Re: The AF war diary

Hi Benjamin,

Thanks for jumping in.

[...]
> Let me bring some light on how it all started. One day somebody¹
> came into my office (which I share with Thomas) and disrupted us
> from our work, like it happens every day.
[...]

You left out some important details like

    * Was it a rainy day or did the sun shine?
    * What did you eat for breakfast that day?

;)

[...]
> ² If I remember correctly, the only things which thwarted him for a
> second was strange literal escaping and the use (and even inconsistent
> use) of literals in CTM.

Can you explain what's meant with "strange literal escaping" and
"inconsistent use of literals"? Or add an issue CTM's bug tracker?


[...]
> Indeed somebody sees a value in it. People are even writing mails and
> ask us to present it on conferences. The main point of the talk we're
> going to give is not how great the result is but the little effort
> which was needed to obtain it.

That's strange. A proof that it needs a little effort to create a
"suboptimal" (not to say almost meaningless) topic map.


> I guess Aki and Lars each spent 10 times more time on thinking about
> it than Thomas and I spent on it together. Now where are your 10
> times better results, Lars? I can see Aki's.

I don't think that I spent 10 times longer to think about it. It's not
that difficult to recognize that the topic map is not "well designed".

I was simply surprised that this topic map has been
accepted/invited/whatever for a Topic Maps conference and that your
company seems to see a value in it which I haven't (otherwise your
company wouldn't have mentioned it at Practical Semantics [1],
promoted it here [2] and here [3]).

That's the reason why I came back with the (seriously meant) question
why the topic map should be interesting and not a sign for
incompetence. I asked myself which facts I have overlooked and gave
the question back to the creators.

Back to your question: "What are your results Lars?": That's a fair
question and I have to admit that I have no results since I didn't
dived into the AF diary log. But some mistakes are obvious:

   * Why isn't aren't the classification levels are topics (i.e. [4])?
   * Why are date/time values represented as string?
   * Why is "complexattack" a string value (i.e. "FALSE")
     Shouldn't that be a xsd:boolean value or maybe even better
     an association?
   * Why is schema:title an occurrence *and* a name?
   * Why inventing another datatype for coordinates?
     We have a lot of ontologies for coordinates, who else uses
     <http://en.wikipedia.org/wiki/World_Geodetic_System_1984> to
     represent longitude/latitude values?

Looking at the original source, I'd ask:

   * Should "UNKNOWN" (i.e. as reporter) be represented as topic?
   * Is "dcolor" an occurrence or a topic?

These are a few questions which could be answered without even diving
into the textual content of the reports (summary, title).

I don't expect that these questions are answered by student who enters
your room by mistake and thinks that it would be cool to do something
with Wikileaks since the Afghanistan logs are mentioned in the press,
but I'd expect that your company answers the question how valuable
this topic map is before marketing it.

Regarding the positive feedback you got for the topic map: I wonder if
anybody ever looked into the source and recognized that it is a 1:1
mapping.


[...]
> It took a while to write down this E-Mail, so maybe I'm now on par
> with Lars' efforts on (working on UNION discussing) the AWD map. Now
> what about your results, Lars?

See above.


> One remark regarding Lars' and Robert's remarks on how to not model a
> topic map: We all know and agree that a 1:1 mapping of relational data
> is not what one may considers a beautiful topic map, but who except
> you cares?

I'd think that a company which carries "Topic Maps" in its name should
care. Everybody who understood Topic Maps should care. Presenting the
topic map in its current state to an audience which has no clue about
Topic Maps and which has no idea how the source looks like, isn't
difficult.


[...]
> It's the same question as if the TMDM is beautiful or not.

I don't see what these questions have in common.

> have their mental models in terms of a topic map. People like lists of
> data and tables of data. That's one reason they use these structures
> all the time. (The other is: they are used to the tools which provide
> easy working with these structures.)

What's your conclusion? Don't use Topic Maps, use lists and tables?

[...]
> needed with currently available tools). Topic Maps come handy when you
> want to describe the existing structure (i.e. the 1:1 mapping) and
> match it with other existing structures (i.e. other 1:1 mappings). I
> think nobody want's to spend hours of time to transform data to
> another structure in the hope of maybe better results. It also cannot
> easily be done automatically.

To stay at the concrete example: All my questions above can be solved
automatically by an algorithm if the user behind it would take care
about. And even without diving into textual analyses and natural
language processing you'll get a better topic map.

> interested in the incidents themselves, so I won't spend a lot of time
> in unearthing subliminal data. In my point of view, it was just a
> quick exercise to see how fast it can be done and how the tools
> available perform.

Funny statement in the light how much fuss was generated around this
particular topic map instead of saying "it's just a half backed
experiment".

It's true that Wikileaks gets a lot of attention these days and I
think the topic map wants to gain some profit from it. That idea is
IMO very good but this particular topic map does not represent the
concepts of Topic Maps very well and does IMO more harm than good
since it shows interested users almost nothing what Topic Maps is
about.

If we'd have a well designed AF topic map, we could i.e. merge the map
with a Cablegate topic map and answer maybe interesting questions like
"What do the cables tagged with AF tell us in the same time period?".
Answering these kind of questions would be much easier with Topic Maps
than with SQL/CVS.


[1] <http://www.practical-semantics.com/program>
[2] <https://twitter.com/tmlab/status/29801644367544321>
[3] <http://www.unternehmen-region.de/de/5546.php>
[4]
<https://github.com/heuer/topicmaps/blob/master/cablegate/classification.ctm>

Best regards,
Lars
-- 
Semagia 
<http://www.semagia.com>

<http://www.topicmaps.de/mailinglist/> German Topic Maps mailing list
<http://tinytim.sourceforge.net/> Open Source Topic Maps engine
<http://mappa.semagia.com/> Mappa - Python Topic Maps engine
_______________________________________________
topicmapmail mailing list
[email protected]
http://www.infoloom.com/mailman/listinfo/topicmapmail