I appreciated the recent comment that made it clear I was conflating several things under “authoring.”
One of those things was the conceptual design topic map, another was the transformation or importing of data into a topic map.
A third one was the authoring of a topic map in the sense of using an editor, much like a writer using a typewriter.
Not to denigrate the other two aspects of authoring but I haven’t thought about them as long as the sense of writing a topic map.
Today I wanted to raise the issue of requirements for a authoring/writing tool for topic maps.
I appreciate the mention of Wandora, which is a very powerful topic map tool.
But Wandora has more features than a beginning topic map author will need.
An author could graduate to Wandora, but it makes a difficult starting place.
Here is my sketch of requirements for a topic map authoring/writing tool:
- Text entry (Unicode)
- Prompts/Guides for required/optional properties (subject identifier, subject locator or item identifier)
- Prompts/Guides for required/optional components (Think roles in an associations)
- Types (nice to have constrained to existing topic)
- Scope (nice to have constrained to be existing topic)
- Separation of topics, associations, occurrences (TMDM as legend)
- As little topic map lingo as possible
- Pre-defined topics
What requirements am I missing for a topic map authoring tool that is more helpful than a text editor but less complicated than TeX?
BTW, as I wrote this, it occurred to me to ask: How did you learn to write HTML?
The last time I seriously looked at Wandora for building a topic map, which was about 3 years ago, I think I had to rule it out for the project I had in mind due to the limitations of their implementation of scope, association roles, and reification. (see http://www.wandora.org/wiki/Reduced_Topic_Maps) I’m not sure if these limitations would be a problem for most beginners.
That said, even when I was a beginner, I wanted to do things with scope that authoring tools wouldn’t let me do. For example, my first topic map, which is the one I created for the class you taught in at UIUC-GSLIS in 2007, was written entirely by hand in XTM 1.0 because I also couldn’t do the things I wanted to do with scope in the OKS Ontopoly editor. (I was, however, able to view the things I did with scope in the Omnigator, which was nice.) It’s entirely possible that I was not a typical beginner in this respect.
Since then I have also tried to use Ontopoly to create other topic maps, but as soon as I want to do something interesting with scope, I get frustrated by the fact that the XTM exported by the OKS is not very human-editable. Personally, I would probably be much less frustrated with the limitations of existing authoring tools if they produced human-editable XTM. Or maybe if they had something like keyboard macros…. for example, writing associations in XTM would be much less painful if you could just use a macro to insert a template of the necessary XTM and then fill it in by hand. (Maybe what I really need is a good XML editor?)
I learned to write HTML from the old NCSA “Beginner’s Guide to HTML”, which doesn’t even exist on the web anymore. That was in 1995; the world of web page creation is very different now and it passed me by long ago. But it may explain why I opted to create my topic map by hand; I was already pretty comfortable typing lots of angle brackets.
This isn’t a very helpful answer, is it? =)
Comment by marijane — April 3, 2013 @ 8:03 pm
Wandora has improved a lot (not in its interface, unfortunately) over the years and a new version is due out on April 30, 2013. I will mark my calendar and try to do a review of it when it appears.
Have you tried to write associations in LTM or CTM? I wrote a cheatsheet for LTM, http://tm.durusau.net/?p=32712. I had the intention of writing one for CTM but searching my hard drive turns up nothing.
I haven’t looked at Ontopoly output in quite some time but possibly an XSLT stylesheet might help with its readability.
Actually it was a very helpful answer! Our experiences as users are always different from others and collecting enough of them highlights common issues or uncommon ones.
Comment by Patrick Durusau — April 4, 2013 @ 3:31 pm