John,
See below in line.
Regards
Sjir Nijssen
Chief Technical Officer
PNA Group
Tel: +31 (0)88-777 0 444
Mob: +31 (0)6-21 510 844
Fax: +31 (0)88-777 0 499
E-mail: sjir.nijssen@xxxxxxxxxxxxx
-------------------------------------------------------
http://www.pna-group.com
-----Oorspronkelijk bericht-----
Van: ontolog-forum-bounces@xxxxxxxxxxxxxxxx [mailto:ontolog-forum-bounces@xxxxxxxxxxxxxxxx] Namens John F Sowa
Verzonden: zaterdag 12 januari 2013 17:32
Aan: ontolog-forum@xxxxxxxxxxxxxxxx
Onderwerp: Re: [ontolog-forum] [External] Re: What is Data? What is a Datum? 2013-01-09-0930
Ed, William, and Sjir,
EB
> I see data as the information that is represented by the marks, and
> you see data as the marks that convey information. We label different
> roles in the same relationship as "data", but we both demand that the
> relationship exists in the eye of the reader.
I more or less agree. But that discussion shows that the terms 'data' and 'information' are more difficult to define than the the more basic triad of mark, token, type.
EB
> It is not sufficient that those marks have an interpretation. If the
> writing on the page is in Mayan, it is not "data" for you; it is just
> marks. Those marks are only data when you can read them. Reading is
> the mental transformation from marks to meaning.
Peirce emphasized that every perception of a mark is a classification as a token of some type. An early explorer who described carvings on a Mayan wall couldn't read them as a language, but the descriptions could still be called data.
In fact, much of the early "data" proved to be quite useful for the linguists who finally decoded the glyphs. All the partial interpretations along the way could be called data.
WF
> ... by the time we are doing data processing, we know what those bits
> in the computer represent, a number, a character, a truth value, a color...
I agree. But you can recognize bits as data of some kind, even if you don't know which bit patterns represent tokens of which types. When you look at a raw bit string, you're like an early explorer looking at glyphs on a Mayan wall.
SN
> I believe the term Conceptual Schema (Complete truly Conceptual Data
> Model) as introduced in ISO TR9007 is quite precise.
I agree that it's a good report. It's over 25 years old, but ISO still charges 210 Swiss francs for a PDF. Is there any way to persuade them to release it?
[[Sjir: The secretary of ISO TR9007, the late Joost van Griethuijsen, informed me that he has tried but it seemed impossible.
What I could do is attaching the proposal (1978) that contained the basis for the most important parts of ISO TR9007, however it is 76 pages and 25 MB and I have been told that such an attachment it not permitted
in this forum.]] (I received a copy years ago from one of the authors.)
SN
> I recommend to this forum to give priority to a better set of
> definitions of ontology...
I agree that clarification is desirable, but with many qualifications.
Absolute precision in the general terminology of any science is not achievable or even desirable. For example, biology is the science that studies life, but nobody has a precise, general definition of life. We could define all known
life forms by requiring some version of RNA or DNA. But that definition would probably be useless for recognizing life forms outside earth.
For computer examples, the term 'operating system' originated with the mainframes of the 1960s. But today, an Android smartphone contains a Linux kernel, and it's still possible to run Linux programs on it.
How would a definition of OS from the 1960s apply?
[[Sjir: I agree with you that definitions have a lifecycle and that every now and then you have to update the definition. A well known example is marriage in The Netherlands.]]
SN
> A good theory and engineering practice can only be built on a much
> more precise definition than used so far.
I agree that each theory requires precise definitions. But as the theories evolve, the same words are used, but the definitions change.
In physics, the concepts and definitions of mass, energy, time, and space have undergone major revisions since the 17th century.
But we still use the same words, and we still use Newton's equations for the motions of macroscopic bodies on earth.
Scientists and engineers continue to use the same terms, even though they use different definitions and approximations for different problems. In fact, engineers routinely use *inconsistent* approximations for different aspects of the
same system.
Basic principle: For each application of each theory, the terms must have the same definition throughout the computation.
But large applications can and *must* use the same terms with different definitions in different parts of the system.
[[Sjir: my experience is that it is possible for the ISO TR9007 concept Conceptual Schema when one consistently starts with observable forms of communication and then work up through the conceptual domain model (including extensive testing)
and from there to the conceptual generic model (including extensive testing). But most people prefer to go the other way around, with little result.]]
John
_________________________________________________________________
Message Archives:
http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr:
http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe:
mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/ Community Wiki:
http://ontolog.cim3.net/wiki/ To join:
http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J