of it.
awoods: we did docker to test the horizontal scaling idea, plus it can be deployed in production should you want it. it generates passwords and what not on build. same with ansible.18:12
awoods: but, i do realize that we have more things to support, and a small community. but, this is already getting folks interested that haven't really be part of the project so far. so, i'll take that as a win!18:13
<awoods>ruebot: that is a win, indeed.18:14
<ruebot>awoods: https://github.com/Islandora-CLAW/claw-docker#quick-start -- if you want to test out my quick start guide :-)18:15
awoods: all that said, i need to update the fedora and camel-toolbox on it all since it is 4.4.018:17
<awoods>ruebot: I will give it a look
awoods: i'm gonna take off now. have a good evening!18:18
* dwilcox joins18:23
* github-ff joins18:28
[fcrepo4-vagrant] awoods opened pull request #42: Enable reindexing service to be exposed to host of vagrant VM (4.5.1-RC...fcrepo-1980) https://git.io/vw8fN
* github-ff leaves
* jrgriffiniii leaves18:34
* github-ff joins19:14
[fcrepo4-vagrant] ruebot pushed 2 new commits to 4.5.1-RC: https://git.io/vw8tE
fcrepo4-vagrant/4.5.1-RC 67dd427 Andrew Woods: Enable reindexing service to be exposed to host of vagrant VM...
fcrepo4-vagrant/4.5.1-RC 2336939 Nick Ruest: Merge pull request #42 from awoods/fcrepo-1980...
* github-ff leaves
* dwilcox leaves20:03
* escowles leaves21:38