Re: Yawas : new Web annotation system

Ka-Ping Yee wrote:

> I wrote:
> > These references are predated by my introduction of the concept
> > of "mediator" in 1995 (http://www.lfw.org/ping/mediator.html).
>
> On Sat, 20 Mar 1999, Ron Zalman wrote:
> > The notion of intermediator used in my paper, is an adaptation of
> > the term defined by Barrett,R., and Maglio, P of IBM. It is similar
> > to the mediator term you had suggested, but is much more general.
>
> As far as i can tell, what Barrett and Maglio describe in their
> paper is the same thing.  They write:
>
>     We define intermediaries as computational elements that
>     lie along the path of web transactions.
>
> That's what a mediator is.  The following is excerpted from my
> 1995 definition:
>
>     ... a mediator is a service that functions simultaneously as a
>     server on its front end and as a client on its back end,
>     which acts not only as a proxy for the user's request but
>     also performs some useful type of processing on the retrieved
>     document.
>
> The second feature which i specified, modifying the link URLs
> in a document to cause subsequent fetches to also use the mediator,
> was my way of implementing a mediator to be universally accessible
> to all browsers, despite possible lack of proxy support in the
> browser and any firewalls that might get in the way.

First, let it be noted that I think your definition should be mentioned, as well
as your important work.
Still, as far as defs. go, I'd have to disagree. What you probably had in mind
when first defining the 'mediator' term, was a proxy based annotation system. For
instance, you write: "...also performs some useful type of processing on the
retrieved document...". This implies that the 'mediator' term is well defined only
for document retrieval. The 'intermediator' term defined by Barrett and Maglio
speaks of ANY web transaction which is more general than document retrieval.

But this observation is pure semantics, and is of lesser importance.
What I find much more interesting is the matter of scalability.
Do you use some kind of distribution ?
Have you made any capacity estimations for crit.org ?
Can you describe the technology used by crit.org to store annotations and to serve
multiple clients ?


Regards,
Ron Zohar-Zalman.

Received on Saturday, 20 March 1999 11:23:07 UTC