Sunday, July 30, 2006

No, Google I Don't mean "Gay Films"!!! Why the semantic web cannot work, the properties free and good need to be porn:OpaqueData and porn:Misleading

Hi all,

I am looking for open source tools to deal with amazon
and that have book interfaces.

After using debian apt-cache search amazon to select a couple of
packages that looked interesting i then googled for them,
" alexandria cowbell gcfilms ".

Although I have moderate search filter turned on,
and personalized google search suggested :
Meinten Sie: alexandria cowbell gay films

I was pretty shocked, and after I turned on strict filtering
it still returned the same thing.
well, I guess google is just catering to its porn clients.

Hopefully this will get slashdotted and google will clean up its act.

Now, on the topic of porn, I would like reiterate on my view of
the semantic web.

The reason why the semantic web cannot work is that
it cannot be used to trick people into looking at pay porn sites.

Let me state a couple of assertions :
  1. the semantic web is a medium
  2. for a medium to be viable, it must be usable to sell porn.
  3. all successful mediums have been used to sell porn
  4. there is no such thing as free porn.
  5. There is no free bandwidth.
  6. the semantic web is disjoint from Opaque and Misleading Content.
  7. the porn industry needs a Medium tha can be used to Mislead you into looking and clicking thru into thier pages.
  8. the porn industry needs to create a misleading meaning of free, thus redefining the term free to non-free.
  9. the semantic web is to eliminate that possibility of creating misleading content.
Therefore the global semantic web cannot be used as a viable medium for misleading porn advertising. Because it is disjoint with Misleading a subclass of Content.

Of course the semantic web could be used to create an ontology
of porn and be used in local semantic web,
and from that web a misleading html web could be created.

But as long as the semantic web cannot be used as a misleading medium
for advertising pay porn and
the misleading ads being mixed in with the supposedly free content creates
the viability for the medium.
but exactly this mix is what is explicitly excluded from the semantic web.

the result would be a pure porn page that allowed peer to peer exchanging of
porn based on semantic tags,
that would fulfill on aspect. But as soon as you get into the ability to
globally tag all porn
then the issue is that most of the porn is bad. Not only is the attribute
free misleading, but also the term good.

so in the end, the isps/search engines cannot bite the hand of
it low quality porn industry that is feeding it,
and will never support the semantic web for the customer fully.

In fact, this brings me to the conclusion that the customer will
always need to be decieved for advertising to be received,
and that for a medium to be successful it will always need to
contain opaque and misleading data in it.

I would like to suggest the following
namespace porn with the classes porn:OpaqueData and porn:Misleading as subclasses
of porn:FreePorn and porn:GoodPorn.

You can run the ontology on pellet and it will prove that the semantic web is not satisfiable

Here is a nice interactive view of the ontology

basically I state that the semantic web as a medium is disjoint from advertising.

pellet say also :
B:Disjoint Classes axiom found: DisjointClasses(SemanticWeb Opaque)
Disjoint Classes axiom found: DisjointClasses(SemanticWeb Medium)
Or: unionOf(Misleading Opaque Advertising)

I look forward to some feedback! please send me your comments

mike