Re: Dynamic vs. Fixed World Views was Re: MARCXML to Topic Maps? MODS to Topic Maps?

Patrick Durusau <patrick-Q/[email protected]>
Newsgroups gmane.text.xml.xtm.general
Message-ID <[email protected]>
Lars,

On 7/17/2010 10:49 PM, Lars Heuer wrote:
> Hi Patrick,
>
> [...]
>    
>> That is we recognize some strings in particular context, say an author
>> field, with some variations, to all identify the same singer and poet.
>>      
> Well, do we talk about a TMDM mapping or TMRM mapping?  TMRM gives us
> some (or a lot of) freedom, but how do you map strings to subjects in
> TMDM?
>
>    
Well, given that the TMDM is the only standardized legend I would 
suggest starting with it. Not to mention that it may be an easier 
stepping stone into topic maps and more widely supported.

*Mapping strings to subject in TMDM*

That is part of topic map authoring/creation that has never been (IMHO) 
adequately addressed in various writings about topic maps. We tend to 
start with topics but never say where the hell those come from. Creation 
ex nihilo I suppose. ;-)

It would be more accurate to say "subject recognition" which is then 
mapped to a topic (or occurrence, etc.).

For example, assume that one of the fields has the string, "ex nihilo" 
in it.

My "subject recognizer" software has *already* recognized that this 
record has a classification of theology, so it "knows" (in the SW sense) 
this is not about "ex nihilo," the sketch site 
(http://exnihilo.mezzoblue.com/), nor is it about a web development 
company (http://www.exnihilo.com/).

Reflecting my personal preferences as any "subject recognizer" reflects 
its creator's preferences, it "chooses" to say this is an occurrence (in 
the topic map sense) of the topic "Ex nihilo" that has the subject 
identifier, http://en.wikipedia.org/wiki/Ex_nihilo.

So it isn't really string mapping but identifying the subject and in 
this case, making it an occurrence of a subject.

If you want to say that is an easy example, you're right, it is. But I 
wanted to illustrate the principle and not get too bogged down in the 
details.

What I suspect would be a fruitful way to deal with the prose field 
issue would be to create an index of those terms with mappings to other 
fields and crowd-source it in terms of what subjects are or should be 
recognized.

The experience with Flickr (which I know only by reports of others) is 
that suggested taggings are remarkably consistent. See my post on an 
article to that effect: http://tm.durusau.net/?p=1126

So, perhaps a "suggested subject" mechanism would be a useful topic map 
application. Indexes a text and then suggests topic for the subjects it 
recognizes.

As I said earlier in this post, this has been a little explored area so 
some of what I have said will be useful and some of it won't be. The 
problem is that I don't know today which is which. ;-)
>> But the same is true for longer "prose" fields. Yes?
>>      
> Yes, I tend to agree if you're talking about TMRM.
>
>    

I think it applies to both the TMDM and the TMRM.

With the TMRM you can write a legend (more on legends later this year, 
hopefully) that avoids things like reification (allows "associations" 
and "occurrences" to have names and other properties), or to avoid 
abusing occurrence and enable representatives of subjects to have 
properties as well as "instances" of those subjects identified in other 
resources. Or to specify complex subject recognition and equality rules.

But all legends depend upon the step that has been neglected up to this 
point: subject recognition by the author of the topic map.

Hope you are at the start of a great week!

Patrick

PS: If you have gotten this far, you are a dedicated reader. Please see 
the blog post I cite above and read Alan Bosworth's article. It is short 
and really should be read every morning before we start working on topic 
map standards and/or technology.

> Best regards,
> Lars
>    

-- 
Patrick Durusau
patrick-Q/[email protected]
Chair, V1 - US TAG to JTC 1/SC 34
Convener, JTC 1/SC 34/WG 3 (Topic Maps)
Editor, OpenDocument Format TC (OASIS), Project Editor ISO/IEC 26300
Co-Editor, ISO/IEC 13250-1, 13250-5 (Topic Maps)

Another Word For It (blog): http://tm.durusau.net
Homepage: http://www.durusau.net
Twitter: patrickDurusau
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.