I was looking at the Afghan War Diary – 2004 at Maiana yesterday.
A couple of things puzzled me so I though I would mention them here.
Take a short look at the ontology for the diary.
I’ wait.
OK, now follow the link for Index of Individuals.
Wait! Err, there wasn’t any category that I saw in the ontology for individuals.
Did you see one?
BTW, scroll down, way down, the listing of individuals. I am assuming that cities and diary entries are both individuals?
I suppose but it looks like an odd modeling choice.
When I think of individuals I think of, you know, people.
I haven’t looked closely but do the reports include the name of persons? That is what I would consider an individual.
Ah, you know what? Individuals = Topics. Someone renamed it.
But how useful is that?
Having every subject represented by a topic in a single index?
That is as unhelpful as a Google search result.
Particularly if your topic map is of any size.
Have indexes of commonly looked for things like geographic locations by name or organizations, etc.
BTW, I don’t think that USMC is of type Host Nation.
If USMC expands to United States Marine Corps then I suspect a type of military organization is probably more accurate.
I stopped looking at this point.
Please forward suggestions/corrections to the project.
Patrick,
the Individuals Index contains Topics that are of any type but are not used as typing topic for other topics. Topics that are used as other topics type are listed in the Ontology. The Master Index then sums up those both indexes.
Your point is absolutely valid for domain specific applications, where those topic types are well-known and certain identifiers may easily be mapped to terms like ‘individuals’, ‘places’ and the like.
But as Maiana is a generic Topic Maps Browser, the term ‘Individual’ should be treated generically, too. I.e., individuals are topics that are not only there for ontological reasons, but carry actual data.
Hope you have a nice day,
Sven
Comment by Sven — January 27, 2011 @ 4:33 am
I think the Afghanistan Diary TM is a good example how you should not model topic maps.
It’s more or less a 1:1 mapping of the SQL dump to Topic Maps which provides no further value. To take a simple example: Why is the “classification level” hidden in occurrences and not a topic by its own? Further, the database provides separate columns for latitude and longitude while the topic map combines both in one occurrence value with an artificial datatype.
Comment by Lars — January 27, 2011 @ 1:40 pm
I have to agree with lheuer.
Just dumping content into TM form without unearthing subliminal structures gives little or no value. Worse, not understanding the domain and the rules and limitation of the original content carrier (RDB, or text files, etc.) leads to simply mirroring the content in its original form.
Ad Sven: “Individuals” is from the terminology of the RDF people, and – unsurprisingly – spreads confusion in TM space. Also the artificial barrier between “types” and “instances” is just that. Artificial. Right from the book. 🙂
Comment by Robert Barta — January 27, 2011 @ 2:47 pm
Patrick, I really can’t understand your confusion concerning the individuals. With the individuals index Maiana is reimplementing the individuals list everbody knows from Ontopia. In this index you see all topics which do not act as type. Very simple.
Besides this I’m very curious to see the first perfectly modelled and curated Afghan war diary, Cables summary or Palestine reports uploaded in Maiana.
Comment by Lutz — January 27, 2011 @ 3:01 pm
… well, you asked for comments …
I wonder if any of the Afghanistan Diary TM supporters have ever looked at the source. I encourage everybody to compare the SQL dump found at http://cryptome.net/afg-war-diary.sql.7z with the generated topic map http://maiana.topicmapslab.de/u/efi/tm/wd2004/download.ctm
The generated topic map simply maps mainly the table columns to PSIs and the values to occurrence values.
What can you achieve with the topic map which isn’t possible with SQL?
Any idea?
Comment by Lars — January 27, 2011 @ 3:34 pm
@Lutz –
Quite honestly I had forgotten that the Omnigator had that oddity.
And at the time I was looking for individuals in the natural person sense.
To me reports that don’t identify an individual or have a traceback to an individual (natural person sense) are nearly useless.
Publication of such reports (topic map or no) may have some titillation value but little else. Accountability means that named persons were responsible for action/non-action, decisions, etc.
I suspect efforts are underway on one of more of the resources you mention but I rather doubt any of them will be “…perfectly modeled and curated…,” as much due to lack of domain expertise as anything.
I doubt any of them will be uploaded to Maiana as “fly-paper” model for content delivery isn’t all that appealing.
Imagine if Apache had said, “…upload your HTML files to our web server…” instead of “…setup your own webserver, that we will keep improving.” Very different delivery models.
Comment by Patrick Durusau — January 27, 2011 @ 4:09 pm
@Lutz:
“””
I’m very curious to see the first perfectly modelled and curated Afghan war diary, Cables summary or Palestine reports uploaded in Maiana.
“””
I’m very curious to see your mapping from the original source to TM.
Further I am curious to see any proof that the provided mapping solves problems which cannot be solved by the SQL dump.
Comment by Lars — January 27, 2011 @ 4:10 pm
@Patrick: Lutz and Sven are referring to Maiana. You can think of Maiana as an Omnigator clone that isn’t Open Source. 😉
Anyway, I wonder how the topic map could beat http://warlogs.owni.fr/ and http://213.251.145.96/iraq/diarydig/
Sven may argue that it’s just a presentation issue and he may be right. And even if it’s just a presentation issue, why should a user switch to Topic Maps if she could use https://github.com/yourcelf/afgexplorer ?
Anyway, the topic map seems to be good enough for Emnekart 2011 which speaks for Topic Maps, I guess
Comment by Lars — January 27, 2011 @ 4:35 pm
@Lars –
Well, for starters, note that http://warlogs.owni.fr/ ask if “…we should investigate further…”
Does this imply that the host has the redacted information?
Topic maps aren’t required to allow others to merge their own information but if I were another national government or wannabe national government, that is why I would use a topic map.
So that I could overcome the semantic impedance of different models. Important to be able to look back at the original model/information.
Not to mention being able to add my secret information along side that captured from others. And preserving the captured information “as is.” Not as it is seen or summarized by some arm chair analyst on the other side of the world. (Or at least being able to drill down to it.)
Comment by Patrick Durusau — January 27, 2011 @ 5:04 pm
Well, I was able to construct several alternative keyword layers for the AWD within few hours using Wandora’s extractors. I would really call that an advantage. The experiment is documented at
http://www.wandora.net/wandora/forum/viewtopic.php?t=56
Comment by Aki — January 28, 2011 @ 3:38 am