Versions Compared

Old Version 2

changes.mady.by.user Jay Alameda

Saved on

compared with

New Version Current

changes.mady.by.user Jay Alameda

Saved on

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.
Comment: Migrated to Confluence 4.0

Outward facing software website - what sorts of things do we want to aggregate?

  • do we need an independent .org site?
    • YES!
      • opensciencetoolkits.org
      • nationalsciencesoftware
      • sciencesoftwarecommunity.org
      • sciencesoftwarecollaboratory.org
      • scientificsoftware.org
  • Available domain names -- please give us your feedback!
    • scientificsoftwaresystems.org
    • openscientificsoftware.org
    • academicscientificsoftware.org
    • softwareforscience.org
    • sciencesoftware.org
    • software4science.org
  • mailing list -
    • start at ncsa, then .org
    •  
    • what else?
  • monthly webinar talks -
    • licenses
    • best practices
    • microtalks followed by discussion (5-10 minutes) -
      • figure technology, get talks set up in advance
      • advertise, recruit speakers
      • find a volunteer for organizing
    • ideas
      • openMRS - open medical record systems
        • succeeded in open source - practices in how they acomplished
    • 5 short presentations - say on
      • build and test?
      • verification?
      • licensing?
      • tools - eg, tools Todd Tannebaum suggested that are useful in Condor
      • workflow management - several open workflow managers - who is working with them, experience, etc
      • integration experiences
      • software engineering
      • consortiums -how (developers, users) deal with consortiums, governance models
      • distributing sw as vm images -
      • education -
        • steve gordon, OSC, heading up xsede education could give a talk on
      • training -
        • xsede training - on how to interact, assist with training
      • Suresh/Marlon on Science Gateways, how communities come together, mechanisms for support
      • John Cazes - community codes, how we work with people
      • Mark Fahey - collaborative support
      • Marlon: presentation on xsede technology insertion service, if you have gone through the process, what do you need to do - how can push back to TIS on what we'd like for them to do -
      • Marlon: topic on EOT section on outreach on underrepresented groups, would like discussion on what people are doing, Richard Tapia or SURA contact, to join/lead this discussion -
      • software sustainability institute /UK - Edinburgh -
        • Steven Newhouse (EGI)
      • software taskforce - would be good to have someone from that group talk about these recommendations -
        • Valerie Taylor
        • David Keyes
      • what I hate - things that make my life miserable/less pleasant -
        • within scope of software...
        • not necessarily deep -
        • eg - time to get allocation
      •  
    • opportunities for cross-project-hackathons
    • postdoc/student exchanges between groups
    • tutorials - in basic scientific computing topics -
      • could brainstorm these topics on this site -
    • XSEDE CI fellows, trying to figure out how to execute this - have someone go on sabbatical to work on CI project - may be useful way to connect XSEDE with SI2 projects -
      • should be announced soon, in final reviews now -
      • work on something "not your own project"
      • campus champions, get them up to speed on advanced support
      • SI2 could offer some of the tools, this could be a cooperative venture
    • Useful to have screen capturing/introductory tutorial - maybe tutorial, tools to do this -
    • many of our software target HPC, how do you leverage potential of XSEDE resources - this would be useful to increase performance of the software -
      • bring people folks in from xsede to share experience
      • system sw tuned for hardware platform -
      •  
  • google summer of code -
    • have
  • (links to)
    • software?
    • documentation?
    • licenses
      • contributor agreements
    • sw engineering practices
  • table of software - software, documentation, licensing, with a hatch mark - for existence (and a link?)
    • divide by discipline
    • figure out guidance on suitability of software, or applicability of software
      • level of experience /recommended use - eg, suitable for teaching undergraduates?
  • things to put in the .org
    • web server
    • wiki
      • mediawiki?
      • confluence?
      • hashtag
      • facebook account -
    • keep this small/focused -
  • what about putting into wikipedia, and actually have we maintain the page -
  • taxonomy -
  • this page should represent another level of collaboration - community on how they map together, this may be an extra reason for our own domain name/wiki/etc -
    • want to avoid having 6 places to update things -
    •  

Community issues - revisit town hall and initial community discussion -

Metrics - what do we think it is important

  • Guidance for the community
  • Marlon: one good metric - how many other SI2 projects or closely related NSF projects are you collaborating with to the point where you give them commit access to your (svn, cvs, github,...)
    • measuring amount of connectivity of your project with this community, projects within this community
  • head counts - get an idea of how big the development community is, how big the user community is -
    • how many groups have code developments
    • how many licenses
    • how many users
    • how many scientists use scientific sw on their research
    • how many citations back to the original paper for a particular code
  • need a way to go beyond this not just counting, but measure their increase - platform for people to increase their impact
    • esp when encouraging cross-collaboration/multi-disciplinary collaborations
  • number of new users/year
    • # of lines of code removed
  • number of phd thesis that used their sw -
  • usage satisfaction - qualitative -
  • # of times the sw is mentioned in (nsf, nih, ...) proposals
  • metrics for code test suite coverage -
    • comment coverage...
  • how complex is your code? (entangled) -
  • # of software that interfaces with your code
    • eg GAMESS - interfacing with MD codes -
      • both input files and APIs
      •  
  • # of institutions that use your code for their curriculum
  • performance changes directly attributable to NSF funding
  • # of SUs used on XSEDE resource
    • useful SUs?
  • front covers of magazines
  • mailing lists, # of items put on mailing lists -
    • what percent of items on mailing lists are from people funded by NSF - to show engagement with the user community
  • quantify extent code is used as a benchmark
  • # of developers graduated from project - demonstrates continuity/sustainability if you can survive lots of turnover
  • # of releases made
  • documentation - and length of documentation
  • bugs filed, bugs fixed
  • code churn -
  • # of training events
    • attendance of training events
  • education modules produced
    • courses using education modules
  • science enabled - in citations
    • impact
  • tracking papers published - h-index -
  • worldwide sw downloads
  •