Dear all,
I'm currently experimenting with using the Exhibit tool to prototype a demo for
faceted browsing techniques on 1000s items, in fact I have used it for over
10,000 items
To enable this to be usable I'm having to nobble Exhibit a little. I only show
the first 20-50 tiles (tabular view is not possible), I disable the "group by"
option, all the facets begin as collapsed and thus are not created unless
require, I do not have facets on properties with many values, the data is
stored in multiple collections and the data is loaded when it needs to be
displayed.
Load time can be up to around 1 minute for large amount of data but browsing
and display are fine.
N
> From: [EMAIL PROTECTED]
> To: [email protected]
> Subject: RE: Questions on exhibit capacity and performance
> Date: Tue, 11 Dec 2007 16:25:07 -0500
>
> I'm currently at 663 events with a map and timeline on my exhibit:
>
> http://www.uss-king.com/TimeLine.shtml
>
> My JSON file is a little over 230 KB. IE7 takes about 25 to 30 seconds to
> load and execute on broadband. It isn't much quicker run on my local machine
> (in fact, I'm not sure it's any faster), so the code execution seems to be
> where the slow down occurs. Firefox (2.0.0.9 with no plug-ins) loads and
> executes in about 8 to 10 seconds.
>
> After it all loads, navigation is pretty quick although it takes IE7 about
> 20 seconds to reset the filters (about 5 seconds for FireFox).
>
> Given the amount of information and the difficulty involved in presenting it
> in other ways, I can live with these execution times for my project,
> although I would love to see it run faster.
>
> Mark D.
>
> -----Original Message-----
> From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
> On Behalf Of Ryan Lee
> Sent: Tuesday, December 11, 2007 4:02 PM
> To: General List
> Subject: Re: Questions on exhibit capacity and performance
>
> As Robert notes, finding a hard limit is probably infeasible given the
> nature of the web and browsers. However, in terms of orders of
> magnitude, we have a general perception that 1,000 is a killjoy, a few
> hundred can be nice, and the ground between depends on your computer's
> specs. There's nothing formal behind that breakdown, and other
> anecdotal experiences are welcome to help flesh out a more useful picture.
>
> Robert Forkel wrote:
> > i doubt there could be really comprehensive testing of this issue,
> > because the environment it depends on - javascript in browsers - is
> > too varied. even in the case of firefox 2.0.x, which i am using, it
> > depends on the number and kind of installed plugins, the number of
> > opened tabs, the time the browser has been running, ...
> >
> > On Dec 10, 2007 7:53 PM, Mark Feblowitz <[EMAIL PROTECTED]> wrote:
> >> Has anyone done any testing to see whether/how/where exhibit
> >> performance degrades as the number of items or the total size of the
> >> exhibits grows? This could be very valuable in deciding whether/how
> >> to use exhibit.
> >>
> >> Thanks,
> >>
> >> Mark
>
> --
> Ryan Lee [EMAIL PROTECTED]
> MIT CSAIL Research Staff http://simile.mit.edu/
> http://people.csail.mit.edu/ryanlee/
> _______________________________________________
> General mailing list
> [email protected]
> http://simile.mit.edu/mailman/listinfo/general
>
> _______________________________________________
> General mailing list
> [email protected]
> http://simile.mit.edu/mailman/listinfo/general
_________________________________________________________________
Free games, great prizes - get gaming at Gamesbox.
http://www.searchgamesbox.com
_______________________________________________
General mailing list
[email protected]
http://simile.mit.edu/mailman/listinfo/general
