- From: Joseph M. Reagle Jr. <reagle@w3.org>
- Date: Fri, 15 Oct 1999 10:46:05 -0400
- To: "IETF/W3C XML-DSig WG" <w3c-ietf-xmldsig@w3.org>
http://www.w3.org/DesignIssues/Fragment.html Tim Berners-Lee Date: April, 1997 Status: personal view, but believed to be my best expression of the underlying architecture for W3C development. Editing status: Good enough fo discussion. [1]Up to Design Issues Axioms of Web Architecture: 1a _________________________________________________________________ Fragment Identifiers on URIs The fragment identifier is a string at the end of a URI which identifies, within a Web document, a part or view to which one refers. For example in the object http://foo/bar#frag the string "frag" is the fragment identifier. (Depending on where you look, the URI is considered to include the fragment identifier, or to have the fragment identifier appended to it. This is important for the BNF, but in practice you will find people using the terms URI and URL loosely to things which do or do not include a possible fragment identifier. Formally, the URI DOES include the fragment ID) In practice, you can divide the processing which occurs when following a link using HTTP into three steps: 1. The client figures out which server to contact by parsing part of the URL, and sends the URL as a request to the server; 2. The server figures out which object is referred to by parsing the rest of the URL, and returns some rendition of it to the client; 3. The client presents all or part of the object to the user The last part typically involves finding some software class which can handle the given MIME type, and passing it the data stream. At the same time, the fragment identifier is passed as an extra function. For HTML, the fragment ID is and SGML ID of an element within the HTML object. Axiom The significance of the fragment identifier is a function of the MIME type of the object This means that the fragment id is opaque for the rest of the client code. The HTTP engine cannot make any assumptions about it. The server is not even given it. It also means that for any new data type one can be creative about using the fragment ID in a relevant way. For example, for a 3D object the fragment ID could give a viewport. For a music object, the Fragment ID could give a section in time, or a set of parts, or it could include a suggested tempo. For future versions of HTML, the fragment ID could be made more powerful to include a range or "ladder" reference to a part or parts of the SGML element tree by position. A very useful fragment ID for plain text would allow ranges to be quoted by line and character number These things are all decisions made when the MIME type is defined. Therefore, The fragment ID spec for a new MIME type should be part of the MIME type registration process. Different MIME types then can have different fragment ID specifications. When HTTP for example negotiates between different content types, it is clearly useful for those types to have a consistent (hopefully identical) fragment ID syntax and semantics. Fragment IDs and Content negotiation If content negotiation occurs across types which do NOT share a fragment ID specification, then rigidly there has been an error. In practice, HTML is the only type (1997) which allows fragment IDs anyway, and other types ignore it. Also, as falling back from a pointer to a specific view to a pointer to the whole document has been considered effective fallback procedure, so no harm has been done. In the future, metadata returned or warnings returned should indicate to the client that this could be a problem. Also, in new access protocols, the fragment ID requested could be shipped to the server as a hint, which would allow the server and client to negotiate and if successful arrange for the fragment ID to be converted to a suitable equivalent value for an alternative MIME type. User awareness of the form of a reference Clearly when a fragment ID is generated and associated with a URI which is generic in any way (language, version, etc as well as content-type), then there is a possible failure of the fragment-id refers to something which is not defined in any specific instance. It would be appropriate for a client, when generating a link (or bookmark, etc) to provide the user with a choice of * A bookmark to the whole living document, or * A bookmark to a specific part of a "dead" version; * Intermediate combinations. As both these options are meaningful and useful, they will have to surface at the user interface level. _________________________________________________________________ [2]Next: Links and the law [3]Up to Design Issues $Id: Fragment.html,v 1.6 1998/03/04 17:24:58 timbl Exp $ References 1. http://www.w3.org/DesignIssues/Overview.html 2. http://www.w3.org/DesignIssues/LinkLaw.html 3. http://www.w3.org/DesignIssues/Overview.html _________________________________________________________ Joseph Reagle Jr. Policy Analyst mailto:reagle@w3.org XML-Signature Co-Chair http://w3.org/People/Reagle/
Received on Friday, 15 October 1999 10:46:07 UTC