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

Using timezone: Eastern Standard Time
* dhlamb leaves00:07
* thomz joins04:30
* jjtuttle_ joins08:16
* jjtuttle leaves08:17
* dhlamb joins08:49
* kefo joins08:56
* mikeAtUVa joins
* whikloj joins09:13
* peichman joins09:19
* apb18 joins09:44
* coblej joins10:01
* bseeger joins10:09
* ajs6f joins10:20
* dwilcox joins10:21
* ajs6f leaves10:49
* thomz leaves10:53
* github-ff joins
[fcrepo-audit] awoods pushed 2 new commits to master: https://git.io/vMziX
fcrepo-audit/master f2bff26 Aaron Coburn: Add test gear for fcrepo integration testing...
fcrepo-audit/master 59138ca Esmé Cowles: Add basic IT...
* github-ff leaves
* ajs6f joins10:54
Can someone give me the phone connection info please?10:55
* github-ff joins10:56
[fcrepo-audit] awoods closed pull request #6: added test gear for fcrepo integration testing (master...fcrepo-1481) https://git.io/vMzPe
* github-ff leaves
<whikloj>ajs6f: https://wiki.duraspace.org/display/FF/2017-01-12+-+Fedora+Tech+Meeting
* github-ff joins
[fcrepo-audit] awoods deleted fcrepo-1481 at 4790baf: https://git.io/vMzPk
* github-ff leaves
<whikloj>ajs6f: or (712) 775-7035 code : 479307#
* peichman leaves10:57
Stupid phone
* coblej leaves
* dwilcox leaves
<whikloj>ajs6f: no worries I am amazingly whip-like
* ajs6f is here
* ad96 joins
* peichman joins11:00
* coblej joins
<dhlamb>you recognize my sweet voice
* escowles joins11:01
* escowles is here
* ylchen joins
* dbernstein joins
* coblej is here
* dshalvi joins11:04
<ajs6f>Just scream loudly.11:05
* barmintor is here
* travis-ci joins11:11
fcrepo4-exts/fcrepo-audit#145 (master - 59138ca : Esmé Cowles): The build passed.
Change view : https://github.com/fcrepo4-exts/fcrepo-audit/compare/2e2a1b6e702f...59138caaa691
Build details : https://travis-ci.org/fcrepo4-exts/fcrepo-audit/builds/191348704
* travis-ci leaves
* dwilcox joins
<escowles>sorry, there's a construction site right outside my window...11:13
<whikloj>awoods: the crickets were very loud11:16
<ajs6f>Acceptance testing, kind of
<escowles>i think it's a good idea to have release testing stakeholders, though i think the ad-hoc release testing of the last couple releases has gone pretty well...
<ylchen>just create a section in that page. I will put my name Yinlin Chen (Virginia Tech) in that page.
* apb18 here late11:17
<escowles>that sounds good to me — good to recognize the good work people are doing, and help get more people involved to share the load11:19
<barmintor>(also particular thanks to escowles for confirming the AF suite runs)
<ajs6f>It will be a yeasty conv11:20
<barmintor>awoods: migration of FCR3 PIDs11:23
<ajs6f>It was such a mistake to expose that stuff in the API.
<dshalvi>Migration from F3 in general
<barmintor>awoods: the fact that PIDs become disconnected as an identifier
I am looking forward to learning what an apple tree is11:24
<escowles>migrating system creation/modification dates11:27
* apb18_ joins
* ajs6f_ joins11:28
<escowles>barmintor: i hope the apple tree has tasty fruit grafted onto hardy rootstock
* ajs6f leaves11:29
<barmintor>escowles: as much as I want a dry cider to be on the other end of this conversation, I have little confidence
* apb18_ dropped off, but view mandatory RI as a pain point.
It's more of a policy thing to me11:30
<escowles>barmintor: alas, probably not
* coblej need to leave call to prepare for another event
* apb18 leaves11:31
<ajs6f_>barmintor come down and visit and I will treat you to a glass of Potter's, our dry, tasty local11:32
<barmintor>ajs6f_: you have mentioned them before, I'd like to11:33
* coblej leaves11:34
<ajs6f_>apb18_: mandatory RI?11:35
<whikloj>ajs6f_: referential integrity (I think)
<escowles>yes, enforcing referential integrity of e.g., object URIs that look like fedora object references11:36
<apb18_>i.e. the repo enforces referential integrity right now, and there's nothing that can be immediately done about it
<ajs6f_>Ah ok talking about the same thing
I thought of the old Resource Index
<apb18_>ha, sorry!11:37
<barmintor>awoods: even the URI translation refactor will be limited in benefit while we rely on MODE's default indexes, which require dereferencing a node to get its path.
* coblej joins11:38
* dwilcox leaves
<ajs6f_>apb18_: it's okay we're just old11:39
<barmintor>awoods: sure?11:40
<ajs6f_>That's a great attitude!11:41
* dwilcox joins
<barmintor>also, if you ever write a JCR implementation, do not support SNS11:42
<ajs6f_>Well, I was going to, but now...
<barmintor>don't do it
<ajs6f_>Just say no.11:43
<escowles>+1 # no reason to merge that to master
<barmintor>Duplicating jcr:path - Not Even Once
<escowles>aspirational versioning?
<dshalvi>We can always get into the J2SE 1.x style of versioning fun11:47
Numbers are free, we won't run out.
<whikloj>It'll be the first one with semantic versioning!!
<whikloj>Remember good ol' Fedora 5.0.0, when we started using semantic versioning11:48
<barmintor>awoods: fix it with codenames11:49
<whikloj>I can wait for that aligning with the API spec, that seems like a good point
<bseeger>+1 for waiting for the alignment with API
<ylchen>first one? really? great!
<escowles>i'm also fine with waiting for the API spec-aligned for 5.0.0, just wanted to make sure we're not setting oureselves up for a repeat of the "fedora 4" brand issue11:50
<barmintor>escowles++ ++ ++
there's a question of how long it's been since a major release, and another about how far ahead you branch for the next.11:53
<ajs6f_>I'm branching now.
<barmintor>but one way to resolve it is to pick your (semi-)annual date, and set cut-offs for features.11:54
<escowles>"Fedora 5.2.1 2018"
<barmintor>maybe parenthesize 5.2.111:55
Fedora 2018 (5.2.1)
* coblej leaves
* coblej joins11:56
<escowles>"Fedora 2017 (5.2.1) Brisbane"
<apb18_>If anybody truly needed RDF/1.0 and <version of fedora that does not support it>, I bet API-X can be used to address those needs without placing burdens on the fedora impl
<barmintor>escowles: you might be joking but that's a pretty accurate description of community release drivers, I'm +111:57
<ajs6f_>This sounds like what we should have done to begin with.
<barmintor>"feature releases are identified by the community meeting they are announced at"
* coblej leaves11:58
<escowles>barmintor: yes, i am joking, but doing an annual release timed for (and named after) OR or some such could work
<barmintor>awoods: it turns out apple tree is kind of how I had proposed our local migration. Glad acoburn did it.11:59
* coblej joins
<escowles>gotta run to another meeting12:00
* ylchen leaves
* ajs6f_ leaves12:01
* ajs6f joins
* ajs6f leaves12:04
* ad96 leaves12:05
* ajs6f joins
* coblej leaves12:06
* coblej joins
* manez_ joins12:07
* manez leaves
* ajs6f leaves12:08
* coblej leaves
* coblej joins12:09
* coblej leaves12:11
* coblej joins12:19
* dshalvi leaves12:21
<barmintor>it is kind of interesting, in the mode of performance requirements in general, to talk about what we imagine the reasonable time to migrate a FCR3 object is.
* ajs6f joins
<barmintor>Or 1M of them.12:22
<awoods>barmintor: something like x-number of resources per y-time?
<barmintor>awoods: yeah
<awoods>barmintor: it would be good to get those requirements12:23
<barmintor>awoods: I am just reflecting on how if you averaged 1 sec per item, 1M => 12 days
maybe you parallelize it, but aggregate time expectations are going to run into network realities
food for thought
<bseeger>awoods, baraminter: just ingest time, or mods->rdf and then ingest time - the later will be somewhat subjective.
<barmintor>bseeger: also a consideration, true12:25
bseeger: I am just imagining migrating, say, 1M JPG files
* barmintor imagines12:26
* coblej leaves12:28
* coblej joins12:30
<ajs6f>barmintor: this is why a major line of thought behind the spec is distribution-- Amdahl's law12:36
* coblej leaves12:39
* coblej joins12:40
* coblej leaves12:49
* bseeger leaves12:53
* coblej joins12:59
* dwilcox leaves13:01
* dwilcox joins13:02
* dwilcox leaves13:03
* dwilcox joins
* dwilcox leaves
* ajs6f leaves
* bseeger joins13:13
* ajs6f joins13:29
* coblej leaves13:32
* coblej joins13:34
* coblej leaves13:37
* ajs6f_ joins13:41
* ajs6f leaves
* ajs6f_ leaves13:43
* coblej joins13:44
* ajs6f_ joins13:48
* ajs6f_ leaves13:51
* bseeger leaves14:02
* bseeger joins14:04
* coblej leaves14:15
* dwilcox joins14:18
* coblej joins14:22
* bseeger leaves15:40
* bseeger joins15:43
* apb18_ leaves16:06
* dwilcox leaves
* coblej leaves16:50
* coblej joins16:54
* mikeAtUVa leaves16:59
* peichman leaves17:01
* bseeger leaves17:09
* coblej leaves17:17
* kefo leaves17:28
* kefo joins
* kefo leaves17:34
* whikloj leaves17:58
* dbernstein leaves19:54
* dbernstein joins20:03
* dbernstein leaves20:26
* dbernstein joins20:39
* dbernstein leaves20:50
* dbernstein joins20:52
* peichman joins22:20
* peichman leaves22:26
* dbernstein leaves23:42
* peichman joins00:13
* peichman leaves00:15

Generated by Sualtam