As we promote the use of markup languages for representing the entire range of digital data, reusing a common format requires adaptable access patterns on the wire. Different network users may emphasize early access to parts of the structure, or only require access to parts of the structure, or perhaps just the structure and not the contents. Time affects the representation of markup instances in three ways: bandwidth, latency, and throughput. We can envision optimizing access along each of these axes by compression, reordering, and tokenization, respectively. YML is a hypothetical "wire protocol" that optimizes access to parsed markup instances in the hope of forestalling fragmentation into "compiled" archival formats.
We have been discussing YML since early 1998 on the FoRK mailing list, but we haven't really written any papers about it yet. It (as well as the concept of XML FORMs) is just a thought exercise for now, but if you'd like to know what we know, check out the following FoRKposts:
Adam Rifkin, http://www.ifindkarma.com/attic/
PhD-Related Documents, Caltech Infospheres Project
Last modified: Tue Jun 16 18:59:27 PDT 1998