Transliteration isn't a particularly useful operation - the 33-
character cyrillic alphabet doesn't map directly onto the 26-character
latin alphabet - and neither of them is reliably phonetic. English
uses something like 44 different phonemes (it varies from dialect to
dialect) and individual phonemes aren't always reperesented by the
same alphabetic characters.
'Googling' also produces the attached information (xpost to a.b.s.e.)
The problem applies to alpha-numeric ordering of electronic component
labels, so it's a problem on two levels. ABCD (order)...is (ABWG)
order, to begin with (or ABVG otherwise).
Am currently trying for a straight transliteration of the labels,
regardless of the loss of sense or order in the process. Actually
proofreading the output of a robot, without loosing the robot's
piquant disregard for meaning or purpose (if you can think of a more
inane activity). With typos out, a real brain can then be engaged to
Looking for working knowledge from a real human being, with experience
in this area. The manufacturers keep the cyrilic fonts for part
numbers in their translated web sites and literature. Is this the real
I'm thinking it would play hell with a microdiodoze spreadsheet, in
varying years of revision.