[LTER-im-rep] [dataone] Announcing DataONE Search: A New Way to Discover Data
Inigo San Gil
isangil at lternet.edu
Tue Nov 17 13:44:50 MST 2015
Hi Amber,
Im glad you could reproduce results. These seem to be either the same
data record, or potentially worse, versions of the same or very similar
record. And they appear to be that way. Do we want a person looking
for the right version amongst all these records? I can see the
frustration coming from a mile away.
In any case, I am not sure what is more confusing from the visitor's
perspective, to have 27 equal versions, or 13 that differ slightly, but
share a very similar title.
Tough problem though -- I briefly saw on the list of providers
(partners) are also groups who provide similar services (data cataloging
and discovery). These are all providing similar if not the same
records. This federated idea (early 2000s) is great, but the thing is
there are many federations competing for the same records.
Other issue raised over here recently, from the data provider
perspective. We may want to take down some of those data versions in an
easy-no-fuss way.
So far the deletions process may involve one or all of the following:
consensus, delegation, forwarding, doodling, stalling or other
interesting dances.
Some of the LTER IMs would welcome an easy and direct way to delete any
of the data records that we originate or we originated in the past, for
whichever reason.
One such reason is precisely seen forty two versions of a data package,
and we would like to alleviate the dizziness that a visitor interested
in our data may have while visiting your warehouse. But there may be
other reasons. Does it sound reasonable?
What do I mean 'delete' easily? not just an easy to use API, but also a
'delete' button next to each dataset that would grant the provider
(partner?) the option of removing unwanted data.
Data are forever, right? Well, not bad data. Bad data are there
temporarily, until we correct and amend it. Yes, we need the delete
option. And the Edit/Update option. And we do not want to see those bad
versions replicating like a cancer around the data federations.
Can you [dataOne] keep bad data? I do not really have an opinion about
it, but if there is a request NOT to offer it to the public, it sounds
logical to me that request should be granted.
But.. a DOI was issued? Yes, thank you. I am sure the DOI can resolve,
for example, to a message explaining that 1) this bad data was taken out
by a responsible data owner, and 2) here is a link to the better version
of the original data and 3) if you must have it, login first to access
it. What, this is not the intended and envisioned use of DOIs? I am
aware of that, but let's be realistic, nothing is forever. Not diamonds,
and certainly not DOIs.
Cheers,
Inigo
On 11/17/2015 12:48 PM, Amber E Budden wrote:
> Hi Inigo
>
> I repeated your search teams and also discovered 58 records with
> apparent duplicates. However, looking through the first set (Rawlence
> et al) I see that they are actually different records, albeit with the
> same author list and repository. The titles do differ.
> In the later records (from LTER), the records share the same title and
> authors but have different package numbers.
>
> DataONE Search delivers results from a search against the metadata
> catalog hosted by our Coordinating Nodes. The way that data are
> packaged (and appear within the search results) is determined by the
> individual Members Nodes. Although the records appear similar, they
> are distinct within the Member Node repository and as such appear as
> separate results.
>
> Hope this helps
> Amber
>
> On Tue, Nov 17, 2015 at 11:36 AM, Amber E Budden
> <aebudden at dataone.unm.edu <mailto:aebudden at dataone.unm.edu>> wrote:
>
> Hi Inigo
>
> Thanks for the feedback.
> Yes, we have been having a few timeouts with the high volume of
> traffic going to the site and are looking into this. I will also
> forward your comments about duplication.
>
> Please do let us know of any thing else you encounter. We
> appreciate all feedback from users.
>
> Best
> Amber
>
> On Tue, Nov 17, 2015 at 11:26 AM, Inigo San Gil
> <isangil at lternet.edu <mailto:isangil at lternet.edu>> wrote:
>
>
> Hi,
>
> Here is some feedback about the new and improved dataOne search.
>
> I wish it was all positive. It is not. It is feedback.
>
> --Lots of duplicates on that search app (58 results, all
> apparent duplicates. two filters keys: biodiversity;
> creator(?) smith).
>
> - No visible "support" link or report problem. No support?
>
> - And then, the site down. Great.
>
>
>
> Go metacat!
>
> Inigo
>
>
>
>
>
>
> On 11/17/2015 10:59 AM, Amber E Budden wrote:
>> *Announcing DataONE Search: A New Way to Discover Data*
>>
>> DataONE is pleased to announce the release of DataONE Search,
>> a web based application allowing users to seamlessly and
>> efficiently discover the 168 thousand publicly accessible
>> data packages* within the DataONE federated network of Member
>> Nodes.
>>
>> DataONE Search allows users to search by keyword and refine
>> by filters including: data attribute, Member Node, creator,
>> year, identifier, taxa, location and whether a complete data
>> package including both metadata and data are available for
>> download. The geographic search interface in DataONE Search
>> provides a convenient overview of the spatial distribution of
>> data sets and allows users to zoom and pan to specific
>> locations of interest.
>>
>> Summaries of the number of data and metadata files, file
>> formats and collection time periods are also made available
>> through the new DataONE Search site. Such summaries provide a
>> convenient overview of the range of data accessible through
>> the DataONE federation from a number of different perspectives.
>>
>> DataONE Search is now the default software for search and
>> discovery of data and metadata within DataONE. The software
>> has been designed to facilitate rapid iteration and
>> deployment of new features and to take full advantage of
>> future capabilities offered by upcoming versions of the core
>> cyberinfrastructure. Notable among these is the upcoming
>> provenance trace within search and discovery (see
>> https://www.dataone.org/webinars/provenance-and-dataone-facilitating-reproducible-science
>> for more information). ONEMercury will remain available to
>> users via the direct URL https://cn.dataone.org/onemercury/.
>>
>> To view demos on how to use DataONE Search, view our
>> screencast tutorials at:
>> https://www.dataone.org/screencast-tutorials.
>>
>> *This number excludes obsoleted records within the network
>> that are publicly readable
>>
>>
>>
>> About DataONE:
>> DataONE enables universal access to data and also facilitates
>> researchers in fulfilling their need for data management and
>> in providing secure and permanent access to their data.
>> DataONE offers the scientific community a suite of tools and
>> training materials that cover all aspects of the data life
>> cycle from data collection, to management, analysis and
>> publication.
>>
>>
>> About MetacatUI:
>> DataONE Search is a customized implementation of the open
>> source MetacatUI search application
>> (https://github.com/NCEAS/metacatui), developed at the
>> National Center for Ecological Analysis and Synthesis, and
>> usable by any DataONE Member Node that supports the DataONE
>> query API. MetacatUI is a client-side web interface for
>> querying Metacat servers and other servers that implement the
>> DataONE REST API. Currently, it is used as the search
>> interface for DataONE (https://search.dataone.org), The
>> Knowledge Network for Biocomplexity
>> (https://knb.ecoinformatics.org/#data), the Gulf of Alaska
>> Data Portal (http://goa.nceas.ucsb.edu/#data), GLEON
>> (https://poseidon.limnology.wisc.edu/metacatui/), and the
>> Montana IoE Data Repository (http://montanaioe.org/).
>>
>> --
>> Amber E Budden, PhD
>> Director for Community Engagement and Outreach
>> DataONE
>> University of New Mexico
>> 1312 Basehart SE
>> Albuquerque, NM 87106
>>
>> Tel: 505-814-1112 <tel:505-814-1112>
>> Cell: 505-205-7675 <tel:505-205-7675>
>> Fax: 505-246-6007 <tel:505-246-6007>
>>
>>
>> _______________________________________________
>> community mailing list
>> community at dataone.org <mailto:community at dataone.org>
>> To unsubscribe or change your subscription, emailsupport at dataone.org <mailto:support at dataone.org>
>
>
>
>
> --
> Amber E Budden, PhD
> Director, Community Engagement and Outreach
> DataONE
> University of New Mexico
> 1312 Basehart SE
> Albuquerque, NM 87106
>
> Tel: 505-814-1112 <tel:505-814-1112>
> Cell: 505-205-7675 <tel:505-205-7675>
> Fax: 505-246-6007 <tel:505-246-6007>
>
>
>
>
> --
> Amber E Budden, PhD
> Director, Community Engagement and Outreach
> DataONE
> University of New Mexico
> 1312 Basehart SE
> Albuquerque, NM 87106
>
> Tel: 505-814-1112
> Cell: 505-205-7675
> Fax: 505-246-6007
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lternet.edu/pipermail/im-rep/attachments/20151117/23e69af2/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/png
Size: 30941 bytes
Desc: not available
URL: <http://lists.lternet.edu/pipermail/im-rep/attachments/20151117/23e69af2/attachment-0001.png>
More information about the im-rep
mailing list