Time
S
Nick
Message
00:30 axfelix joined #dataverse
00:40 axfelix joined #dataverse
01:36 axfelix joined #dataverse
02:35 garnett joined #dataverse
06:19 metamattj joined #dataverse
07:29 bencomp joined #dataverse
11:47 blind-dataverse- joined #dataverse
11:48 blind-dataverse- left #dataverse
11:48 romainM joined #dataverse
11:49 romainM
hello world ?
12:37 pdurbin
romainM: hi!
12:38 romainM
yes, a living person ! :D
12:39 pdurbin
heh. only two bots here. iqlogbot and searchbot :)
12:39 pdurbin
oh, maybe sbmarks is a bot ;)
12:40 romainM
are you sure they're not all bots ?..
12:40 * pdurbin
slaps bencomp with a trout
12:41 * bencomp
throws a herring with raw onions at pdurbin
12:41 pdurbin
bencomp: just don't give me any of that double salt licorice
12:42 bencomp
:D
12:43 pdurbin
romainM: after reading https://groups.google.com/d/msg/dataverse-community/HPJZ4kWgC2U/JO4plaSUBAAJ I'm very curious what metadata fields you want
12:43 romainM
oh
12:43 romainM
at first, I was looking for geo fields
12:43 romainM
then, you showed me something I was too blind to see :D
12:44 romainM
but I will maybe need some fields kinda ... specifics
12:44 romainM
it's more like some "personal notes" fields, I have to see with my people if I can't just use an "unused" field
12:44 romainM
to store into it
12:45 pdurbin
romainM: did you look at our geospatial metadata block? https://docs.google.com/spreadsheet/ccc?key=0AjeLxEN77UZodHFEWGpoa19ia3pldEFyVFR0aFVGa0E#gid=4 is linked from http://guides.dataverse.org/en/4.1/user/appendix.html
12:45 romainM
I did, yes
12:45 romainM
but there is also some fields like "archive arrangement level description"
12:46 romainM
that ... didn't look to anything
12:46 romainM
(fields that correspond to my people's metadata)
12:46 pdurbin
romainM: so... you're looking for a field for the curator/archivist to use for notes about the dataset?
12:47 romainM
actually, I got like 10 fields on 45 that I don't know what to do about :D
12:47 romainM
fields that don't correspond, in the name, at fields I can fill in dataverse
12:48 pdurbin
hmm
12:48 pdurbin
romainM: if you want you can send the list of 45 fields to the mailing list
12:48 romainM
That's not "a problem", because they don't know for now that I'm looking at dataverse
12:48 romainM
as a solution for their needs
12:49 romainM
so maybe they will find some other fields that correspond to their needs
12:49 romainM
ah ?
12:49 bencomp
romainM: what kind of data are you managing?
12:49 romainM
I'm working as a dev for the cdsp
12:49 romainM
a mix of cnrs and sciences po
12:50 romainM
we are doing quantity and quality surveys
12:50 bencomp
interesting!
12:50 bencomp
searchbot: lucky cdsp
12:50 searchbot
bencomp: http://cdsp.edu/
12:50 romainM
it's social science, I woudl say
12:50 romainM
hum
12:50 romainM
not that one cdsp
12:51 romainM
:D
12:51 romainM
http://cdsp.sciences-po.fr/
12:51 romainM
it's more like this :p
12:51 romainM
http://cdsp.sciences-po.fr/index.php?lang=ANG&idTexte=accueil
12:51 romainM
in english
12:52 bencomp
that makes more sense :)
12:53 bencomp
romainM: all the important people in CDSP are now at the RDA Plenary, aren't they? :)
12:53 romainM
what is that ? 0o
12:54 bencomp
research data alliance conference
12:54 bencomp
it's in Paris right now
12:54 romainM
ah
12:54 romainM
maybe
12:54 romainM
well, I'm a new employee here, so I'm not really "aware" of all that things
12:54 bencomp
I bet Mercè is there too, isn't she pdurbin?
12:55 bencomp
:)
12:55 romainM
(it's like 3 months now, and I had nothing to do with socio-political datas :D)
12:55 romainM
*before
12:56 romainM
now that you mention it, that's maybe why my boss isn't there today
12:56 bencomp
hehe
12:56 bencomp
so are you comparing storage systems?
12:57 romainM
my boss is comparing
12:57 pdurbin
bencomp: looking at https://twitter.com/mercecrosas doesn't offer me a clue
12:57 romainM
and my boss told me to see what we could do with dataverse
12:59 pdurbin
romainM: well, we're here to help :)
12:59 romainM
and with some specs
12:59 romainM
(the duplicate function, for example, was one of them)
12:59 romainM
and I'm here to have help :D
13:01 pdurbin
:)
13:02 romainM
I was wondering, with the python api
13:02 romainM
is it intended to get only our own created dataverses through the Connection.get_dataverses() function ?
13:03 bencomp
pdurbin: I'm also here to be critical ;)
13:03 pdurbin
bencomp: I think you'd love Haskell :)
13:04 bencomp
a colleague of mine is a big fan of Haskell, but I haven't really looked at it yet
13:04 pdurbin
romainM: it looks like get_dataverses is getting the SWORD Service Document: https://github.com/IQSS/dataverse-client-python/blob/f84b980545aec73325eba88417c36a61deb79a9a/dataverse/connection.py#L88
13:05 bencomp
I'm thinking learning Java 8 and Scala is more urgent
13:06 pdurbin
romainM: and the Service Document returns a list of dataverses on which you have the AddDataset permission: https://github.com/IQSS/dataverse/blob/v4.1/src/main/java/edu/harvard/iq/dataverse/api/datadeposit/ServiceDocumentManagerImpl.java#L63
13:07 pdurbin
the purpose of the Service Document is to tell you where you can create a dataset. in which dataverses
13:07 romainM
hum
13:08 romainM
I'm not a big sword user, mb
13:08 romainM
I'm not a sword user at all, actually
13:08 romainM
:D
13:08 pdurbin
well, SWORD is based on XML . We also have a "native" API based on JSON
13:08 romainM
so, that's what I thought, and what I actually "feared" :S
13:09 romainM
ya, I will probably see that
13:09 romainM
I wanted to make some specifics export functions
13:09 romainM
but if I can't access the others dataverses, that won't do it ^^
13:10 bencomp
romainM: there is a search API
13:12 pdurbin
SWORD can pretty much only be used by owners of datasets and admins of dataverses. if you're trying to export published data from other people's dataverses, I would use the native API : http://guides.dataverse.org/en/latest/api
13:14 romainM
I will try that then
13:15 pdurbin
romainM: it will export the dataset metadata at JSON . Are you interested in any particular format?
13:16 romainM
json is ok for me, I will format it myself for what they want
13:17 pdurbin
great
13:17 romainM
they want a custom json format and XML -EAD format, as far as I can tell for now
13:17 pdurbin
searchbot: lucky xml-ead
13:17 searchbot
pdurbin: http://www.loc.gov/ead/
13:18 pdurbin
haven't heard of that one
13:18 pdurbin
I'm actually working on this right now: Export Descriptive Metadata From Dataset · Issue #907 · IQSS/dataverse - https://github.com/IQSS/dataverse/issues/907
13:19 romainM
I haven't myself before seeing the specs :D
13:22 bencomp
EAD is fun, especially since version 3 is brand new and incompatible with EAD 2002
13:23 pdurbin
heh
13:23 bencomp
if you're starting fresh, are you looking at EAD 3, romainM?
13:23 romainM
I don't know
13:23 romainM
well, I don't know what "they" use
13:24 bencomp
"incompatible" = there is no trivial translation
13:24 romainM
if it was my choice, I would go straight to v3
13:25 bencomp
I haven't had an in-depth look at it yet, but I believe it's richer and more suitable for machine interpretation
13:27 pdurbin
bencomp: if you like Scala you could hack on https://github.com/IQSS/DataTaggingServer :)
13:30 bencomp
:)
13:30 pdurbin
something for everyone :)
13:33 bencomp
https://github.com/IQSS/dataverse/issues/2576 for IQSS ;)
13:35 bencomp
pdurbin: why did someone copy source files for a plugin system? did they look at OSGi?
13:36 pdurbin
bencomp: copy source files? where?
13:37 bencomp
package edu.harvard.iq.dataverse.ingest.plugin.spi;
13:38 bencomp
ServiceRegistry and PartiallyOrderedSet
13:39 * pdurbin
looks at https://github.com/IQSS/dataverse/blob/v4.1/src/main/java/edu/harvard/iq/dataverse/ingest/plugin/spi/ServiceRegistry.java and https://github.com/IQSS/dataverse/blob/v4.1/src/main/java/edu/harvard/iq/dataverse/ingest/plugin/spi/PartiallyOrderedSet.java
13:40 pdurbin
bencomp: I have no idea what this code even does. You might want to mention it in the issue as an example.
13:43 bencomp
done
13:43 bencomp
edited the original description
13:47 pdurbin
thanks
14:49 romainM
the native works perfectly for me
14:49 romainM
cool
14:51 romainM
except some ... untitled datasets in the root ?
14:53 pdurbin
romainM: no titles? are they test datasets?
14:53 romainM
dunno
14:53 romainM
got the ids if interested
14:53 romainM
22, 30, 49 and 52
14:53 pdurbin
meh, if it's on a test server I wouldn't worry about test datasets :)
14:54 romainM
well, all the datasets of the root have no title field
14:54 romainM
that makes maybe more sense
14:54 romainM
dataset 22 : {u'protocol': u'doi', u'authority': u'10.5072/FK2', u'persistentUrl': u'http://dx.doi.org/10.5072/FK2/81ZOVB' , u'identifier': u'81ZOVB', u'type': u'dataset', u'id': 22}
14:55 romainM
not saying if it's a bug or not, just wondering if it's intended
14:55 romainM
because dataverses do have their title, for example
14:55 romainM
{u'type': u'dataverse', u'id': 16, u'title': u'Philip Durbin'}
14:56 pdurbin
oh, you mean you'd like the python client to give you the title
14:56 romainM
nope
14:56 romainM
that's the native here
14:56 romainM
what I mean, is that by asking for the root content
14:57 romainM
with a get http://"+HOST+"/api/dataverses/:root/contents
14:57 romainM
I got the list of dataverses and datasets of the root
14:57 pdurbin
oh. but no titles for datasets?
14:57 romainM
yep
14:57 romainM
that's strange to show titles for the dataverses, but not datasets
14:57 pdurbin
romainM: can you please create an issue at https://github.com/IQSS/dataverse/issues ?
14:58 romainM
that's completly minor, just wondering ^^
14:58 romainM
ok :)
14:59 pdurbin
I haven't used that "contents" endpoint much. There are known issues with it: https://github.com/IQSS/dataverse/issues/1160 and https://github.com/IQSS/dataverse/issues/2122
14:59 axfelix joined #dataverse
15:00 axfelix joined #dataverse
15:11 romainM
done
15:12 romainM
dunno if I was clear enough ^^"
15:24 * pdurbin
looks at Native API : contents json result , got title for dataverses but no title for datasets ? · Issue #2577 · IQSS/dataverse - https://github.com/IQSS/dataverse/issues/2577
15:24 pdurbin
romainM: looks great. I edited it to add code fence blocks (a Markdown thing... ```code```)
15:46 bencomp
pdurbin: did you save your edit?
15:47 pdurbin
bencomp: whoops! I must have only previewed it. saved now
15:47 bencomp
:)
15:48 pdurbin
thanks
16:01 axfelix joined #dataverse
16:12 metamattj joined #dataverse
17:30 metamattj joined #dataverse
17:44 petrichor left #dataverse
17:53 cnk joined #dataverse
19:19 petrichor joined #dataverse
20:36 garnett joined #dataverse
21:39 garnett joined #dataverse
