Thanks to everyone for the suggestions. I'm inclined to think that <fw> 
is probably the right solution here; these page numbers also appear only 
on rectos, so modern numbering practice wouldn't match the original even 
if it were right.

It does strike me, though, that if we're making up a modern numbering 
system for the documents, which departs in a significant way from the 
original, it might as well be done with XPath at render time.


John Carlson wrote:
> For the purposes of processing, of course, a sequentially-ordered unique 
> identifier for each page break is useful - on several projects I've 
> worked on, this has been accomplished using @xml:id assigned a value 
> created from a four character identifier, an abbreviation for page 
> break, and a sequential number independent of anything printed/written 
> on the page (e.g., xml:id="DIST-pb-0209"). When the numbering is 
> regular, this alone would be enough to generate labels using automated 
> processes (XSLT with the xpath substring functions, etc.). For 
> leaves/pages with irregular numbering, one could use @n or @rend to 
> record the 'erroneous' label to be processed differently (XSLT: 
> <xsl:when test="@n">, etc.). In other words, only pages with 
> non-sequential or repeated numbering would need the extra attribute:
> <pb xml:id="DIST-pb-0288"/>
> ....
> <pb xml:id="DIST-pb-0289"/>
> ....
> <pb xml:id="DIST-pb-0290" n="209"/>
> ....
> I wouldn't think this is too unusual a situation nor always the result 
> of error - several kinds of books come to mind that might contain 
> repeating page labels (omnibus editions of multi-volume series, works 
> with two parts that are independently numbered, etc.).
> John
> On Fri, Jan 16, 2009 at 3:14 PM, Martin Holmes <[log in to unmask] 
> <mailto:[log in to unmask]>> wrote:
>     HI folks,
>     We have a set of printed books in which page-numbering is frequently
>     erratic; numbers are omitted and repeated, and sometimes the order
>     of digits in the page number is wrong.
>     I'd be glad to know how anyone else has handled marking up this
>     problem. One approach we took initially for digit-ordering was to do
>     this:
>     <pb n="146" rend="164" />
>     where what should have been 146 was printed as 164. But on another
>     volume, I've found that page numbers 18 and 19 were repeated,
>     meaning that everything subsequent to that is "wrong"; that's made
>     me reconsider what I mean by "wrong" in this context, and whether a
>     page number might just be better viewed as a label rather than a
>     necessarily unique identifier for a page.
>     Cheers,
>     Martin
>     -- 
>     Martin Holmes
>     University of Victoria Humanities Computing and Media Centre
>     ([log in to unmask] <mailto:[log in to unmask]>)
>     Half-Baked Software, Inc.
>     ([log in to unmask] <mailto:[log in to unmask]>)
>     [log in to unmask] <mailto:[log in to unmask]>

Martin Holmes
University of Victoria Humanities Computing and Media Centre
([log in to unmask])
Half-Baked Software, Inc.
([log in to unmask])
[log in to unmask]