3. Knowledge does not supposed to be only
theoretical. Neverthless, a logic-checkers (semantic analog of the
spellcheckers) could work for proper formatted "theories" (texts). A sort
of the scientific CNL can help to avoid ambiguity of scientific terms and
phrases in thesis by adding "red marks" to unrecognized logical movements. Being
adopted by academical institutions and scientisiv funds such "semantic" norm
(established standards for logical checkers) could help the semantic
software to spread in the real life. Why not?
"both in one" is a very rare case. Knowledge
keeper does not know XML, RDF, Sparql, OWL and other funny things. And she is
right. But she does not know FOL. It's a pity;-)
Programmer usually
knows only these funny things. But he does not know FOL as well. It's a
shame;-)
Anyway everybody has some knowledge to share. And how many
knowledge programmers do we have right now?
Before, we had knowledge
guru with their enigmatic Knowledge Base Management Systems. And SW has at
least one merit: now any programmer can write ontology. Well, it's too
much:-)
With CNL any plodding knowledge keeper can;-) Is any
narrative text is an ontology, at least a factology (aka A-box)?
With
Simple English we have more interesting situation: many type of text may be
simplified grammatically without to loose or corrupt a knowledge. And still
being readable. A problem is that between CNL-statement and a formula from
some logical theory there is a gap: we should know exactly which theory she
keeps in mind (so named context), i.e. transformation is not universal.
That rules of "translation" are theory specific.
For many areas we
still do not have theory itself. Theory as a system of definitions at
least.
Alex
PS а за пределами форума можно пообщаться и на
русском. буду рад:-)
I like your point but how can you distinquish "programmers" and "end
users" epecially if they are "both in one" )
BTW follow your post I have found (from the
"Simple English" http://simple.wikipedia.org/wiki/Simple_English_Wikipedia )
that "... people use the Simple
English Wikipedia because the simple language helps them understand
difficult ideas..."
If it was true we would have
Simple Math and Simple OWL
for academics and presidents )
Regards
Yuri
P.S. Many of us use English as a second
language (for ex. my native is Russin), so I am sorry if it looks like
a joke ;)
Surely there is a distinction to be made between what is
modeled and what it is modeled in. Changing the syntax or the presentation
is not the same as changing the content of the model, and in the end a
model is either a model of something in the problem domain or a model of a
(usually proposed) solution.
Mike
On 08/11/2010 08:14, Alex
Shkotin wrote:
John,
as XML, RDF are data models and
ways for information coding, it is not for users but for
programmers. XML is accepted by programmers for their
needs. Internal structure of SW is a programmers deal.
We have
5(!) syntaxes for OWL 2. I think none of them for end user. These are
for a new kind of programmers: knowledge coders;-)
And as usual
for end user we need forms to fulfill, manage and query
knowledge.
It seems one additional "user interface" we may get
now is a very simple CNL. We get it while verbalizing RDB data, for
ex. And for me it is better than "metadata", as end user can write
CNL.
We have know spellcheckers working background when we type.
It would be crucial step forward for the web to have a background
CNL checker the user accept. Not everybody but who'd like to type
knowledge;-)
Even for knowledge interchange it may be better to
have CNL. Do we need a standard for CNL?
> Even if Martin only published GoodRelations in RDF/XML
format, > the conceptual schema isn't inextricably to RDF
formats.
Yes. That's the point I was trying to
emphasize.
Nobody develops alternatives to good notations.
The fact that there are so many variations is a sign that the
edict to make the XML serialization the normative version was a
disaster.
People constantly say "Oh, you don't have to use it
if you don't like it." But making the XML serialization the
normative version is a terrible example of a premature optimization
in favor of the semantically least important and most trivial
aspect.
The first step toward a rational reconstruction of the
Semantic Web is to demote the XML serialization of the languages to
a legacy status. Then sponsor a design competition for better
alternatives.
--
Mike Bennett
Director
Hypercube Ltd.
89 Worship Street
London EC2A 2BF
Tel: +44 (0) 20 7917 9522
Mob: +44 (0) 7721 420 730
www.hypercube.co.uk
Registered in England and Wales No. 2461068