Java Slowdown?
 

I went to this site and got the spinning beachball. Safari seemed to handle it fine. I went to the Java homepage and OW passed the Java test.

[url]http://www.santafeinformation.com/[/url]

Anyone else notice any problems loading Java content?

The site loaded (and regularly refreshed) for me without a problem - though I *did* have to close the annoying Java box...

Peter

No problem here, either.

i got the spinning beachball of death ... had to force quit omniweb.
my computer: iBook 600 MHz G3

I got the spinning beach ball as well...had to force quit twice for it to actually close.

OW 5.5.4 (v607.17)
OS X 10.4.8
Intel Core Duo 2.16GHz
2GB RAM

No problem here.

OW 5.5.4
10.4.8
2GHz Intel Core Duo

Not a problem here either.

PowerMac G4 Dual 867
1.5 GB RAM
OS X 10.4.8

Looking at that site's source code*, I'd wager that this is yet another occurrence of the damnable JS/garbage collection bug in the current version of WebKit which is why some people are experiencing a beach ball and others are not. Probably not related to Java at all.

* Unsurprisingly, it contains a lot of JS.

Explain that, please
 

Jonathan, for the novice, where is the WebKit leak occurring, and why would it affect some users differently than others?

Beginning with 5.5.3, I have experienced the SBBOD almost every time I am away from OW for an extended period (an hour or more), leave it up on my screen, and then attempt to return to it.

It isn't a leak, it is a bug in the JS garbage collection. The reason why it will affect some users and not others (or so I am told - I don't know the details as I am no programmer) is that the garbage collection only starts once a certain threshold has been reached. Therefore, if this is the first JS page you load, you will probably not experience the bug, but if it is the [I]n[/I]th page in a session to contain JS, then you more than likely will. In other words, the longer you leave OmniWeb open for, the more browsing you do, the more sites you open, the more likely you are to experience the bug (and it can happen on any JS containing page, not just this one).