• Created a personography file.
• Entered two lists of people: editors and informants. The Informants list contains one <personGrp> for the dual informant "Y"; otherwise it's all <person>.
• Generated a new schema to support the <particDesc> element and to add @corresp to <bibl>. In this I was frustrated by two things: firstly Roma has a bug (which I've reported on SF) whereby including the corpus module but then excluding the majority of its elements results in the exclusion of <particDesc> even when <particDesc> is not explicitly excluded, so I was forced to add the entirety of the corpus module. Secondly, oXygen has an annoying habit of deciding to validate files using tei_all instead of the explicitly-linked schema, which means you often don't realize your schema is screwed for ages.
• Ran some XSLT with today's regex of the day (constructed with an XPath concat()):

  concat('(^', ., '[\d]+)|([\s]+', ., '[\d]+)|([\s]+', ., '[\s]*$)|(^', ., ';)|([\s]+', ., ';)|(^[\s]*', ., '[\s]*$)')
  

  to formalize all references to people in <bibl> (through added @corresp), @who and @resp so they use a psn: pseudo-protocol which we will map to the personography file.

Next stage: go through SMK's markup documentation and integrate its key info into the formal documentation, and at the same time explain the use of the pseudo-protocol, to bring our docs up to date.