When C-R is hit, or View-Reload, the browser should confirm that the user will lose information if they continue. This happens, in the least, with Cache-Control: no-store flag in the server's response header. In order to remain compliant with RFC 2616, the content must be deleted upon a refresh. However, in the event that the user has entered information into the browser, the loss of their input, without question or hesitation, is somewhat ... evil ... as well as a source of enormous frustration for me, personally. I believe it should be addressed, and I believe it to be a relatively simple fix. See Bug 133946 for more details on my embitterment. :)
How does one tell whether the user has entered data on the page? Or do we want to trigger this for all no-store pages?
------- Additional Comment #1 From Boris Zbarsky (out of town till mid-April) 2002-04-02 17:49 ------- How does one tell whether the user has entered data on the page? Or do we want to trigger this for all no-store pages? -- The two apparent options that arise are: (1) page dirty flag or (2) assume dirty and trigger for all no-store pages. I prefer the page dirty flag, which can be implemented functionally by comparing all form data to the default data from the HTML/DOM. However, this is a more complex solution, with no added benefit, IMO. Prompting in all no store cases is hardly a nuisance in comparison to the extreme alternative. The issue in question is the permanent destruction of user data, which can happen with a single absolute operation, C-R, which just happens to be adjacent to C-T, the ever so popular new-tab key set.
uid is being phased out.
related: bug 48333
Reassigning obsolete bugs to their respective Seamonkey owners (i.e. nobody). If you want this fixed for Firefox, change the Product and Component accordingly and reassign back to me.
Filter "spam" on "guifeatures-nobody-20080610".