Re: shopping cart in net.data

Pete Helgren wrote:

Larry,

I recall there were differences in the way that Firefox handled sessions and the way IE does. We had an application that would only work correctly in IE when we opened multiple sessions on the same PC because IE created unique sessions and FF reused the same session ID across browser instances.

This may NOT be what you are experiencing but I remember chasing our tails on this one.

Yes, this is a definite issue in Firefox. If you open two IE instances from the desktop, you will get two different session IDs. If you open two Firefox instances, you will get the same session ID. It's tied into the fact that all Firefox instances share the same process. That in itself is a little bit of an issue: if one Firefox sessions goes unstables, you can't cancel it without taking down *all* your FF sessions. It also means secure sites stay logged in even if you close one instance of Firefox, as long as another instance stays open.

This issue has been around for a little while. Since 2001, to be exact. And it really exemplifies one of the biggest drawbacks to the open source community, namely the fact that if nobody wants to work on something, it doesn't get done.

https://bugzilla.mozilla.org/show_bug.cgi?id=117222

Read the comments, especially those by Michiel van Leeuwen. It's an enlightening look into the development effort, especially the part that it's been marked an enhancement, even though it's clearly a basic requirement. I'm not saying that commercial software is always better; more than one BPCS modification request got changed to E status (enhancement) because someone just didn't want to do it. But the fact that a highly requested bug (some 80 duplicate bugs) has been open for six years is a very interesting statistic.

And I love Firefox...

Joe