 |
 |
Paul Miller wrote: > As we saw it, there are a number of things that might be done, none of which > is perfect for various reasons. These are; > > (i) - change the recommendation in [2], such that XML/RDF > implementations label elements with a leading capital letter. > > The obvious problem here is that this approach is directly at odds with > emerging practice in the wider XML/RDF community. > > (ii) - change the recommendation in [1] - and retrospectively convert the > existing installed base of HTML - so that elements are labelled solely > in lower case. > > It's unlikely that all those implementors out there would be happy > to go back and change all their existing metadata. > > (iii) - recognise that case doesn't matter in HTML, so leave the existing > installed base alone. Change the recommendation in [1] so that new > implementors use lower case element names. > > (iv) - recognise that the HTML installed base uses a leading capital > letter. > Recognise that XML/RDF uses all lower case letters. To allow > conversion between the two, declare the Dublin Core element names > to be case-insensitive, and leave it to HTML/XML tools to sort it out > according to whether the element is being used in XML or HTML. > > (v) - Do nothing. > > What do you all think? (I *know* which one I prefer, but then I'm biased) I don't really see the difference between (ii) and (iii). I mean -- we can decide what case we specify in the drafts -- but if we opt to change [1] to recommend lower case -- but we can't effect any change on the installed base. It will be up to those sitting on HTML-encoded metadata to decide whether they want to convert their data to reflect the change i [1]. Anyway: My vote is for (iii) and I'll just as well vote for (ii). Rationale: ---------- The installed base of DC metadata is still very small. Alternative (iii) will not create any immediate problems for the few implementers who has relied on the case recommendation in Kunze's draft. Also, since the draft is still unpublished outside these lists -- anyone who has seen the draft must be a member of the active community and should be well informed about the situation because of the very discussion taking place here and now. This mimimizes any confusion that may arise from choosing alternative (iii). The real problem will come when the Internet community start migrating from HTML to XML. At that point in time, there _may_ be some comfusion among the small number that already has marked up their collection relying on Kunze's draft -- but unless the draft is changed to recommend the use of lower case in HTML, there will be a much larger confusion (because we will have a much larger HTML installed base with much looser connection to the DC community after the DC in HTML is becomes available to the general public). This rules out alternative (v) as far as I am concerned. Alternatives (i) and (iv) will make us _very_ unpopular in the XML community (and deservedly so) -- as soon as you've decided to use Unicode as your character set, there is no such thing as case insensitivity. References ---------- > [1] - http://www.mailbase.ac.uk/lists/dc-general/1999-05/0063.html > [2] - http://www.ukoln.ac.uk/metadata/resources/dc/datamodel/WD-dc-rdf/ -- - gisle hannemyr ( [log in to unmask] - http://home.sol.no/home/gisle/ ) ------------------------------------------------------------------------ "Use the Source, Luke. Use the Source." -- apologies to Obi-Wan Kenobi ------------------------------------------------------------------------ %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%