The number in the URL obviously relate in some way to the enumeration - but not in a simple fashion. I checked a few more (9L1, 9L2, etc) and they are all on the same or adjoining page to the Sullivan family. But they are definitely not listed in order 9L1, 9L2, 9L3 etc. The could be using another coding system such as a Gray code to count through the entries (easy to implement, but not so easy for users to decipher). It could be possible to decode the system if enough data was collected in order to see the pattern.
Just looking at the page in question, the end of the previous page and the start of the next page:
922 Catherine Crowley 9
92L Mary Crowley 6
92G Margaret Crowley 3
92P David Crowley 1
925 Michael Regan 40
92R p48 end Margaret Regan 35
92T Null
92Y Null
92B p49 start Bridget Regan 18
921 Daniel Cremmins 40
9LM Mary Cremmins 40
9L9 John Daley 28
9LS Bridget Daley 25
9L3 Matthew Daley 3
9LQ William Daley 2
9L7 Catherine Hyde 33
9LW Nicholas Hyde
9L4 William Brian, 30
9LH Mary Brian 30
9LC Denis Sullivan 24
9LZ Mary Ann Sullivan 18
9L8 Daniel Sullivan 30 brother
9LD Betzy Sullivan 35 sister in law
9L6 Julia Ann Sullivan 8
9LX Bartholomew Sullivan 6
9LF Robert Sullivan 2
9LN Michael Regan 38
9LJ Catherine Regan 40
9LV Mary Regan 16
9LK William Sullivan, h, mar, 40, Rope Dealer, b. Ireland
9L2 Ellen, wife, 40, b. Ireland
9LL Jeremiah, son, 5, Scholar, b. Glam, Cardiff
9LG p49 end Ellen, dau, 6 mths, b. Glam, Cardiff
9LP Null
9L5 Null
9LR p50 start Timothy Morrison, lodger, unm, 24, Shipwright, b. Ireland
9LT Null
9LY Thomas Mahoney 33
9LB Mary Mahoney 26
9L1 Patrick Mahoney 2
0 (zero) and the vowels A, E, I, O, U and lower case letters return error messages.
This shows that 9L does not relate to a household or even a page - it's just part of the counting system used.
The sequence appears to repeat - though the 'null's, particularly the 'T' needs a little more confirmatory work.
Pages appear to be separated by two 'nulls', empty properties return a 'null'. Presumably blank lines will also return a 'null' and there may well be some other way (eg three nulls) to indicate a break between enumeration districts.
The possible sequence = M9S3Q7W4HCZ8D6XFNIVK2LPG5RTYB1 them M again.
Note that the same sequence appears to work on each 'digit' of the alphanumeric 'number' - 92 is followed by 9L as 9L2 is followed by 9LL. So 9L should be followed by 9P etc up to 91 then SM etc. Feel free to test this theory ...
This sequence may well be a standard counting system of some sort (definitely not a Gray code) - if anyone recognises it I'd be interested to know.
This might help users of familysearch to see the order of the enumeration - but not the important breaks between households.
