Community Dynamics Toolbox

h1. OBSOLETE, All notes Moved to Github at https://github.com/NCEAS/commdyn/wiki

- Do not edit or add to these notes here.  Use github and/or etherpad instead.

h1. Project Team Meetings

h2. 7/18/2014

Matt: good idea to combine codefest with ABI workshop because there are other people working on similar technical issues. We may need a little time for specific community dynamics discussion.

Work on scientific manuscripts during codefest and add Thursday afternoon and Friday morning. Travel on Friday afternoon and Saturday morning.

Codefest will offer great exchange on scientific R programming. R open Sci will be there. Analysis and infrastructure.

Corinna send e-mail to commDyn group. Get quick notice of who wants to come. Ask what people still need to do for their manuscripts. Add Peter, Chris and Lauren to logistics e-mails. Put together an agenda: Thursday: present and do some usability work on the new collaborative tool and get feedback. Friday: work on manuscript, data analysis, 

Codefest Tuesday - Thursday noon.

Matt report: collaboration between three projects (DataONE, commDyn, GeoLink) built in capabilities to share code during development. Library of community dynamics metrics. Lauren has started to develop a prototype. Have a prototype to show during codefest. Get input on how to extend for work in commDyn.

h2. 6/16/2014

h2. 5/9/2014

Syd, Matt, Scott, Corinna

Syd and Scott at ASU working with Julie: 

late breaking poster for ESA

8 Konza datasets

7 Sev datasets

richness, timelag, 

comparing taxa at same site

grass community is changing, not so much creosote; composition is changing, not so much richness

not many community datasets at LTER sites

VCR small mammals are not usable

not as extensive a comparison as originally hoped

Konza data show tipping point with shrub encroachment

characteristics of datasets:

multiple communities at one site, producers, consumers

Matt: Lauren Walker has started working on project. Ben and Peter will join later.

work across several projects that are trying to do the same

define a model that allows people to upload R scripts, datasets and analytical output. All linked to understand what was done.

Share everything as the analysis is going on, through the DataONE repository.

How should that be related/linked to github. Git is good for actively working, but not archiving. I.e., the archival copy within DataONE needs a link to the living code in git. Version in DataONE will stay archived with DOI, while the version in git may go away.

Drawing up some preliminary design ideas.

Syd should document what worked and what didn't in git.

software un-coference in September, venue locked in. Science/software work-a-thons.

Scott: Follow up with other working group (Lauren, Lizzie, Elsa et al.) 

h2. 4/4/2014

Scott, Matt, Corinna, Syd

Scott and Syd going to ASU end of April to work with Julie

Syd: report from last group call

larger group participated, decided that each working group should meet separately

group 1: Syd, Julie, Eric

Eric posted diversity metrics on github

challenge: choose the metrics

datasets: how will they be chosen

plant community dataset, annually sampled, plus insect community dataset, or small mammal community - multiple datasets from the same sites, annually sampled. Limitation is co-location of datasets. SEV, KNZ, JRN, maybe VCR

Matt: discussion with Ben and Lauren where to go with the repository. Initially a web interface to drop things and search and find. Next step is to access them within the working environment (R, Kepler).

h2. 3/19/2014

Scott, Matt, Corinna

Annual report: Scott has submitted

reconvene the workshop group to continue work on papers. Possibly with ISEES meeting: possibly September 2 - 4 followed by commdyn workshop: September 5 - 7 

Action items:

Matt to write up and circulate specifications for Ben to work on starting in May, 2014, based on recommendations from the workshop

collaboration tool to share workflows should work across platforms, not only Kepler

Corinna send thank you e-mail to workshop participants. 

Corinna move everything from redmine to github.

set up times for paper working groups to meet online. Start with the one Syd already initiated, then follow with other groups.

set up structure in github for the four groups.

h2. 2/27/2014

Scott, Matt, Syd, Corinna

Possibly have a follow up workshop in September together with ISEES meeting in Santa Barbara

Important to keep the good momentum going for the different working groups that have started.

Syd:

write e-mail to remind everyone of what we agreed to do.

h2. 2/13/2014

Scott, Matt, Corinna

Discussion of the workshop outcomes.

h2. 11/22/2013

Scott, Syd, Corinna

Workshop preparation: What to ask people to do in preparation for attending:

bring a dataset - multi species and multi sampling

how do you want to analyze your data

bring R code if available

What questions can be asked across several datasets of different taxonomic groups and habitats

Syd:

list of indices that we already have identified on this website

R code to handle the four example datasets

Corinna:

Kepler workflow to access the data via EML and pass on a R dataframe

Or R code accessing datasets in DataONE

h2. 10/11/2013

Matt, Scott, Corinna

plans for the workshop:

treat part of the workshop as hackathon, combine ecologists, R specialist, and workflow specialist in groups to actually implement an analysis.

Data sets: people should bring their data sets.

Syd to find appropriate data sets and give feedback on how to find them. 

goals of workshop: 

list of analyses, list of data sets, what is need to move from raw data to analyses and visualization.

https://docs.google.com/spreadsheet/ccc?key=0AgZm2NvfJ87OdFlBdWRQMkNwWnpCT3FkcEJDb2gtZ0E&usp=drive_web#gid=0

h2. 9/17/2013

Matt, Scott, Syd, Sam, Corinna

Started list of potential invitees for the workshop in google docs.

h2. 8/27/2013

Matt, Scott, Syd, Sam, Corinna

reports:

Matt is still trying to hire a programmer

Corinna announced project at LTER IM meeting and got several responses regarding datasets available. Specifically GCE, LUQ, and VCR

Syd has started R programming for some community change indeces

workshop planning:

start document for goals, participants and agenda

action items:

Corinna start document for workshop planning

Syd and Sam communicate about R routines for data manipulation

Syd and Scot get names of datasets from SEV and KNZ to Corinna

Corinna and Sam work with Kristin and Adam to get EML for those datasets

h2. 6/13/2013

Matt, Scott, Syd, Sam, Corinna

Agenda

Action items from last meeting

Scott - report from the Science Council meeting re. datasets

Matt - intro to project management system and start of tracking Community Change indices

Corinna - start list of participants for first workshop at NCEAS

Notes

Scott announced the project at the LTER Science Council meeting

Important to remember is that we have support to clean data and document them in EML. We are not expecting the information managers at other sites to add this to their tasks. We can take datasets that have at least 10 data points at even time intervals.

Action items:

Matt: add functionality to this project management system to track datasets from a general idea of its existence to actually being in the NIS/DataONE

Corinna: start adding Community Change indices from the proposal to redmine. 

h2. 5/2/2013

Matt, Syd, Corinna

Today we filled Syd in a bit on her role in the project and talked some more about the first workshop in spring 2014.

Action items:

Corinna - doodle for the community workshop in February/March and set date

Syd - download and explore Kepler

Scott and Syd - start list of community change metrics that we want to encode

Matt - hire programmer

Next project team meeting: 6/13/2013 1 pm MDT