Difference between revisions of "Telecon (2019-02-06)"

Latest revision as of 14:28, February 8, 2019

Return to Archive

Attendees

Kristin Vanderbilt
Colin Smith
John Porter
Stevan Earl

Agenda

Identify potential topics for a spring hackathon

Discussion

What would be a good theme for the next hackathon:

Build on EML AssemblyLine to make EML generation easier. Integrate with Metabase project – using a common set of database view as basis for generating EML. Kristin would like a spreadsheet form of the attributes component of the EML AssemblyLine.
Visualizations: feed EML to document to service and get visualization output. Hard part is to determine x and y dimensions. Use this software not for analysis, but for determining data fitness for use. R DataMaid might produce the kinds of graphs we need. Emery has a system that already does visualizations like this – check with him on how he implemented it. Visualization software would be implemented using a web service so it can be called from anywhere. Interactive graphs are a desirable feature. Mark Servilla recently excited by opportunity to use Google Earth Engine for spatial viz.
EML 2.2 and Annotations. LTER used to have a tool that you could put text in to and it would provide a list of keywords from the LTER Controlled Vocabulary. John used this to return a snippet of EML to go into VCR EML. But some people want just a comm-separated list of terms, others want the snippet. Need a web editor for EML to directly import the output of this service. New Morpho may have capacity for annotation; need to associate terms provided by the metadata creator with the PURL. Mark Servilla has talked about making value-added metadata based on machine-learning. Annotation challenge: How does one identify the appropriate concept with which to annotate. E.g., if one has a variable generated by using a particular method, how to you locate the concept for that method?
Metadata exporter tool – to produce NetCDF, for instance. Could have capacity to subset. Would this need to be implemented by EDI portal? Not necessarily. Could be a web service. Amazon Web Services are dandy but cost money.

Action items

Decide on a hackathon topic

@@ Line 1: / Line 1: @@
-Return to [[IM Code Registry]] main page
+Return to [[Archive]]
-== Controlled vocabulary ==
+===Attendees===
+* Kristin Vanderbilt
+* Colin Smith
+* John Porter
+* Stevan Earl
+===Agenda===
+* Identify potential topics for a spring hackathon
-==== Overview ====
+===Discussion===
-[http://vocab.lternet.edu/vocab/registry/index.php The IMCR Controlled Vocabulary (CV)] is designed improve discoverability of registered software. The IMCR controlled vocabulary is organized around [https://www.dataone.org/data-life-cycle the data life cycle] with the notable exclusion of the "analysis" phase of the cycle (we consider analysis tasks outside the scope of information management and thus the IMCR). Extension of the controlled vocabulary to "science domain" is underway. Contact Colin (colin.smith@wisc.edu) if you have comments or questions.
+====What would be a good theme for the next hackathon:====
+* Build on EML AssemblyLine to make EML generation easier. Integrate with Metabase project – using a common set of database view as basis for generating EML. Kristin would like a spreadsheet form of the attributes component of the EML AssemblyLine.
+* Visualizations:  feed EML to document to service and get visualization output.  Hard part is to determine x and y dimensions. Use this software not for analysis, but for determining data fitness for use. R DataMaid might produce the kinds of graphs we need. Emery has a system that already does visualizations like this – check with him on how he implemented it. Visualization software would be implemented using a web service so it can be called from anywhere. Interactive graphs are a desirable feature. Mark Servilla recently excited by opportunity to use Google Earth Engine for spatial viz.
+* EML 2.2 and Annotations. LTER used to have a tool that you could put text in to and it would provide a list of keywords from the LTER Controlled Vocabulary.  John used this to return a snippet of EML to go into VCR EML.  But some people want just a comm-separated list of terms, others want the snippet. Need a web editor for EML to directly import the output of this service. New Morpho may have capacity for annotation; need to associate terms provided by the metadata creator with the PURL. Mark Servilla has talked about making value-added metadata based on machine-learning. Annotation challenge:  How does one identify the appropriate concept with which to annotate.  E.g., if one has a variable generated by using a particular method, how to you locate the concept for that method?
+* Metadata exporter tool – to produce NetCDF, for instance.  Could have capacity to subset. Would this need to be implemented by EDI portal?  Not necessarily.  Could be a web service. Amazon Web Services are dandy but cost money.
-==== Action items ====
+===Action items===
-* Extend the CV to cover "science domain". Anyone is welcome to contribute to [http://bit.ly/imcrvocab this working document] in which development of the "science domain" categories and terms is occurring.
+* Decide on a hackathon topic
-* Refine the CV. Suggest new terms and definitions or refine existing ones by contacting Colin (colin.smith@wisc.edu).
-* Get software terms form Kepler ontology (Kristin is doing this).
-==== Links ====
-* [http://vocab.lternet.edu/vocab/registry/index.php The IMCR Controlled Vocabulary]
-* [http://bit.ly/imcrvocab CV working document]
-* [https://groups.niso.org/apps/group_public/download.php/12591/z39-19-2005r2010.pdf Guidelines for the construction, format, and management of monolingual controlled vocabularies]