Hi Mark and Brian,

Thanks for your replies. I still feel that it should be OK to have <daogrp><daoloc><daodesc> in that order, because the stylesheet can display these in any way that we wish. I guess if we'd realised that the order was going to matter we would have developed our new template differently, but now its going to be more work to enforce this order, but for no apparent benefit other that this is what the DTD asks for. Oh well...



Hello, Jane

Order in XML is central feature, though specific ordering may be specified as irrelevant. Think of sibling paragraphs in a text. If the order changes, the sense of the text is lost.

For the element daogrp, the DTD specifies zero or one daodesc elements followed by one or more (in any order) of daoloc, ptrloc ...
So if you had a daogrp with no daodesc but only, say, refloc and arc, then the order would not matter. But since you do have a daodesc, it is constrained to preceeding the other elements. Now, why the DTD specifies that, I don't know.

I hope that clarifies somewhat.


> -----Original Message-----
> From: Encoded Archival Description List [mailto:[log in to unmask]] 
> On Behalf Of Mark Carlson
> Sent: 17 January 2009 00:54
> To: [log in to unmask]
> Subject: Re: Order of daogrp elements
> I'm afraid that this is just one of the restraints of EAD.  
> There are many instances where order matters.  For instance, 
> c02 is a child of c01, but it can't be placed as a direct 
> child of c01, it must come after the <did> and then any of 
> the specific sub-elements if they are used.
> One nice thing about XMetaL, is that you can create "Mini 
> Templates" so that when you double click on <daogrp> it will 
> always populate the child elements in the order that you 
> specify in the Mini Template that you created.  This removes 
> much of the guess work.  This is also useful as EAD allows 
> many many encoding practices that may not be desirable.  For 
> instance, the ability to embed many subelements into 
> infinitum. This is one of the reasons that many repositories 
> don't edit directly in XML but use some other "front end" and 
> then convert that to EAD XML.  There are pluses and minuses 
> on both sides.
> Mark
> Jane Stevenson wrote:
> > Hi all,
> >
> > I was wondering if anyone has come across this issue: We 
> are introducing <daogrp> and within that <daoloc> and 
> <daodesc>. However, the EAD description won't validate if 
> they are in that order, i.e. we have to enter <daodesc> first 
> of all, before <daoloc>. I cant' find any information on why 
> this is the case, and its quite difficult for us to impose 
> this order in our Data Creation Template.
> >
> > The DTD entry is:
> > <!ELEMENT daogrp (daodesc?, (daoloc | ptrloc | extptrloc | refloc | 
> > extrefloc | arc | resource)+)>
> >
> > My knowledge here is not great, but I believe this does 
> mean that <daodesc> must come first. But I'm really not sure 
> why. It seems to me that one of the features of xml is that 
> the order of sibling elements is irrelevant, unless there is 
> a very good reason otherwise.
> >
> > Cheers,
> > Jane.
> >
> > Jane Stevenson
> > Archives Hub Co-ordinator
> > Mimas
> > The University of Manchester
> > Kilburn Building
> > Oxford Road
> > Manchester M13 9PL
> > [log in to unmask]
> > +44(0)161 275 6055
> >