- From: Timothy Holborn <timothy.holborn@gmail.com>
- Date: Mon, 2 Feb 2015 20:50:15 +1100
- To: public-rww <public-rww@w3.org>
- Message-ID: <CAM1Sok0Sxzrbw0KHMrXsF_2DmxisaK5pMMZLHnwrxrxuGjb=HA@mail.gmail.com>
Hi all, I woke-up the other day, dreaming about semantic bookmarks... I found an old project [1] that seems relatively inactive; and, aspects of the concept were not contained therein; My other thoughts were in consideration of Http Signatures, and credentials [2]. Essentially, it's a small document containing RDF rather than simply a file that stores a link. To make it work, the definition would need to be addressed; followed by application support (browsers or browser plugins, etc.). Below is the text i wrote about the concept... I'm not sure how much time i've got to focus on this concept. Would be happy to collaborate, of course... Tim.H. Semantic BookMarks (Or; Semantic Library File, or semantic link/object file (?)) In this document the word RDF is interchangeable with Linked-Data, Semantic and ‘structured data’. CONCEPT The concept is to create a file format for bookmarks that uses RDF to create a set of resources about a web place or thing. The bookmark acts both as a classical bookmark, as well as providing a gateway to data-services listed in the bookmark file itself. Semantic links may have numerous varieties from client only, to symmetrical, signed services. A Semantic Bookmark can be used by an account as a services resource, enabling data-services to be supported between the bookmarked location (or related service) and the user-environment (browser, account, device/machine, etc.) The semantic bookmark would provide RDF describing the resource the bookmark has been created about. This may include a description and/or use-information about the services and other structured data relating to the resource. The concept follows the idea of a bookmark in the way that it is a single downloadable or linkable file. The concept also follows the idea of a library file, as the file itself can store an array of linked-data (triples), as RDF, about the service related to the resource. Therein; the bookmark does not simply provide a ‘shortcut’ file enabling access to the site. A Semantic Bookmark is a file that includes resources relating to the location and a description of the resources available in relation to that location, which in-turn may be used by other programs to navigate data-services resources available for the user, identified via notations in the semantic bookmark. CONSIDERATIONS / ASPECTS Linked-Data enables a web of information to be networked together on web-scale. Overtime the amount of structure data is envisaged to expand exponentially. As this occurs search functions will change dramatically, and an array of new functionality will become available more broadly, than improvements to search, through related technologies. By creating RDF bookmarks / Library shortcuts, it is envisaged that web-computing improvements can be achieved, including (but not exclusive to) search, browser functionality, data-services and memberships. FUNCTIONALITY CONCEPTS Improve Desktop Search - Create categories - Create ‘linked files’ - ‘Link’ visited websites, establish search within sites that have been bookmarked Improve ‘auth’ - Enable access to online / cloud resources where user has authenticated to portfolio of existing bookmarks (containing AUTH info) Semantic Experiences - Provide Modules via ‘bookmarks’ to compatible applications. - Bookmarks can be searched for information stored in relation to the bookmark (ie: websites / web-databases, etc.) - Bookmarks can be used to insert content - Bookmarks can link to browser /desktop (web) apps - Bookmarks can relate to schemas for import/export of structured data to/from address/systems. (ie: bookmark wiki article, import into EPUB3, CSV, etc.) - Archive - should provide capacity to store historical versions as well as most recent version, where versioning applies… - Ability to convert WebHistory and/or find SemanticBookmarks from WebHistory. - RDF should include the ability to categorise the bookmark files - Bookmark should include capacity to include Indexes - Bookmark should include capacity to provide credential for private access (ie: AUTH - Password/userid) - Bookmark should be capable of obtaining data (ie: foaf file) USE CASE EXAMPLES (not exhaustive) Dictionary.com bookmark enables via RDF inclusion to desktop dictionary language file. Address book Bookmark file = access to users Profile Document (ie: FOAF or vCARD) Store access credentials in relation to website location A wikipedia page shortcut might be used to support extraction of the data (and references). A short-cut to an image might reference the source, copyright info, license info, etc. CLASSES Different classes of 'semantic bookmarks' might be defined as part of the ontology design. A-Symmetrical A Client-side document Symmetrical Bookmark A client side document that can update from server side Signed / unsigned Using HTTP signatures, a bookmark file can be signed. (must be symmetrical form though?) PingBack The capacity to send info, identify if update is available, track usage, etc. NOTES: [1] http://www.w3.org/2003/07/Annotea/BookmarkSchema-20030707 [2] http://opencreds.org/
Received on Monday, 2 February 2015 09:50:44 UTC