Roadmap

This is the monthly plan for growing the project and helping execution towards growth of the project, which hopefully means increasing its usefulness!

• Sign up for a task by placing ~~~~ at the end.
  • This is an example task Jon Phillips 03:30, 14 June 2011 (UTC)
• When a task is completed, add <strike>Some Task</strike> to show its completed.
  • This is an example closed out task Jon Phillips 03:30, 14 June 2011 (UTC)

We can add features all day, but the projects weakness is not enough contributors and papers. Thus, we will error on the side of getting more contributors and papers first.

2011.12

• Move site to new infrastructure Jon Phillips 04:27, 14 June 2011 (UTC)
• add ads to the site (mlinksva + fabricatorz)

2012.01

• send emails to top schools departments, followup as humans (pete, spencer, greg-g, rejon)

• Clean up search interface (christopher)
  • design (christopher)
  • implementation (christopher)
  • The lists of Journals, Authors, and Paper Titles should be made easier to read/scan (perhaps by breaking into clear columns)
  • The instructions should be state-aware (e.g. "Get fewer results" or "Get more results")
  • The search box could have auto-complete based on the meta-data

• AcaWiki Announce list archives are down - http://lists.ibiblio.org/pipermail/acawiki-announce/
  • Opened a ticket with ibiblio Christopher Adams 12:58, 1 July 2011 (UTC)
  • Announce archives are now visible but don't go back to the beginning of the list.

• Develop our own metrics and expose on site on the design (cube)
  • number of contributors
  • number of users
  • number of edits
  • length of summaries
  • number of linked to summaries
  • number of views of a summary
  • max # of contributors to a summary

• Handle pages with blank summaries (spencer)
  • An assessment system does not resolve all problem associated with BLANK summaries. In that case:
  1. blank summaries should not be prominent in search engine results
  2. blank summaries should not appear in AcaWiki's browse summary (at least not as blue links--the metadata script can fix that)

• Implement standard for Sentence case titles (greg-g)
  • AW:SC

• Fields for journal volume and issue are reported missing from the "Post a Summary" form (who will do this?) (spencer)

• consider mechanical turk solutions to increase yield (thinking again of researchers to work with) Jon Phillips 04:55, 28 July 2011 (UTC)
  • http://projects.csail.mit.edu/soylent/
  • http://groups.csail.mit.edu/uid/turkit/
  • what researchers are working on text summary?

• Classification system - a system of summary assessment as far as completeness, accuracy, quality
  • Cf. Creative Commons
    • Creative Commons Case Studies Evaluation
    • Case study of a simple but highly effective use of Semantic MediaWiki on the CC Wiki
  • Cf. WikiProject Council/Assessment FAQ
  • Employ the metadata script to displays the quality assessment of every summary

• Quality control lists for AcaWiki content (greg-g)
  • We need lists of Articles pages which do not have:
    • URL
    • DOI
    • Subject
    • tags
    • summary text
  • Some of these are valid (not everything has a DOI or URL), but creating lists like that would help with maintenance.

• Get data import from Fin, see email thread. Jodi is hooking up, Fab team will import (Jon Phillips)
  • http://neuro.imm.dtu.dk/wiki/Main_Page

• Read over and assess AcaWiki:Related_people_and_groups
  • Move relevant leads to the roadmap and start working on
• complete questions here
  • http://en.wikipedia.org/wiki/Functional_Requirements_for_Bibliographic_Records

• Look at sharing content with Journalist's Resource - http://journalistsresource.org/
  • http://www.timeshighereducation.co.uk/story.asp?storycode=416847
• investigate import of bibliographic sources of academic papers, reality assessment
  • open library?
  • http://www.readwriteweb.com/archives/33_gb_of_scientific_papers_-_and_a_manifesto_-_upl.php

2012.02

• investigate machine summaries of academic papers (rejon)

2012.03

• Improve LaTeX support, needed for any serious effort in mathematics and related fields. Contact for advice, Michael Kohlhase at Bremen (who will do this?)

His group has state-of-the-art tools for LaTeX to XML/HTML support and are the current maintainers/developers of Bruce Miller's LaTeXML

2012.04

Metadata improvement

Some of these make sense to collaborate on with other open content communities.

• Normalize conferences

Conference venues in CS (where they dominate journals as an organizational structure for the literature) have been indexed extensively by DBLP http://www.informatik.uni-trier.de/~ley/db/ and Microsoft Academic Search http://academic.research.microsoft.com/?SearchDomain=2 Look at the data dumps and APIs provided by these services, and adopt whatever normalization will allow users to frictionlessly import data from such sources. Better still, do a bulk harvesting from these and other sources and support an API to query for basic venue information which could be used to provide javascript dropdown for auto-completion when user starts to type in a venue name.

• Normalize journals

Basic information should be normalized (full name, common abbrevs, issn, publisher, date range, scope description, urls ). There are a number of more or less open aggregations already available which could be pooled for this purpose. I want to see creation and maintenance of an authoritative open index of journals, available with a RESTful API. The number of partners who have told me "sure we can do that, lets just do it" and then been unable to deliver has incremented by about 1 per year for the last 10 years. Why is this so hard? There are only a few hundred K journals. Cant we crowdsource the maintenance of basic information about them?

Past

• Roadmap 2011