Wikipedia’s automatic citation generator relies on Citoid to extract citation metadata from web sources. However, these metadata are sometimes incorrect or incomplete.
Until recently, the only way to fix this was by writing or fixing a Zotero translator. However, this requires JavaScript knowledge, and must wait for approval from the Zotero community.
Web2Cit is a tool under development to collaboratively define procedures to extract basic citation metadata from web sources, requiring much less technical skills than those needed to write a Zotero translator.
In this session for Web2Cit early adopters we will show you how to use the current features of this tool to improve metadata extraction from the web sources that you or your community usually work with (T308452).
- When?: Friday May 20th, at 4 PM UTC.
- Where?: "Localization + Small Wikis" room
- Username for contact: @diegodlh & @Scann
- Session Duration: 50min
- Language of session: English, but you may use Web2Cit with web sources in any language!
- Prerequisites: Knowing some XPath and regular expressions may be useful (but not mandatory).
- Good for newcomers?: Yes!
Remember bringing a web source you'd like to work with (preferably a local newspaper), or pick one from our list of problematic URLs (make sure that it's still not working OK). Note that Web2Cit does not support PDF files.
You may also check our introduction guide beforehand!
I know some JavaScript. Is this session for me?
If you know some JavaScript, you may prefer writing or fixing a Zotero translator. JavaScript translators are likely to be more robust than procedures collaboratively defined with Web2Cit.
However, writing a translator may take some time (even for skilled JavaScript programmers) and require approval from the Zotero community. So you may still find Web2Cit useful!