#duraspace IRC Log


IRC Log for 2013-03-13

Timestamps are in GMT/BST.

[2:49] * eddies (~eddies@unaffiliated/eddies) Quit (Read error: Connection reset by peer)
[3:35] * eddies (~eddies@cm107.eta98.maxonline.com.sg) has joined #duraspace
[3:35] * eddies (~eddies@cm107.eta98.maxonline.com.sg) Quit (Changing host)
[3:35] * eddies (~eddies@unaffiliated/eddies) has joined #duraspace
[6:47] * eddies (~eddies@unaffiliated/eddies) Quit (Quit: Leaving.)
[6:47] * eddies (~eddies@cm107.eta98.maxonline.com.sg) has joined #duraspace
[6:47] * eddies (~eddies@cm107.eta98.maxonline.com.sg) Quit (Changing host)
[6:47] * eddies (~eddies@unaffiliated/eddies) has joined #duraspace
[6:50] -zelazny.freenode.net- *** Looking up your hostname...
[6:50] -zelazny.freenode.net- *** Checking Ident
[6:50] -zelazny.freenode.net- *** Found your hostname
[6:50] -zelazny.freenode.net- *** No Ident response
[6:50] * DuraLogBot (~PircBot@atlas.duraspace.org) has joined #duraspace
[6:50] * Topic is '[Welcome to DuraSpace - This channel is logged - http://irclogs.duraspace.org/]'
[6:50] * Set by cwilper!ad579d86@gateway/web/freenode/ip. on Fri Oct 22 01:19:41 UTC 2010
[9:28] * eddies1 (~eddies@cm107.eta98.maxonline.com.sg) has joined #duraspace
[9:28] * eddies1 (~eddies@cm107.eta98.maxonline.com.sg) Quit (Changing host)
[9:28] * eddies1 (~eddies@unaffiliated/eddies) has joined #duraspace
[9:31] * kshepherd (kim@ec2-184-73-177-234.compute-1.amazonaws.com) has joined #duraspace
[9:36] * eddies (~eddies@unaffiliated/eddies) Quit (*.net *.split)
[9:36] * kshepher1 (kim@ec2-184-73-177-234.compute-1.amazonaws.com) Quit (*.net *.split)
[10:19] * eddies1 (~eddies@unaffiliated/eddies) Quit (Quit: Leaving.)
[10:58] * eddies (~eddies@108.202-128-171.unknown.qala.com.sg) has joined #duraspace
[10:58] * eddies (~eddies@108.202-128-171.unknown.qala.com.sg) Quit (Changing host)
[10:58] * eddies (~eddies@unaffiliated/eddies) has joined #duraspace
[11:33] * eddies1 (~eddies@bb116-15-142-48.singnet.com.sg) has joined #duraspace
[11:34] * eddies1 (~eddies@bb116-15-142-48.singnet.com.sg) Quit (Changing host)
[11:34] * eddies1 (~eddies@unaffiliated/eddies) has joined #duraspace
[11:35] * eddies (~eddies@unaffiliated/eddies) Quit (Ping timeout: 252 seconds)
[12:16] * eddies1 (~eddies@unaffiliated/eddies) Quit (Quit: Leaving.)
[12:31] * eddies (~eddies@cm107.eta98.maxonline.com.sg) has joined #duraspace
[12:31] * eddies (~eddies@cm107.eta98.maxonline.com.sg) Quit (Changing host)
[12:31] * eddies (~eddies@unaffiliated/eddies) has joined #duraspace
[12:56] * mhwood (mwood@mhw.ulib.iupui.edu) has joined #duraspace
[13:32] * mhwood (mwood@mhw.ulib.iupui.edu) Quit (Quit: Leaving.)
[13:54] * mhwood (mwood@mhw.ulib.iupui.edu) has joined #duraspace
[15:07] * fasseg (~fas@HSI-KBW-078-043-007-220.hsi4.kabel-badenwuerttemberg.de) has joined #duraspace
[19:10] * hpottinger (~hpottinge@mu-162198.dhcp.missouri.edu) has joined #duraspace
[19:14] <hpottinger> Oh, right, time change, meeting doesn't start until 3pm CDT… I'll just hang out
[19:14] * bram-atmire (~bram@94-225-36-191.access.telenet.be) has joined #duraspace
[19:24] <bram-atmire> Hi
[19:25] * fasseg (~fas@HSI-KBW-078-043-007-220.hsi4.kabel-badenwuerttemberg.de) Quit (Quit: Leaving)
[19:29] * bram-atmire (~bram@94-225-36-191.access.telenet.be) Quit (Quit: bram-atmire)
[20:06] <hpottinger> hey, mhwood, are we the meeting today?
[20:07] <mhwood> So far, it seems so.
[20:07] * hpottinger hands the gavel to mhwood
[20:07] <hpottinger> need one more to reach quorum, I think
[20:12] <hpottinger> well, we were talking about something last week, IIRC, something about data objects and making them usable?
[20:24] <mhwood> I see some talk about Cocoon 2.2, Cocoon 3, and "after Cocoon".
[20:25] <mhwood> PeterDietz was talking about bolting new UIs on through REST.
[20:26] <hpottinger> ah, yes, after Cocoon, I remember that talk
[20:27] <mhwood> I'm still trying to get to C3. I managed to use NetBeans to build and run the sample "you're so smart you made Cocoon 3 work" webapp.
[20:27] <hpottinger> I was poking around the other day looking for other XML-based engines
[20:27] <hpottinger> kept running into Xul-based things
[20:28] <hpottinger> blundered into this list: http://java-source.net/open-source/xml-user-interface-toolkits
[20:29] <mhwood> Ugh, a common infrastructure that's always built with the application because Mozilla never got Firefox and Thunderbird onto the same XULrunner release....
[20:29] <hpottinger> found this: http://luxor-xul.sourceforge.net/
[20:30] <mhwood> Wow, that is seriously too many choices...and nothing familiar.
[20:30] <hpottinger> I agree, it's in my list of "explore this, later"
[20:31] <mhwood> Luxor looks like another XML aspect, unrelated to what Cocoon does for us (the pipeline).
[20:31] <hpottinger> Oh, I see, there was a "Open XUL Alliance challenge" that might have encouraged "diversity"
[20:32] <hpottinger> I am pretty sure there isn't really anything like Cocoon out there
[20:33] <hpottinger> this searching I did (by accident) all came up with lots of UI-centric development
[20:33] <hpottinger> which is interesting reading and might be fun to play with, but doesn't really count as web app development
[20:34] <mhwood> Yes. I'd really rather wind up with *fewer* (like, 1) UIs in the DSpace box and let people create alternatives out-of-tree, once we give them a fairly stable API to code to.
[20:35] <hpottinger> Agreed. And it doesn't even really have to be XSLT/XML at all, just a simple template language would be sufficient
[20:36] <mhwood> (There are some things that everybody seems to have a need to do: UIs, fonts.... We wind up with huge collections of different ways to do the same thing, that are hard to tell apart.)
[20:36] <hpottinger> Anything to make the job of random library coder/devops worker more familiar
[20:37] <mhwood> Actually you could rip the UI off of the pipeline, too. C3 is concentrating on the pipeline and paring away stuff that has been done by others, arguably better, since C2.1 or so.
[20:37] <mhwood> I would kinda miss aspect-oriented page building.
[20:38] <hpottinger> what are people using for UIs on C3?
[20:38] <mhwood> Dunno, I only got one demo app. to run so far.
[20:39] <hpottinger> Lenya runs on Cocoon, wondering if they have C3 in the works?
[20:39] <mhwood> You *can* still translate XML to XHTML using Cocoon, but the emphasis has shifted away from Cocoon running the show.
[20:42] <hpottinger> no mention of C3 on the Lenya roadmap (http://lenya.apache.org/index/roadmap.html) and they are both Apache wards.
[20:43] <mhwood> Not surprising. Cocoon looks stagnant to anyone not on the dev. team.
[20:45] <hpottinger> ugh, the dates on Lenya's road map are a decade past...
[20:46] <hpottinger> You know, we are kinda shining beacon on a hill when it comes to keeping the community informed...
[20:47] * PeterDietz (~peterdiet@ has joined #duraspace
[20:47] <PeterDietz> hi all
[20:48] <hpottinger> hey there, peterdietz, mhwood and I are just boggling at out of date road maps (other projects, not ours, mind you)
[20:48] <PeterDietz> so.. Not much agenda. Probably don't want to discuss too much without the horde
[20:48] <hpottinger> 3 = horde
[20:48] <hpottinger> :-)
[20:48] <PeterDietz> Ohh, (reading scrollback), cocoon. Yeah, I don't want anything to do with that one.
[20:49] <PeterDietz> Hardy, did you say you mentioned the rest api demo site to someone.
[20:49] <PeterDietz> Also, I know helix will kill me, for not updating the elastic search statistics page, but at OSU, we've been doing a crazy amount of improvements to elastic search statistics page
[20:50] <hpottinger> yes, peterdietz, I did tell another developer here about the hedtek Demo on Demo
[20:50] <PeterDietz> One of the interesting things, the default data range is "last five years", then we report on a granularity of monthly downloads within that collection...
[20:50] <mhwood> [Robin Williams] Demo demo!
[20:50] <hpottinger> he's a developer for the campus repository we just merged with our DSpace instance
[20:50] <PeterDietz> (oops, the interesting is in this line). So, when you request a time period that is a month or less, you get output in days
[20:51] <hpottinger> crazy? improvements?
[20:52] <PeterDietz> And, my "Al Gore Rhythm" for choosing that allows you to bend your date range around a month.. i.e. what was our usage over xmas break: Dec 20 - Jan 10,, get days
[20:53] <PeterDietz> also, we've upgraded to using elastic search 0.19.11, (default dspace is 0.18.6), not sure if there's anything excited about the upgrade, but
[20:53] <hpottinger> anyway, back to REST-API demo, faculty are interested in being able to display a list of their resposited materials on their own pages.
[20:53] <PeterDietz> but, we changed our back-end architecture, and put elasticsearch on its own dedicated virtual server (since we have several webapps that run on ES)
[20:54] <PeterDietz> kompewter: define reposited
[20:54] <mhwood> Deposited in a repository?
[20:54] <mhwood> But I'm not positive.
[20:54] <PeterDietz> yeah, so being able to display information in DSpace, outside of DSpace requires you to do something to share that data.
[20:55] <PeterDietz> A few approaches.. 1) Hacky XMLUI DRI way, Have researchers home-page have an ajax script that grabs raw data from xmlui's DRI thing, and embed that... probably will not work
[20:56] <PeterDietz> 2) Have a REST API that allows for searches, and read. So you could say /api/search?q=author:"Albert Einstein"&results=Items or something (I just made up that syntax for search)
[20:57] <hpottinger> ah, yes, peterdietz, you weren't at OR11, where I think the verb "deposit" was coined
[20:57] <hpottinger> gah, silly fingers, "reposit: verb, to deposit in a repository"
[20:58] <PeterDietz> you can have little servlets to do novel things within your current site. i.e. we are using mhwood's content-statistics servlet for some of the basics... (semi-expensive operations) http://kb.osu.edu/dspace/content-statistics
[20:59] <PeterDietz> <dspace-repository-statistics date="20130313T205713Z">
[20:59] <PeterDietz> <statistic name="communities">170</statistic>
[20:59] <PeterDietz> <statistic name="collections">1425</statistic>
[20:59] <PeterDietz> <statistic name="items">51055</statistic>
[20:59] <PeterDietz> <statistic name="bitstreams">106717</statistic>
[20:59] <PeterDietz> <statistic name="totalMBytes">95917</statistic>
[20:59] <PeterDietz> <statistic name="images">78111</statistic>
[20:59] <PeterDietz> <statistic name="imageMBytes">7937</statistic>
[20:59] <PeterDietz> </dspace-repository-statistics>
[20:59] <hpottinger> planning to use that servlet soon
[21:00] <PeterDietz> ... but you could do things like this in a more organized fashion, and connected with structured ways to request this.. and for all objects.
[21:00] <PeterDietz> err, could do this better with an API..
[21:00] <hpottinger> open search can probably suffice for this stuff, I think
[21:01] <mhwood> Well, it's a degenerate API. One niladic operator.
[21:01] <PeterDietz> I'm not an expert for sure, and don't know exactly what being a RESTful API means vs what is there now. But, its certainly the base layer I would prefer to build upon.
[21:02] <hpottinger> but, I agree, if you're going to invest dev time, building against REST seems like a safer investment
[21:02] <PeterDietz> yep.. for instance the little toy-app I build, gives you a way to easily build against your repositories data: http://dspace-rest-client-play.herokuapp.com/item/45
[21:02] <mhwood> Indeed, I am hopeful that my little servlet can be slurped up into the REST code and disappear in favor of /rest/site/statistics or some such.
[21:05] <PeterDietz> And then, something more programmatic (in an institutional sense, not in a thing built by programmers sense), is that you could build tools to help talk with your API.. i.e. Ruby Gem that lets any developer in Ruby config the url to their API (maybe also auth token), and then you can say foo = @dspace.communities.all
[21:07] <PeterDietz> ...and then being cool Acmeists ( http://acmeism.org/ ) that we all strive to be, we'll also build starter tools for users to rock&roll with php, Java, cobol, etc
[21:07] <hpottinger> mhwood, you could hedge your bets and make pull requests against hedtek and Wijiti github forks.
[21:07] <PeterDietz> we do need to "bring back" the rest api
[21:07] <hpottinger> cobol, narf!
[21:08] <mhwood> Ooh, I haven't written any COBOL in two, three years.
[21:09] <PeterDietz> I haven't played with Wijiti, but I imagine that its nothing scary.
[21:09] <PeterDietz> I have played with Hedtek's API implementation, and it is nothing crazy
[21:10] <hpottinger> R/W, with no authZ, kinda scary
[21:10] <PeterDietz> Last time I played with dspace/rest-api thing, it was scary, created infinite loops, and took down tomcat. So as-is, we've got nothing to lose
[21:14] <PeterDietz> If we had more horde present, I'd say I'd like to get input on possibly replacing current DSpace/dspace-rest with Hedtek's DSpace REST
[21:14] <PeterDietz> But, in actuality.. We really ought to support full CRUD..
[21:15] <PeterDietz> So, that was a premature statement. I'd like to inspect wijiti's before discussing bringing any one of them in.
[21:15] <hpottinger> I'm OK with CRUD as long as we can also do authZ
[21:15] <PeterDietz> Figuring out authZ should be doable..
[21:15] <PeterDietz> I'm also needing to figure out if these two have compatible endpoints.
[21:16] <mhwood> Gotta let you two figure it out...'bye 'bye.
[21:16] * mhwood (mwood@mhw.ulib.iupui.edu) has left #duraspace
[21:16] <PeterDietz> So you could take your DSpace-REST-client, and point it at a demo of either one, and for read operations, get similar/same data ...
[21:16] <hpottinger> if that's not currently possible, it should be a small tweak to make them line up
[21:17] <hpottinger> (but I'm a MVC guy, so who knows…)
[21:17] <PeterDietz> And at the same time, I'd like to re-write (start de novo) the DSpace-rest-client that I've built, and make it in rails..
[21:17] <PeterDietz> I'm also tempted to throw away the current rest implementations, and use the webMVC to manage building the rest api.. That was a very comfortable framework to use to build that webmvc webapp
[21:18] <hpottinger> solo many paths....
[21:18] <hpottinger> s/solo/so/
[21:18] <PeterDietz> so.. I need a handler, who can help organize this thought into something that gives us good direction long-ish term, and makes good use of our time
[21:19] <hpottinger> doh, wrong syntax, try that again, s/so/solo/
[21:19] <PeterDietz> If I do all that I just promised above.. I'll forget to eat for atleast 2 weeks.. ;)
[21:19] <PeterDietz> kompewter is dead..
[21:19] <PeterDietz> anyways.. I've got some getting-back-to-work to do...
[21:25] * hpottinger waves to all the crickets an clicks the close button.
[21:25] * hpottinger (~hpottinge@mu-162198.dhcp.missouri.edu) has left #duraspace
[21:55] * khkjhk (52292725@gateway/web/freenode/ip. has joined #duraspace
[22:01] * khkjhk (52292725@gateway/web/freenode/ip. Quit (Quit: Page closed)

These logs were automatically created by DuraLogBot on irc.freenode.net using the Java IRC LogBot.