UW Mangrove Project
The calendar unifies previously scattered information regarding colloquias, "590s" (research reading groups), department events, and courses.
Basic information about all these events is automatically extracted from existing web pages, for instance the university course time schedule page.
In addition, anyone in the community may add information to the calendar in order to:
Add a new event to the calendar. See how to annotate.
Augment existing information. For instance, by default the 590's on the calendar have just basic meeting time information. To improve this so that the calendar also displays each week's topic and presenter, annotate the web page that already exists for that 590.
The Who's Who dynamically compiles general, contact, and personal information about members of the community. This service was initially seeded with information from a previous, manually compiled Who's Who of graduate students and a department summary of some graduate student contact info. To improve your Who's Who entry (for instance, to add a picture or correct stale information), annotate your personal home page.
The search service provides a simple web-search-like interface to search for useful information within the community. We offer two distinct types of search:
Tag-only search. Extract tag information from any annotated web page and display the results in a convenient form. For instance, the query "<grad Student> <name>? <office>?"
returns a list of all annotated graduate students with their name and office.
Combined tag and text search. Users cannot be expected to annotate every item of data on a web page, or they may have only initially added a few semantic tags to their page. Thus, we provide a search service that can looks for both textual and semantic information within a single page. For instance, to find pictures of all of the assistant professors in our department, even if "assistant professor" is not semantically tagged, use the search "Assistant Professor <facultyMember> <portrait>?"
See Search for more information about the format of a search query.
Our Research Publication Database compiles a searchable database of
publications produced by members of our department based on the
information in home pages and project pages. This service is able to
avoid repeated entries by detecting duplicate publications, and is
able to infer missing information (e.g., an author of a paper) from
context (e.g., the paper was found on the author's home page).
To have your publications included, simply annotate your page with <publication> tags.
You may optionally add further tags to specify the year, authors, etc. -- these may improve the classification of your publication but are not necessary to included in the database.
Click on the heading above to finding out about automating common tasks via semantic email.