Peter Cooper wrote:
Denver
I really don't think that ISC will have turned off the JVM hyperevent - it wil break to much code already out there
Suspect that they will make it an optional attribute somewhere <csp:object useXMLHTTP=1
or some such construct - if it's not there use JVM if it is use XMLHTTP
==== My own view is that it is a welcome replacement for having to dl and install a JVM
But it still is browser dependent - useable if you have control over the browsers - intranet - but not ready for public internet prime time too many browsers not supporting it *yet*
It's the old probem - you can make use of whizzy features but the more you do the more you restrict the audience
for my own part I am using simple Javascript with very simple DOM fiddling this works on 98% (I believe) of browsers - but even then it wil fail if the user has turned off JavaScript But this is reasonable - so many public sites use JS that I think it's OK
XMLHTTP - don't work on Mac's a all, flakey on anything less than the
latest Moz - too large a percentage audience
or so my research indicates
Peter
On Sat, 07 Aug 2004 14:03:32 -0400, Denver Braughler <[EMAIL PROTECTED]> wrote:
Bill McCormick wrote:
Just to give you an incentive to drool over 5.0.11 I can tell you all that the complete XMLHTTP implementation of the hyperevent broker sans JVM has made it in this kit including all the configurable bits
:)
I like to make people happy sometimes
I certainly am happy to hear that XMLHTTP is going to be in 5.0.11.
Is this an all-or-none arrangement where all CSPs must be re-compiled, or is there a migration path? Will attribute INSERTBROKERAPPLET become a no-op? Is cspbroker.class still going to be used with browsers that do have a JVM but don't support XMLHTTP?
