Conversation with #duraspace at Wed Oct 8 08:13:39 2014 on mhwood@chat.freenode.net (irc) (08:13:39) #duraspace: The topic for #duraspace is: [Welcome to DuraSpace - This channel is logged - http://irclogs.duraspace.org/] (08:13:39) #duraspace: Topic for #duraspace set by cwilper!ad579d86@gateway/web/freenode/ip.173.87.157.134 at 21:19:41 on 10/21/10 (09:59:24) kohts [~kohts@212.233.127.186] entered the room. (10:15:36) hpottinger [~hpottinge@mu-161174.dhcp.missouri.edu] entered the room. (10:25:36) peterdietz [~peterdiet@server112.longsightgroup.com] entered the room. (10:54:57) robint [81d7ec36@gateway/web/freenode/ip.129.215.236.54] entered the room. (11:01:23) robint: Hi everyone (11:01:54) mhwood: Hi. hpottinger will be back shortly. (11:02:43) robint: Am I right in thinking Tim is away this week? (11:02:57) mhwood: That is what I recall. Who was going to moderate today? (11:03:47) robint: We sort of left it vague (11:04:03) mhwood: So I see. (11:05:57) mhwood: So: status of 5.0, anyone? (11:06:00) kstamatis [25067cc3@gateway/web/freenode/ip.37.6.124.195] entered the room. (11:06:17) peterdietz: Hi All, I can be the moderator. Will need some audience assistance though (11:06:43) hpottinger: I'm back (11:07:09) hpottinger: agenda for today: https://wiki.duraspace.org/display/DSPACE/DevMtg+2014-10-08 (11:07:11) kompewter: [ DevMtg 2014-10-08 - DSpace - DuraSpace Wiki ] - https://wiki.duraspace.org/display/DSPACE/DevMtg+2014-10-08 (11:07:24) robint: I'm just digging out Tim's regular Jira list from last weeks notes... (11:07:41) robint: https://jira.duraspace.org/browse/DS-1582?filter=-4&jql=project%20%3D%20DS%20AND%20issuetype%20in%20%28%22Code%20Task%22%2C%20Improvement%2C%20%22New%20Feature%22%2C%20Task%2C%20Sub-task%29%20AND%20fixVersion%20%3D%205.0%20AND%20status%20%3D%20%22Code%20Review%20Needed%22%20ORDER%20BY%20createdDate%20ASC (11:07:42) kompewter: [ Issue Navigator - DuraSpace JIRA ] - https://jira.duraspace.org/browse/DS-1582?filter=-4&jql=project%20%3D%20DS%20AND%20issuetype%20in%20%28%22Code%20Task%22%2C%20Improvement%2C%20%22New%20Feature%22%2C%20Task%2C%20Sub-task%29%20AND%20fixVersion%20%3D%205.0%20AND%20status%20%3D%20%22Code%20Review%20Needed%22%20ORDER%20BY%20createdDate%20ASC (11:07:44) kompewter: [ https://jira.duraspace.org/browse/DS-1582 ] - [DS-1582] All DSpaceObjects should have metadata support - DuraSpace JIRA (11:09:10) hpottinger: DSPR#654 is already in master (11:09:11) kompewter: [ https://github.com/DSpace/DSpace/pull/654 ] - Support Metadata On All DSpaceObjects by KevinVdV (11:09:24) hpottinger: and has spawned a few issues (11:09:55) hpottinger: DS-2179, DS-2164 (11:10:00) kompewter: [ https://jira.duraspace.org/browse/DS-2179 ] - [DS-2179] Shibboleth (and probably LDAP) logins fail after merge of Metadata For All - DuraSpace JIRA (11:10:01) kompewter: [ https://jira.duraspace.org/browse/DS-2164 ] - [DS-2164] Metadata4All Bug with EPerson - DuraSpace JIRA (11:10:42) peterdietz: Kevin's got a fix, about 2 hours ago for the 2164 (11:10:45) peterdietz: I haven't tested it yet (11:11:08) hpottinger: both issues have PRs, DSPR#687 is high on my TODO list for today, if it fixes 2179 for me, I will likely merge it, unless there are objections (11:11:10) kompewter: [ https://github.com/DSpace/DSpace/pull/687 ] - [DS-2179] Shibboleth (and probably LDAP) logins fail after merge of Metadata For All by KevinVdV (11:11:16) mhwood: I've inspected both, and they seem correct, but have not tested yet either. (11:12:21) robint: If these are small scale fixes for bugs then I think they should just be merged (11:12:53) hpottinger: I agree, standard procedure, just thought I'd let you all know my intentions :-) (11:12:56) peterdietz: According to Github, there are 36 open features. I think this is the big think to keep track of (11:13:13) peterdietz: think -> thing (11:13:49) peterdietz: I've noticed some trends / duplicates.. That could collapse this number.. i.e. 2 CAS implementations, 2 BatchImportUI implementations ... (11:14:28) mhwood: I think the CAS implementations are for different versions. (11:14:36) hpottinger: https://github.com/DSpace/DSpace/pulls?q=is%3Aopen+sort%3Acreated-asc+label%3Afeature (11:14:37) kompewter: [ Pull Requests · DSpace/DSpace · GitHub ] - https://github.com/DSpace/DSpace/pulls?q=is%3Aopen+sort%3Acreated-asc+label%3Afeature (11:14:49) kstamatis: same for BatchImportUI, XMLUI and JSPUI (11:15:00) peterdietz: Sorry to derail.. I'm wondering if we should look at things like ORCID, or LOD, or author pages.. They seem big and feature-y (11:15:04) hpottinger: One of the CAS PRs was closed yesterday, I think (11:15:48) hpottinger: next on my list today, after I can login with Shibboleth again, is to get back to testing the ORCID PR (11:15:49) robint: peterdietz: good plan, the sooner they get reviewed and committed the better (11:16:36) hpottinger: LOD: who's testing? (11:17:09) peterdietz: I've actually been working on @Mire's LOD project: https://github.com/dspace-oceanlink/DSpace/tree/oceanlink-4_x/dspace-lod (11:17:09) kompewter: [ DSpace/dspace-lod at oceanlink-4_x · dspace-oceanlink/DSpace · GitHub ] - https://github.com/dspace-oceanlink/DSpace/tree/oceanlink-4_x/dspace-lod (11:17:38) peterdietz: define[working on] = Working on integrating,, not actually developing the feature itself (11:18:03) peterdietz: That is not the same as Pascal's LOD project though (11:18:04) hpottinger: is there a PR for @mire's LOD project? (11:18:38) robint: Do they conflict? If so is there agreement as to which one gets merged? (11:19:03) peterdietz: No PR at this time. I think it's generic, but it was tailored for a marine biology data science group, atleast, files have the name OceanLink all over the place (11:19:20) peterdietz: I have no idea of an interesting SPARQL query, so I'm finding it impossible to test (11:19:33) peterdietz: umm, webapp loaded [check] (11:19:46) peterdietz: data populated [check] (11:19:59) kdweeks1 [~Adium@2001:468:c80:a103:6860:79c4:61e2:5ff0] entered the room. (11:20:23) hpottinger: (robint: I don't think @mire as made a PR for their work, so I'm not sure if it's up to us to consider it, though we *can* try to not get too much in their way with accepting a competing PR (11:20:38) mhwood: hpottinger++ (11:21:19) robint: Ideally Pascal and @Mire should collaborate (11:21:21) peterdietz: It will be Mark Diggory's job to jump in at some point, to tell us why to hold off on Pascal's version.. speak now or forever hold your peace? Or. it could be like what happened to XOAI.. i.e. a future release overhausl (11:21:41) peterdietz: Pascal is aware of @mire's work. He's already made a PR (to dspace-oceanlink) (11:22:06) peterdietz: I'll let them figure it out.. I think Pascal is on holiday (11:22:11) hpottinger: mdiggory isn't here, so peterdietz is talking to an empty chair :-) (11:22:28) peterdietz: 8am US West (11:22:36) peterdietz: wait, 9 (11:22:50) robint: He'll be surfing before work (11:23:06) peterdietz: According to the Weezer song, you drive your car to work, I'll take my board (11:23:14) mhwood: So the question is: accept Pascal's work for 5.0 (still some time to decide) and hope that they work together to bring in anything novel that @mire has, or make LOD wait for 6.0 to give more time to merge them or pick one. (11:23:32) kdweeks left the room (quit: Ping timeout: 260 seconds). (11:23:45) robint: Would it be worth emailing both parties to ask that question? (11:23:52) hpottinger: I'm leaning towards giving the advantage to code we can see (11:24:55) mhwood: We can start testing Pascal's code now, while they coordinate and perhaps produce more commits. (11:24:57) robint: hpottinger: I agree, but it would be nice to give the @Mire gang the chance to say " hold on, because..." (11:25:35) hpottinger: I agree, I think friendliness is in order here (11:25:44) peterdietz: yeah. Also, Looking at dspace-rdf (Pascal's version). It looks small-ish, it could get tucked into dspace-rest maybe (11:26:35) robint: peterdietz: as sort of another endpoint? (11:26:44) hpottinger: how about this: we say: hey, @mire, we want LOD in 5.0, we have a current PR, do you want to exercise your veto? (11:26:48) peterdietz: yeah.. look how small dspace-rdf webapp is: https://github.com/pnbecker/DSpace/tree/dspace-rdf/dspace-rdf (11:26:48) kompewter: [ DSpace/dspace-rdf at dspace-rdf · pnbecker/DSpace · GitHub ] - https://github.com/pnbecker/DSpace/tree/dspace-rdf/dspace-rdf (11:28:16) peterdietz: Deadline was just on the order of "hours" ago, we could grant a late arrival, if they wanted to propose their version. (I've also noticed some additional PR's have been arriving) (11:28:46) peterdietz: But yeah. I'd say get @mire and Pascal to make an official discussion (11:28:50) hpottinger: post-deadline PRs? (11:29:22) peterdietz: look for anything without a label (11:29:27) mhwood: Just because there's a PR doesn't mean that more commits cannot be accepted, right? With the RT's approval. (11:29:30) robint: Is there another PR for the rdf stuff that would go into dspace-api? (11:30:06) hpottinger: speaking of waiting on PRs, we have about 9 feature PRs that bear the label of "work in progress" https://github.com/DSpace/DSpace/pulls?q=is%3Aopen+sort%3Acreated-asc+label%3Afeature+label%3A%22work+in+progress%22 (11:30:06) kompewter: [ Pull Requests · DSpace/DSpace · GitHub ] - https://github.com/DSpace/DSpace/pulls?q=is%3Aopen+sort%3Acreated-asc+label%3Afeature+label%3A%22work+in+progress%22 (11:30:27) peterdietz: robint: same one. I just shared a subdir https://github.com/pnbecker/DSpace/tree/dspace-rdf/dspace-api/src/main/java/org/dspace/rdf (11:30:28) kompewter: [ DSpace/dspace-api/src/main/java/org/dspace/rdf at dspace-rdf · pnbecker/DSpace · GitHub ] - https://github.com/pnbecker/DSpace/tree/dspace-rdf/dspace-api/src/main/java/org/dspace/rdf (11:31:11) robint: peterdietz: doh! :) (11:31:31) hpottinger: we had 11 WIP PRs yesterday, so I think 2 must have been closeed (11:32:20) mhwood: Haven't read all, but I wonder if #668 should have that label removed given recent comment by submitter. (11:33:38) hpottinger: DSPR #668 aha, thanks, mhwood (11:34:06) robint: And some of the others look a bit dormant to be honest (11:34:16) mhwood: Yes. (11:34:45) hpottinger: WIP label removed for #668 (11:35:00) hpottinger: shall we merge it? (11:35:05) robint: How about I volunteer to chase up the owners of the WIP PRs and update their status? (11:35:45) hpottinger: robint: I did add a comment to all of them yesterday, but, the personal touch would probably help, thanks! (11:36:39) robint: hpottinger: you are way ahead of me (11:37:42) robint: getting back to the LOD stuff, what is our plan of action? (11:40:04) hpottinger: our RC seems to be testing the @mire work... (11:40:34) peterdietz: I don't have any preferred LOD solution ( I really am not an LOD expert, and am not familiar with Pascal's version). The one that IS a PR is probably our most likely candidate though (11:40:49) hpottinger: (BTW, all new PRs post-deadline were bug fixes, I've labeled them all as such) (11:41:08) peterdietz: Plan of Action should be to email mdiggory and pnbecker, to get a statement from each (11:41:27) hpottinger: peterdietz: mind if we put that on your TODO list? (11:41:34) peterdietz: i can send it (11:41:38) robint: Do we just need a volunteer to contact Pascal and @Mire? (11:41:50) ***hpottinger points at peterdietz. (11:42:06) robint: Sorry, lagging behind here (11:43:49) hpottinger: so, I've been using the "feature" label... but, there's this "improvement" label... (11:44:19) hpottinger: https://github.com/DSpace/DSpace/pulls?q=is%3Aopen+sort%3Acreated-asc+label%3Aimprovement (11:44:19) kompewter: [ Pull Requests · DSpace/DSpace · GitHub ] - https://github.com/DSpace/DSpace/pulls?q=is%3Aopen+sort%3Acreated-asc+label%3Aimprovement (11:44:57) robint: I'm afraid I've got to duck out early this week, cheers all (11:45:13) robint left the room (quit: Quit: Page closed). (11:46:24) hpottinger: before I forget, who wants to join the RT? :-) (11:47:48) peterdietz: Do we want to talk about DSPR673 ? Rehab DCValue (11:48:07) mhwood: OK (11:49:43) hpottinger: DSPR 673 is a +1 for me, anyone with local modifications that rely on DCValue *have been warned* sufficiently (11:50:03) mhwood: I understand Bram's concerns, and it sounds like this should go in late if it goes in at all, to concentrate any breakage into one PR rather than many. (11:50:46) mhwood: Documentation will be much smaller problem than it seems. Most of those hits are on code snippets. (11:51:53) peterdietz: The change is essentially just this: https://github.com/mwoodiupui/DSpace/commit/1a8eb2e73cf87f2dba3c207672f7c83de412a5f7#diff-ae2c72907adcf21a7ce831528fb20606 (11:51:56) kompewter: [ [DS-1883] Un-deprecate DCValue and rename it. · 1a8eb2e · mwoodiupui/DSpace · GitHub ] - https://github.com/mwoodiupui/DSpace/commit/1a8eb2e73cf87f2dba3c207672f7c83de412a5f7#diff-ae2c72907adcf21a7ce831528fb20606 (11:52:18) hpottinger: mhwood: are you OK with taking on the weight of rebasing #673 as required? (11:52:27) mhwood: Yes, I can do that. (11:52:55) hpottinger: OK, let's just merge last (11:52:58) mhwood: 119 of those 120 files only change "DCValue" to "Metadatum", I think. (11:53:23) peterdietz: DCValue renamed to Metadatum.. Remove the @Deprecate.. (11:53:40) peterdietz: I forget.. What was the grand scheme.. The ideal metadata-value container? (11:53:41) mhwood: Yes, it's very simple but wide-ranging. (11:53:50) hpottinger: and get ready to see the other deprection warnings you may have missed (11:54:35) mhwood: Tim convinced me that Metadatum has a different function than MetadataValue. We may want to replace individual fields inside MetadataValue with a Metadatum, though. (11:54:41) peterdietz: MetadataValue: https://github.com/DSpace/DSpace/blob/master/dspace-api/src/main/java/org/dspace/content/MetadataValue.java (11:54:42) kompewter: [ DSpace/MetadataValue.java at master · DSpace/DSpace · GitHub ] - https://github.com/DSpace/DSpace/blob/master/dspace-api/src/main/java/org/dspace/content/MetadataValue.java (11:55:56) mhwood: I should look at the code to be sure, but one seems to represent an abstract metadata value while the other represents a metadata value that is linked to a particular object and backed by the database. (11:57:03) ***hpottinger has a grudge against that big wall of deprecation warnings (11:58:03) mhwood: Yes: "Database access class representing a Dublin Core metadata value." (No longer strictly Dublin Core.) That's different from a simple structure without all the database implications, and IIRC we need both. (11:58:59) peterdietz: Hmm. Lacking the PhD, I don't see the edge to gain by joining / or keeping seperate Metadatum and MetadataValue... I'll just pay attention to the part where it's no longer implying DublinCore (11:59:33) mhwood: That bit is probably just a C/P error. (11:59:53) peterdietz: But, I'm +1 to no longer implying DublinCore, and no longer having a giant deprecation block. (12:00:23) peterdietz: If someone know further best-practice for this (i.e. extra mile).. Then we're all ears, but this looks sufficiently useful. (12:00:43) peterdietz: Gotta merge it last. Similar to the maven consolidation in the past dspace-xmlui/dspace-xmlui-api (12:01:09) mhwood: Keeping it updated should be simple if tedious. (12:01:49) hpottinger: that would be a fun thing to script: keep_this_branch_rebased (12:02:21) ***mhwood imagines writing an AI in shell. (12:02:56) hpottinger: you could just cron it, rebase every morning and push the result if it's clean, or e-mail you if there are conflicts (12:03:16) hpottinger: like CI, only really dumb (12:07:31) mhwood: OK, I will keep this one updated. Nudge me if needed. (12:09:35) hpottinger: I think we've wrapped up? (12:09:51) hpottinger: except for the part where everyone volunteers to be on the RT, that is (12:10:26) mhwood: Discussion has quieted down. I think someone could all "meeting adjourned". (12:10:35) mhwood: s/all/call/ (12:10:36) kompewter: mhwood meant to say: Discussion has quieted down. I think someone could call "meeting adjourned". (12:10:49) peterdietz: {gavle bang} (12:11:18) peterdietz: This court session has adjourned, we shall re-convene in 7 days. In the mean time, keep fixing up code (12:11:29) hpottinger: meeting adjourned, where did I put my tea? (12:11:55) hpottinger: and no more feature PRs for 5.0, OK, unless you ask *really* nicely (12:12:20) kohts left the room (quit: ). (12:12:55) peterdietz: oops Gavle is the home town of a Swedish coffee company Gävle -> Gevalia.. Gavel is the word for a judges wooden mallet (12:13:25) hpottinger: no more Swiss typos, peterdietz (12:13:56) peterdietz: No more cultural criss-crossing the Swiss are famous for the Alps. The Swedes are famous for IKEA, and Volvos (12:13:57) hpottinger: sweedish, that is, darn me (12:14:22) peterdietz: ok, jumping to unlogged #dspace (12:14:55) peterdietz left the room. (12:23:38) infralumpen_ [c8037802@gateway/web/freenode/ip.200.3.120.2] entered the room. (12:23:57) infralumpen_: Hi (12:44:21) kstamatis left the room (quit: Quit: Page closed). (12:46:53) infralumpen_ left the room (quit: Quit: Page closed). (15:30:58) peterdietz [~peterdiet@162-231-22-3.lightspeed.clmboh.sbcglobal.net] entered the room. (15:55:16) peterdietz left the room (quit: Ping timeout: 244 seconds).