David Huynh wrote:
> David R. Karger wrote:
>
>> I'm not certain the distinction between "view" and "lens" can really
>> be formalized---obviously, every view is only showing some aspect of
>> the object. But there do seem to be very different ends to the
>> spectrum, that perhaps need to be handled differently?
>>
>>
> In practice, when one writes a lens in Haystack, it is often with hopes
> that the lens can be used for more than one type of information (e.g.,
> an "address" lens can be used for people and companies). When writing a
> view, it is often with concession that it can only be used for one type
> of information, in a limited context, for a particular task. The view is
> specially laid out and cannot be guaranteed to generalize.
Right on! I like the idea of separating between a view and a lens based
on its rate of generalization.
On the other hand, I can't stop thinking that this "degree of
generalization" is really a float and not a boolean.
> When we created the view architecture, we conceded that we did not know
> how to separate content (e.g., set of property-value pairs to show) from
> presentation. Then we created the lens subsystem and tried to ignore
> presentation altogether (i.e., we laid those pairs in a table,
> regardless of whether there were more effective layouts). That is, a
> lens is a selection of content to be shown, while a view is both a
> selection of content and a (specialized) layout of that content.
I think we are converging on the idea that there are at least two layers
for content:
[selection] ---> [presentation]
and the scope of this effort is to trying to minimize the ontology so
that it works in the [presentation] layer.
Note that there are three aspects that we still haven't considered:
- adaptation: the act of adapting the content to some given needs
- interaction: the act of capturing user events and translate them
into further selection or adaptation
- modification: the act of modifying the content in a persistent way
I was talking to David yesterday and we agreed that while we should try
to modularize the ontologies as much as possible, it's also pretty hard
to draw a line without knowing where the other things on the table would
go when needed.
What do you people think about that?
--
Stefano Mazzocchi
Research Scientist Digital Libraries Research Group
Massachusetts Institute of Technology location: E25-131C
77 Massachusetts Ave telephone: +1 (617) 253-1096
Cambridge, MA 02139-4307 email: stefanom at mit . edu
-------------------------------------------------------------------
Received on Wed Oct 13 2004 - 19:09:59 EDT