> Hmm. I'm not sure I see this as a bug. What about a different scenario?
> 1. You are editing a page.
> 2. You realize you didn't mean to erase those two paragraphs.
> 3. So, you hit the attach button to cancel the edit.
But that's not what the attach button means. For me it means that it is
something I can do in parallel....
mmmmm what about help, changes.... ? I have been using IE and did not
noticed this. I have a different view on how it works.
Would it make sense to notify people before going to the possible
scenario of loosing changes?
> 4. You hit the edit button and you come back to the badly modified one.
I am not suggesting attach should save (before going to the attach
page), or pages should be cached. I was suggesting that "attach" should
be disabled when editing. But the same should be for all the other
navigation buttons...
I am not sure.
> This isn't great either. I wonder if this is a serious problem (i.e.
> screws people up constantly) or it is one that goes away (i.e. after
> having your saved text gone the first time, you don't make that mistake
> again).
You are right. I should have learned already ;)
> One of the problems with fixing this scenario is that IE and Netscape work
> differently. Right now, I am going with the concept that no swiki page
> should be cached, because it could always have changed. Netscape conforms
> to that well. IE, on the other hand, caches pages when people hit back no
> matter whether the server tells it that it should not be cached. So, is
> there a solution that would make everyone happy. Unfortunately, no.
I agree.
Saludos,
-casco
