>> 1) browser-applet interaction is
>>     * highly restricted
>>     * buggy
>>     * different in every browser
>
> Browser/applet interaction is restricted, but the interaction between
> the applet and the servers it was served from is unrestricted.  Since
> the data that a user is working on is on the server, I do not see why
> very much browser/applet interaction would be needed.

Then read the notebook source.  If you capture a key in a cell, it might affect 
the rest of the notebook.  Interrupting, opening print / help windows, etc.  
Evaluating the bottom-most cell spawns another cell.  The list goes on.



>> 3) I frequently use worksheets with hundreds of cells.  This, in turn, runs 
>> my computer out of memory.  This is not using java.  Add the overhead of an 
>> applet for every cell... you get the idea.
>
> Sun's Java-SE implementation has included Class Data Sharing since version 
> 1.5:
>
> http://java.sun.com/j2se/1.5.0/docs/guide/vm/class-data-sharing.html
>
> So, I do not think that one can assume that having multiple applets
> running will use too many resources without testing this.


I remain skeptical.


>
> If the resource usage was too high, then an alternate strategy of just
> using one applet and then binding it to the current cell being edited
> could be explored.

Absolutely not.  We tried a "change the cell when you click on it" scheme for 
about two weeks.  My ears are still ringing from all the complaints I heard 
about it.



--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/
-~----------~----~----~----~------~----~------~--~---

Reply via email to