Monday, April 25, 2005
On the discussion of data
regularity à On the Stephenson Cyber Attack
Taxonomy à
Chapter in book on Computational Semiotics
April 26, 2005
Dear Paul, First of all ... thanks for your effort ...
I know that time is precious to you, and I appreciate your endeavor in order to contribute to the book. Regarding now the outline .... I will make some comments, all of them with the purpose of enhancing your text. So, please accept them as a contribution to improve its readability. I feel the subject is ok for the book, but I will suggest some improvements ...
Initially ... I need a title ... so you should provide one.
Second, it will be interesting to have a gradual introduction to the theme, and that the importance of semiotics is clearly declared within the text. For an attentive reader, this importance is implicit. But I think it is important to make it explicit. You started saying ... A "Second School of Semantic Science" approach to ontology mediation of human communication has been developed." Well ... it will be interesting to talk a little bit about the current school of semantic science, and why it failed in approaching meaning.
So, the reader should understand better why a second school was required. It should be clear to the reader what exactly is the problem you are dealing with .... e.g. the study of the connection between words and the phenomena in real world they refer, and the technique you use to consolidate that by computational means, ... or e.g. the study on how computational (or human) agents may be communicating each other, using natural language (or quasi-natural language) to communicate.
It will be important that the reader is able to contextualize your work and compare them with the current approaches being pursued at the mainstream. Remember, we are out of mainstream ... so we need to make ourselves clear on what we are doing, and what are the differences. It would be interesting if we can also characterize what we are considering mainstream here ... as there are a few independent streams of research ... "Semantic Web", "Evolution of Language", etc.
You proceed with "This approach does NOT use description logics, " ... it will be useful here to explain what is description logic (with references, if possible), why description logic is undesired, and what are the problems related to its use. I didn't understand well your point on a "managed vocabulary".
What exactly are you calling here a "managed vocabulary" ? Are you using an ontology as an interlingua, and creating a dynamic mapping between some exercised set of terms and this ontology ? Can you be more clear here ?
You further relate that to "complex behavior". I became lost here too ... it is not clear to me on how do you connect words to behavior ... are you talking about the overall algorithm running your system, or are you talking about an agent's behavior, triggered by the interpretation of messages ?
At last, it will be useful if you could provide also some references. Well, this is it, to begin. Please, (again), consider my effort here as a contribution, not as a critique. Sometimes we are so embedded in our ideas, that we avoid redundancy in order to be synthetic and pragmatic. But the effect of being synthetic and pragmatic sometimes makes our texts more difficult to read. I feel that Computational Semiotics today has a major challenge, which is to be understood by the mainstream on artificial intelligence and computer science.
A lot of people say to me they are not able to understand computational semiotics, because we use a discourse that is suitable only for those that are already initiated, and the result is that people are not able to fully understand what we are doing. So I have this concern on being didactic.
Now, regarding your second question, about your book on Foundations of Knowledge Science ....I would guess that maybe the publishers will be interested. They publish many tittles regarding Knowledge Management and other stuff ... but I am not a member of their editorial board, so I don’t have the authority to talk by them. I think you should give it a try. Their web site, with instructions for potential authors is here: http://www.idea-group.com/requests/author.asp
This is what I have done in the case ... I wrote them a proposal, and they accepted it. You should do the same yourself.
I will be expecting a new version of your abstract ... and the most important thing here is the title. If you are able to work on the other comments too, it will be fine. But otherwise, this can be done later, if you wish.
I feel your work is representative of the people working with Computational Semiotics, and would really want to include a chapter from you within the book. I will work with you in order to make adjustments on the language in order to make your work more available to a wider audience, if you accept.
My best regards, Ricardo