Ideas and proposals for realizing WikiCite and related projects. These instructions and schedule are speculative. Anyone can post here.
Instructions: what you can do with WikiCite
Query citation data in Wikidata
Seeing the results of queries about citations in Wikidata is the usual reason why new people become excited about the WikiCite project. Popular queries include the following:
- Show all the publications by a single person, generating a scholarly profile for that individual
- Show all the publications at a university, showing many things including what sorts of papers at the university are most cited
- Show all the publications on a certain topic, showing which papers are most cited, the popularity over time of publishing on the topic, and who is prominent in the field
There are two ways for beginners to do queries. One way is to use a Wikidata front end which generates queries using WikiCite information. Typically front ends are easy enough to use that a new user can access interesting information in 2 minutes.
The other beginner starting point to use the default and native Wikidata Query Service. This approach is powerful enough to permit any query but also displays a dual input interface of human-readable query requests and a translation of that into SPARQL. Different new users have mixed results with this; some queries are easy to do in 1 minute; other queries fail; and other users become comfortable only after hours of practice. The Wikidata Query Service is the recommended path for beginners who intend to engage regularly with Wikidata.
Manually edit the Wikidata item for a publication
Learning how to manually edit Wikidata is necessary background knowledge for understanding how to automate information management with WikiCite.
see the data model
Learn enough to be able to discuss WikiCite
WikiCite participants wish for any interested persons to have meaningful conversations about the benefits of open citations and WikiCite's role in making open citations available to everyone. The following publications and videos are popular beginner introductions for anyone who wants to become informed enough to join conversations about open citations in general and WikiCite particularly.
Learn about Wikidata and Wikimedia projects in general
Participants in WikiCite typically support Wikimedia projects including Wikidata and Wikipedia. These projects share a common philosophy of advocating for open content, a public commons of information, civil collaboration, global participation, and a sincere desire to improve everyone's quality of life through knowledge sharing. Having a working understanding of Wikimedia projects is useful context for understanding the nature of WikiCite.
To learn more about Wikidata consider the following:
To learn more about other Wikimedia projects then consider the following:
Experiment with batch editing tools
WiDaR is a collection of tools for making many edits to Wikidata quickly. All of them are experimental but it is helpful to know they exist, and in the future these or updated tools will offer these functions.
Mix'N'Match is a game which makes it easy to disambiguate Wikidata items to items in external databases.
Curate and suggest datasets for others to ingest
Ingest a dataset
Quickstatements is a tool which allows a person to ingest a dataset into Wikidata. A user who is comfortable with Wikidata's data model, and who has experience integrating some data manually, can scale up their engagement by presenting a CSV file or Google Sheet link to Quickstatements.
There is no formal schedule of content/software development as WikiCite - like Wikimedia in general - is a crowdsourced campaign built by volunteers. The 2019 timeline below began a guess of one person, Bluerasberry.
Wikicite v1 (2005), pre-Wikidata (Talk)
Wikicite v2 (2006) (Talk).
Wikicite v3 (2010) (Talk)
Overviews: Meta page | Wikidata page & roadmap | Website: wikicite.org
Events: Conference series (2017-202x)
Development: Microgrant series (2018-2021) | Shared citations proposal
Shared citations (sister project + core MediaWiki): WikiCite/Shared Citations
Potential timeline (bluerasberry)
By end of 2019
- Ingest easy-to-get citations
- 60% of PubMed
- 10% of CrossRef
- Whatever anyone shares ad hoc
- 10% author disambiguation
- Match with ORCID
- Match with a few university partners
- Develop front end visualizations
- Establish high familiarity in the Wikimedia community
- 10% of self-described highly engaged Wikimedians know of Wikicite
- Public discussions about relationship between WikiCite content and other Wikimedia projects outside of Wikidata
- Anticipate public outreach beyond people who care about citations
By end of 2020
- increase content volume, taking it where it exists
- Begin to repeat process with harder-to-get citations
- Establish data models for cataloging other academic sources beyond journal citations
- Videos of lectures
- Suggest data model to other projects managing publication
- All media in any context?
- prepare to replicate institutional engagement
- document Wikibase
- dockerized deployment for local WikiCite deployment
- produce marketing materials
- internal management of finance
- self-published marketing
- establish eligibility for funding from government and foundations
- model Wikibase as a possible commercial sector
- compare to Internet Archive's Archive-It service
- compare to the commercial MediaWiki environment
- develop product comparisons with other consumer products
- pilot diversity efforts
- secure funding for paid content translation
- develop realistic documentation about the relationship between wiki volunteers and basic language infrastructure
- layout market research re: target language and and regional communities, particularly against WMF outreach investment plans
- execution some translation to make WikiCite engagement and further development possible with reasonable effort
- model examples in each identified field of academia and media
- cover all departments in universities, so that every field has a space
- cover every sort of media, so that for any publication there is an example
- plan for some regional coverage
- some sample collections in English language targeting topical interests in every country
- some sample collections in mixed languages covering some organizations in every country
- Wikimedia platform integration
- model modes of inter-wiki integration
- Links from Wikipedias to WikiCite profiles, for example, scientists to their bibliographies
- share searches which provide data about citation usage
- model outreach which encourages citation usages as a measurable success metric
- enable pulls from Wikidata to other projects
- pilot Wikimedia outreach
- move slowly and cautiously...
- ...but get some experiments in plan in as many Wikimedia projects and languages as possible...
- ...anticipating a multi-year discussion which needs to start now
- Design and test a Shared Citations database (see detailed timeline plan) ~> needs updating, as not immediately implemented by WMF
- Implement a working beta version of a Shared Citations database (see detailed timeline plan)
Used across the wikis, widely reused off-wiki, citoid helps generate structured citations from URLs, DOIs, and other strings; and supports their reuse within a single page.
- Currently Citoid relies on parsers for a range of strings, and can be extended w/ additional parsers?
- Reuse: If something like Shared Citations existed, citoid would be the primary conduit for reuse across pages.
- Most citations are generated using a number of high-traffic templates, such as Cite Q.
- Some projects focus on improving adoption, translation, and use of the major templates
Some article- and book-metadata projects focus on gathering metadata about sources, both through Citoid-style parsing of target URLs, or more detailed extraction from the documents they refer to.
- Some of these try to better integrate existing repositories of metadata, like Zotero. These include Cita and Web2Cit.
- Source-assessment projects focus on gathering data about the reliability of source domains and articles, or the topical relevance of sources.
- Some of these gather data in wikitables, like Perennial sources. Ex: CiteUnseen
- Others also incorporate data from external reviewers, like NewsQ. Ex: the Sourcery underlay
This includes bots for ingesting and updating citations, deduplicating existing citations, and linking cites to wikidata records and Wayback Machine snapshots.