cocoon-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Stefano Mazzocchi <>
Subject Re: [RT] Cocoon Emotional Landscapes
Date Fri, 13 Oct 2000 21:38:04 GMT
Peter Donald wrote:
> >This is where I see the parallel: Cocoon and friends are now what the
> >web was 10 years ago, only with much more hype attached due to the XML
> >label.
> I agree
> >It will take another 4 years for XML to really take off and create the
> >critical mass to starting having semantic searching.
> do you really think there will ever be semantic searching?

Yes, I do. This whole project is nothing but the liftoff platform for
the semantic rocket. :)

I know you guys don't see this yet, I know.

I'll reveal more about this in my ApacheCON 2000 speech :)

(that I will make available right after the conference to everybody thru
the site and even, if you like, included in the distribution as I did
for the last speech).

> After working on it for a while I have basically decided that it 
> wont ever happen.

Believe me, I entirely know your feeling about this, but something deep
inside tells me something and shows me the way.... you know I have a
strange ability to see where the wind will blow before it starts

> I looked at RDF and thou - naah too complex which means your basically 
> left in same situation you are in now.

Yep, but please answer my question: would XML publishing have sense
without a solid and portable transformation capability?

Now, does RDF makes sense for the semantic web without XXX?

The problem is defining what XXX is, but the parallel is strong: both
XML and RDF are way too general to mean something, you need more stuff
on top.

The RDF model has a weak RDFSchema on top that is highly overlapping
with XMLSchema with features and highly rejected by people because of
its complexity and apparent uselessness.

Just like you're doing now.

But semantic searching is too important to let the idea die.

> Basic textual analyisis and hit n miss datamining
> - is there any reason to believe this will change?

No, but here we are talking about the biggest collection of coherently
formatted and heavily hyperlined structured text in the history of

It would make ancient Alexandria look a elemntary school library.

It's not just text anymore, so regular expressions and friends are poor
man tools compared to even basic contextual searching or even something
as simple as mechanical semantic inference.

No AI will be required, just a good strategy and some luck.

> I can't see Joe
> Six-pack conforming and maintaing correct meta information and considering
> most of the web is made up of our friendly neighbourhood Joes - unless we
> move to corporate web - I can't see it happening. What do you see that
> makes you think otherwise ?

Because Cocoon3 will give them the ability to perform semantic searches
on their local sites: Joe Six-pack will have to use text-based searching
for its site and loose market if others provide a much more efficient
semantic search.

Don't get me wrong: I do share your vision about what RDF and friends
are today, but I'm also aware of the fact that in a few decades the
amount of available information on the web will be so huge and
(hopefully) coherently organized that textual searches will be like
searching for textual strings in a library of images: useless.

Imagine a world of Cocoons and a web with thousands of different
user-agents and file formats that grows exponentially every year: how
can something like Google keep up with it?

The whole search experience must be redesigned from scratch.

And this is want I want to do with Cocoon3.

If you want to see it happening and help making the history of
technology, just stick around for the next couple of years :)

Oh, yeah, last thing: try to stop me :)

Stefano Mazzocchi      One must still have chaos in oneself to be
                          able to give birth to a dancing star.
<>                             Friedrich Nietzsche
 Missed us in Orlando? Make it up with ApacheCON Europe in London!
------------------------- http://ApacheCon.Com ---------------------

View raw message