Skip to content

GoalsMilestones

Jump to bottom Edit New page
Arlin Stoltzfus edited this page Feb 2, 2018 · 23 revisions

Goals for 2018

At the f2f meeting, we imagined a vision of success by the end of the year, and set goals based on that.

Vision for what we see in 12 months (if successful)

  • Lots of users using our tools and providing feedback
  • Researchers reaching out about our frameworks, automated reasoning, etc.
  • We are responding to scalability challenge with redundant hosting
  • People see that depositing trees gets them citations, so they deposit more
  • There will be a phylotastic community, perhaps organized via a central website

2018 Goals based on above vision

  • Software and services
    • Support top 5 (priority 1) use-cases to the extent feasible (Q2)
      • Implement IUCN status and habitat service (Q1) (Abu)
      • Implement portal support for viewing table of data (Q2)
      • Implement portal support for downloadable data (csv, NEXUS, etc) (Q2)
      • Implement popularity service or other taxon sampling (Q1) Arlin
      • Implement common name service (Q1) Abu
        • Explore contracting more capable version from Yan (Arlin)
    • Explore feasibility of trees on wikipedia (Q1) (Luna)
    • Workflow composition and execution
      • Complete, tested method to generate similar workflow from plan plus revised requirements (Q1)
      • Workflow execution module recovers from failure by re-planning and re-execution (covering set of desired exceptions) (Q2)
      • Support responsiveness to a range of user preferences. (Q3)
        • Define range of flexibility in user needs
        • Decide how much of that range to support
        • Implement support
    • Toolboxes
      • Submit compliant Rphylotastic package to CRAN (Q1, BCO, Luna, Abu)
    • Services
      • Implement phylogenerator as third tree generation service. Satisfy all the requirements (implementation, WSDL, documentation with examples) (Q4, Abu w Brian)
    • Portal
      • Integrate workflow planning. Portal implements user interface that accesses automated workflow planning, displays results in integrated fashion following portal conventions for metadata and workflow description. (Q3)
      • Robust repeatable deployment. docker run phylotastic/portal generates a portal that works on whatever computer does it, does not require further intervention or customization to support web users. (Q2)
      • Improve reliability and robustness. (Q1)
        • Monitor server failure and automate restart upon failure. (Abu)
        • Monitor failure of services (used by portal) and automate restart of project services; notify in case of external services.
        • Identify workflow failure point 100 % of time
        • Upon workflow failure, notify user with plain-language description of failure and instructions.
  • Education and outreach objectives
    • Grow our user base, grow community (Q4) TBN, Arlin
      • Social media (BCO and postdoc(s)).
      • Sharing buttons on portal
      • Regular tweets or other posts (needed: website with restful URLs)
      • Presentations at meetings,
      • Project stickers, posters
  • Monitor use and share results every month (AS)
    • Manuscripts
      • Datelife paper 1: Manuscript submitted to journal (finish tests, write ms, post to bioRxiv, submit ms to journal). (Q2, Luna and Brian)
      • Phylotastic project paper 1. Manuscript submitted to journal (write ms covering portal, services, toolbox, registry, . . . ) (Q2) Arlin
    • Capitalize on pedagogical value of phylotastic resources. With educator collaborators, develop lesson plans for HS and university levels that are (or will be) used in classrooms, that are discoverable and accessible online. (Q3) Enrico
    • Present at teacher conferences (NABT meeting in Nov: https://nabt.org/Events-2018-Conference)
  • Project management goals (Q1 through Q4) Arlin
    • Focus on use-cases. Revisit list of use-cases at least quarterly, consider which ones are being used and which ones present strategic opportunities.

Publication targets (old, needs refreshing)

Through Q1, 2017

  • All - description of phylotastic system
    • Syst Biol application note
    • Soft release to OT and others: 15 Oct
    • Initial plan: 15 Oct
    • Manuscript: 15 Dec
  • Brian (DateLife) - DateLife overview paper
    • To do: more robust testing
    • Bioinfo journal
    • Could include character-based (BOLD) scaling service
    • Integrate scaling into portal and wrap as services
    • Initial plan: 15 Nov
    • Manuscript: 15 Jan
  • Arlin, Dail, Abu, collaborators - interoperable tree annotations
    • Evol Bioinformatics,
    • Initial plan: 1 Dec
    • Manuscript: 15 Mar
  • Dail, Arlin, Enrico - phylotastic in the classroom
    • Audience is educators teaching university intro bio or evolution or taxon-specific courses (insects, plants)
    • Need to identify venue
    • Include assessment (probably have to fill out IRB)
    • Could ask NSF for supplement for this
    • Connect with NMSU Ed Research Office
    • Now: Explore assessment needs and make plan to get approvals
    • Initial plan: 15 Dec
    • Manuscript: after formal assessment, if needed

Longer-term publication targets

  • Dail, Dima, Brian - holes in the ToL
    • Update to Drew, et al. (2013) Lost branches on the Tree of Life.
    • Compare to taxonomy and species with data in NCBI, find fraction of nodes supported phylogenetically for different phyla
    • Could also be in scope: how to assess “support” for a branch, conflict [bipartitions, quartets, etc]
    • To do: initial assessment of feasibility
  • Enrico, Thanh, Van - description of services and semantic framework for composition
    • Venue: applied computing journal; conference proceedings
    • Use portal for demonstration
  • NMSU team - metadata generation of description for services used using natural language generation
    • Venue: applied computing, AI conferences
  • Dima - name services (either standalone or part of phylotastic paper)
    • ZooKeys - venue for taxonomic publishing to semantic web
    • Problems & opportunities in name-finding & resolution & integration
  • Dima et al : common names
    • Bioinformatics journal describing the approach used
    • Common names from images (Watson API)

Quarterly Milestones (old)

Milestones for Q1 2017

  • Image service : support for user preferences (image popularity)
  • Interoperable tree annotations
    • Create pilot version
    • Export the right format from the portal
    • Ensure there are at least three other tools (ETE, Evolview, Mesquite) that will ingest this format
    • Make validator
    • Conversion to New Hampshire extended or other formats
  • Metadata paper
    • Implement workflow description via metadata from services
  • Web services composition
    • Prototypes for full web services
    • Registry web services descriptions
    • Service reliability and usage metrics [soft constraint in planning for composition]
  • Supporting assessment of coverage
    • Identify and display nodes supported by phylogeny vs. taxonomy

Milestones for Q4 2016

  • Publications (see pubs plan above)

    • General manuscript (15 Dec)
    • DateLife ms plan (15 Nov)
    • PhyloViz interop plan (1 Dec)
    • Educational uses plan (15 Dec)

  • Dev targets for general phylotastic paper

    • Mobile app to use list server, making lists accessible on portal
    • UI testing and improvements to portal
    • Portal saves tree viz annotations and modifications internally
    • Soft release up to 50 people

  • DateLife: methods to implement

    • Average consensus trees
    • Interface with rest of phylotastic

Milestones for Q3 2016

  • admin
    • report on first year
    • revised plan for second year
    • plan and execute f2f meeting
  • portal
    • implement viewer redesign
    • implement other UI redesign
    • ready for alpha release by early September
  • services
    • must have at least 2 tree sources (not just OT)
  • planning and composition
    • minimal demonstration
  • major release - Oct?
    • portal
    • manuscript
    • services & registry
    • documentation with worked examples
  • outreach
    • develop short slide presentation
    • begin outreach at meetings
    • make a dedicated effort to get educators involved
  • metadata
    • initial version of metadata model

Milestones for 2nd quarter of 2016

This will be supplemented by some 1st-quarter goals that we expect will be incomplete.

  • back-end services
    • in conjunction with testing (below), get reliability up to 80 % for name extraction and tree acquisition
    • add EOL image service (get image for species)
    • add species page-rank service
    • add DateLife scaling service
    • add phylotastic gallery service
  • generalized web UI
    • derive prioritized list of features based on user feedback (Dail)
      • minimally viable product (MVP)
      • additional features
    • build MVP (Van, Dail)
    • bring consultants on board for evaluation and redesign
  • mobile app
    • MVP redesign with better workflow & user control
      • list-capture-repeat-list-tree workflow
      • list management (add, rename, edit)
      • how-to
      • informative error messages
    • share with alpha testers (Arlin)
    • make plan to go ahead (or not) based on feedback
  • tree viewer criteria and evaluation
    • develop criteria for viewer
    • evaluate available viewers
  • design and implement minimal metadata model for trees
    • design model to cover date, methods, sources
    • implement model in mobile app and web tool
    • [design model for metadata in NeXML]
  • controller that implements automated planning (June 1)
  • manuscript
    • decision on main points
    • outline
    • figures
    • draft

Milestones for 1st quarter of 2016

  • Communication & outreach strategy (Jan 30) Arlin (Feb 24)
    • plan to cultivate community of testers & commenters
    • plan to educate, stimulate use of API
    • plan to publicize project progress
    • plan for publications to ensure professional credit
  • Services registry (Jan 30) Abu, Enrico (Feb 24)
    • Move service registry to a NMSU-CS server
    • Move all current services to the project server
    • Develop new services on the server
    • Implement public API
    • Set up portal
      • document API
      • document how to register new services (defer to Q2)
      • handle requests to register new services (defer to Q2)
  • New services we need for web UI and other applications - Abu Saleh
    • get species for taxon
      • get species from taxon & country
    • species list server
    • ? a second name-extractor (other than GNRD)
  • Services ontology - Thanh, Enrico
    • Develop ontology for description of web services
    • explore how to use ontology for web discovery
    • explore how to use ontology for service composition
    • add to service registry and to APIs
    • document
  • library API implementation in at least one language (January 30) - Enrico
    • Perl, Java, Python, possibly look at Galaxy, or R
    • what is the minimal workflow?
  • taxon match processor (Feb 5) - Dail, Arlin, Dima, Van, Enrico
    • develop use-cases (Dail)
    • design interactive version (Dail)
    • implement and test interactive version
    • design rules and evaluate interactively
    • design automatic system
    • implement automatic system
  • generalized web UI (Feb 5) Van, Dail, Arlin, Enrico
    • initial design and mockup (Arlin)
    • initial implementation (Van) March 2
      • extract names from source
      • tree retrieval
      • tree display
      • output features
    • build out features
      • name matching with user approve-delete
      • taxon sampling
      • gallery features
    • improve usability (about, FAQ, feedback, videos)
  • broad suite of tests (March 1)
    • black-box tests of accuracy
    • tests for edge cases, errors, exceptions and special conditions
    • tests for metadata and informative error messages
    • tests of speed and capacity load testing (jmeter)
    • regular uptime tests-- build up history on each registered service
  • mobile app (completed 17 Mar)
    • initial implementation in Android and iOS
      • accumulate list of species (incremental tree build)
    • initial team testing
      • capacity to distribute app
      • functional testing in museums and zoos
Add a custom footer
Add a custom sidebar

Clone this wiki locally