Lolu chungechunge lwabekwa kunqolobane. Uyacelwa ubuze umbuzo omusha uma udinga usizo.
Why do certain websites default to bottom of page upon opening?
When I am opening up sites like REI.com and mountainhardwear.com the web page defaults to the bottom of the page. It only happens on certain websites. I can open gmail, hotmail, amazon - and it stays at the top of the page.
All Replies (18)
Thanks everyone for responses on/to this unresolved issue. I have installed Chrome and am getting happier with it by the hour. So I am bidding Firefox au revoir until v 13 is released, and then- we'll see.
I loved my Red Cats (Blue) theme, add-ons, and customization (made it look as much like the old FF as possible, hated the silly magic button or whatever it was called). But all good things get screwed up eventually, so I'm movin' on while waiting to see if Mozilla developers can ever find and address this issue.
Same problem "solved" for me be removing custom hosts file.
Don't know if the OP is still interested, but I have been experiencing exactly the same jump-to-the-bottom problem when loading pages with Firefox 12. I notice it particularly when accessing pages on newegg.ca and imdb.com.
On another discussion of this issue somewhere, someone suggested that it might be to do with using a custom hosts file to filter ads, trackers, and malicious sites. I have for years used the MVPS hosts file to do this. I tried replacing my hosts file with a stock (essentially empty) one and then restarted Firefox. This change completely eliminated the jump-to-the-bottom behaviour.
I'm not really happy with this solution, because I like having that hosts file as a great security measure. I realize it's a little outdated as a method, but I'm not sure if Adblock really provides the same benefits. I will have to look into that.
As for why using a hosts file causes Firefox to trip all over itself like this, I have no idea. I have to agree that the user experience from Firefox was much better before they started this silly accelerated update cycle.
i use mvps hosts file also, along with many of mine own entries. that was the first area i addressed. i used a blank hosts file no change. i then reverted back to version 11, put mvps back in place with my customs entries, and the problem was resolved.
could very well be an update in v 12 that conflicts with certain servers in custom hosts.
newegg was the worst for myself. interesting !!!
Tried the same trick: renamed a blank file as HOSTS and renamed original MVPS hosts file, and voila! Problem gone. Put the original HOSTS file back, and the problem returned. So Eureka! You have found it. Nice job!
Unfortunately, that solution will NOT work for me, as I will not give up my MVPS hosts file- the protection from scammy ads and unwanted downloads is just too important. So I'll stick with Chrome until Mozilla figures out what it is in v 12 that conflicts with the MVPS hosts file.
Glad to hear that that "worked" for you too mrnews. I agree, that it's not a satisfactory solution, though. Hope Mozilla figures it out soon.
there are two beta versions of FF out there, v 13 and 14. anyone tried those ?
For those using the MVPS HOSTS file and having a problem with pages positioning incorrectly, what do you get if you try to navigate directly to a redirected host? For example, if you follow this link:
With the MVPS HOSTS file in place, my Windows 7 system (I haven't tested on a Windows XP system) returns a 504 (gateway timeout) and Firefox displays this:
"The connection to the server was reset while the page was loading."
Do you find that a web server or proxy on your computer is actually responding?
On Chrome on Win XP:
"Oops! Google Chrome could not connect to www.doubleclick.net"
That's the purpose of the HOSTS file: to block these creeps from displaying their ads/scams/spam on any web page I go to. Also blocks connection with their servers, so they can't download malware via an ad.
I have used MVPS Hosts files for YEARS- the only thing you notice is that ads in boxes come up as 404s, which actually makes me happy! They can't load- Yay!!
Re alpha & beta versions of FF 13 & 14-- given the performance of the release version of 12, I'd be loathe to trust a beta version....
Okulungisiwe ngu Mr. News
Firefox 14 is Not a Beta version as only 13.0 is currently in Beta testing as 13.0b3.
that site is doubleclick by google. its domain is not listed in the mvps list. there is one similar, but maybe it should be.
on pale moon FF v 12 , newegg site is fine and so now i have a v 12 with my hosts list
http://majorgeeks.com/ <link to Aurora build removed removed by Mod.>
i was referring to this one, but i have not checked it out, my bad
Not a beta !
Okulungisiwe ngu cor-el
Majorgeeks also fails to inform users that the Aurora branch nightlies can get checkins almost every day so there is usually a update each day as a result due to checkins.
im batting .1000 today. just ignore me
Here are the results of a few more hours of investigation.
(1) The problem pages have an <iframe> with the src set to a URL on a redirected host. Because the host isn't found, Firefox fills the iframe with an error page based on an internal script (docshell/resources/content/netError.xhtml).
(2) In some cases, these <iframe> elements are set to 0x0 so you can't see them. In other cases, e.g., the TV review page, the <iframe> is visible.
(3) Whether the <iframe> is visible or not, Firefox scrolls the page down to the <iframe>. Why?
Comparing the netError page served in Firefox 12 with the one served in Firefox 11, there is a very notable change (displayed in bold):
<button id="errorTryAgain" autocomplete="off" onclick="retryThis(this);" autofocus="true">Try Again</button>
That's the new HTML5 attribute to give focus to an element. Therefore, Firefox will give the focus to (select) the Try Again button. I believe that Firefox automatically scrolls the page to bring the focused element into view.
On pages with invisible frames, you can't see this, but in the LCD TV review above, you can see that Firefox has selected the "Try Again" button in the little frame: it has the dotted outline showing that the button has the focus.
Probably this autofocus attribute should be omitted in an iframe, since the user probably doesn't think it is the highest priority to try again to load an <iframe> that isn't already in view.
Mr. News, do you want to update your bug?
Edit: Mr. News, I added a reference to your bug on the following. Since 40 people got cc'd by my comment, I don't suggest chiming in there! 301471 – XUL error pages should have 'Try Again' button focused, eliminating the need of additional Tab pressing.
Edit: Mr. News, I added a comment on your bug with my analysis.
Okulungisiwe ngu jscher2000 - Support Volunteer
Thanks for your hard work investigating this jscher2000.
You were right, that was the bug responsible for introducing this problem. But I did a google search against bugzilla hoping to luckily stumble across a bug already filed about this. And I did. I'd post it here, but would like to avoid any additional comments on it unless anyone can report it's not working once again.
It's been fixed as of 5/3/2012 and included starting with Firefox 13 Beta 3.
Download it here and let me know if it's still busted for you guys:
The problem was discovered by other people who realized pressing the space bar to move down the page wasn't working anymore for them. Instead they noticed the space bar was focusing a iframe with a error page inside. That usually happens when a ad is blocked by the hosts file. They connected the dots, filed a bug, found a regression range, and the devs got to work and actually fixed it.
Okulungisiwe ngu NoahSUMO
Thank you all! I installed the v 13 beta just now, and the problem is resolved on all the web pages I cited on p. 2 of this thread. Nice job!! My only add-on disabled by v 13 (so far) is ColorZilla, which will hopefully be updated by the developer soon, as it's an important web design tool for me. My QuickTime plug-in also needed updating, which I am doing, and hope that resolves the issue I've had for years with embedded QuickTime files not playing and triggering a "you need this plug-in" message when I already have it...
I'm pleased that this issue was tracked down and addressed, which somewhat restores my faith in the Mozilla machine. It also had the peripheral "benefit" of introducing me to Chrome, which has its strengths and weaknesses. The browser start-up time on Firefox still seems to be the fastest between FF, IE, and Chrome.