Karen, you reasoning and examples is of great help. I'll discuss this with the developers of our interrelated projects. I understand that there's no such thing as "single truth", and I appreciate any efforts to clarify best practices in the online documentation. Thanks, Matthias On 11-Apr-07 at 17:39 -0700 Karen Coyle wrote: > Matthias Steffens wrote: > > > >I see your point. However, I thought that the recommended way of > >encoding total number of pages in MODS was: > > > > <extent unit="pages"> > > <total>402</total> > > </extent> > > > OK, now I've gone back and looked at the documentation, and there isn't > anything called "number" -- just start, end, total, list. Number is in > the "detail" field, and to me, using "detail" here rather than "extent" > doesn't make much sense. Detail appears to be for the enumeration of > serials. Enumeration is different from extent. I could imagine "detail" > being used for chapter number or names, but then you'd use extent to > give the pages included in the chapter. I can't explain why that makes > sense to me, but it does. > > Using "extent" would mean that an article on just one page has to be in > "start"; "end" with the same number confirms it's only one page. "Start" > with only one number really means the starting page. And I have seen > citations that use the starting page and the number of pages (used > mainly for ILL requests), so you could have: > > <extent unit="pages"> > <start>37</start> > <total>5</total> > </extent> > > This means that you could also do: > > <extent unit="pages"> > <start>37</start> > <total>1</total> > </extent> > > Rather than: > > <extent unit="pages"> > <start>37</start> > <end>37</end> > </extent> > > You could also have "start" alone because no other information was > given, and you don't know how long it is or where it ends. > >Please correct me if this is wrong. Thanks again, > > > > Matthias > > > You may have already guessed that there are only varying degrees of > right in this activity. ;-) It *would* be great to clarify the best > practices for some of these cases that we can predict. And there will > always be the unpredictable exceptions.