How To Create a Hello World Page with structr
Guide to creating a “Hello World” page with structr, which is a Neo4j-based CMS.
While looking at the guide, it occurred to me that most users are only going to create pages. I can’t imagine most sysadmins giving users the ability to create domains, sites or even templates. Users are going to author pages. And when their applications open up, it is going to be inside a domain/site and have only certain templates they can use, as part of a workflow to others.
Shouldn’t that be the same case for topic maps? That is the average user does not author a topic map, does not author default subjects or identifiers, probably doesn’t author identifiers at all. And for that matter, whatever they do author, is part of a workflow with others. Yes?
Has the problem, at least in part, been that topic map explanations explain too much? That for some specific domain, we say what to do and it simply works?
Take www.worldcat.org for example. A topic map authoring interface to that resource should allow users to select one or more entries, returned from a query, which all share the same ISBN, as being the same item.
For example, search for “Real World Haskell.” Six items are returned, with the first two obviously being the same title. The first entry has the following ISBN entry: 9780596514983 0596514980, the second entry has: 0596514980 : 9780596514983. That’s right. Addition of a colon separator and the ISBN numbers are reversed. Rather than two entries, a topic map should allow me to mark this as one entry and to process the underlying data to present it as such. Including all the libraries with holdings.
That should not require any more effort on my part than choosing these entries as identical items. Ideally that choice on my part should accrue to the benefit of any subsequent users searching for the same entry.
The third and fourth items are the same text but in Japanese. My personal modeling choice would be to merge them and the sixth item (the Safari edition) as language and media variants respectively. Might need language/media variant choices.
No thorny theoretical issues, immediate benefit to current and subsequent users. Is that a way forward?
Deduplication of the WorldCat files by automated means is certainly possible and a value-add. But given the number of users who already consult WorldCat on a daily basis, why not take advantage of their human expertise?