Simon Brooke wrote:
> On Tuesday 15 May 2007 19:29, David Huynh wrote:
>
>
>> Simon, have you had a chance to check out Exhibit? It seems related
>> to PRES 1.10.0 beta 2:
>>
>> http://simile.mit.edu/exhibit/
>>
>> Having had world enough and time... we then want some scatter plots,
>> too...
>>
>
> OK, questions:
>
> (i) Why is the data file a JavaScript array, rather than (e.g.) XML? Are
> there some specific (e.g. efficiency) reasons for doing that, or is
> that just the way you happened to do it?
>
JSON is a bit more readable/writable than XML, especially for arrays,
e.g., compare
{ label: "David Huynh",
has-degree: [ "B.A.Sc.", "S.M." ]
}
with
<item>
<label>David Huynh</label>
<has-degree>B.A.Sc.</has-degree>
<has-degree>S.M.</has-degree>
</item>
> (ii) Have you played with any scenarios in which the user can modify or
> add to the data, and save it back?
Not quite yet but I'm hearing demand for that. I'm thinking of using
Google Data API.
> If so, how do you resolve issues
> where two different users are making modifications simultaneously? Must
> the locking granularity be at the level of the whole data set, or would
> it be possible to use some finer granularity? How would you produce a
> visual 'diff' of two data sets, or a user interface for resolving
> differences?
>
>
>> http://people.csail.mit.edu/dfhuynh/projects/factbook/factbook-people
>> .html
>>
>
> In this example, if you select e.g. 'EU' and 'LITERACY VS. EXPECTANCY',
> the data is highly clustered and hard to discriminate. It would be nice
> to be able to zoom the graph a la Google Map, or to change the scaling
> from linear to log.
>
The plot can be set to log scale in the HTML. But yes, dynamic control
would be desirable.
> But... very interesting and promising.
>
> I confess I'm particularly going to study how you achieved the
> semi-translucent map pins!
>
:-) Just translucent PNGs.
David
_______________________________________________
General mailing list
[email protected]
http://simile.mit.edu/mailman/listinfo/general
