DISQUS

Sounds like you've got a bad extension sucking RAM. I recommend a new profile and see how that goes.

http://support.mozilla.com/en-US/kb/Managing+pr...

http://dotnetperls.com/chrome-memory

"Here we saw an experiment that tested the memory usage of Firefox 3.5 and Google Chrome 3.0, along with Safari 4.0 and Opera 10 (some pre-release versions). The memory watcher program that recorded and accumulated memory every three seconds over 19.2 minutes of intense browsing in four browsers at semi-random intervals reported figures that place Firefox 3.5 in the clear lead in memory usage. Firefox 3.5 showed the best memory efficiency in the average, maximum and final measurements."

http://dotnetperls.com/chrome-memory

"Here we saw an experiment that tested the memory usage of Firefox 3.5 and Google Chrome 3.0, along with Safari 4.0 and Opera 10 (some pre-release versions). The memory watcher program that recorded and accumulated memory every three seconds over 19.2 minutes of intense browsing in four browsers at semi-random intervals reported figures that place Firefox 3.5 in the clear lead in memory usage. Firefox 3.5 showed the best memory efficiency in the average, maximum and final measurements."

http://dotnetperls.com/chrome-memory

"Here we saw an experiment that tested the memory usage of Firefox 3.5 and Google Chrome 3.0, along with Safari 4.0 and Opera 10 (some pre-release versions). The memory watcher program that recorded and accumulated memory every three seconds over 19.2 minutes of intense browsing in four browsers at semi-random intervals reported figures that place Firefox 3.5 in the clear lead in memory usage. Firefox 3.5 showed the best memory efficiency in the average, maximum and final measurements."

Sam, I've been watching these trends for a number of years and have access to the actual daily number of Firefox users so I can add some insight here.

All browser usage, (all of it, really, for each browser) fall _dramatically_ on weekends. The number of people going online with Firefox falls almost 25% on weekends and holidays.

How does that square with Firefox's share going up on weekends and IE's share going down? Remember first that share is from a consistent 100%. Even when the absolute totals rise or drop, the total share is always 100%. So, if Firefox's absolute user base falls by 25% on weekends, yet its share grows by several percent, the only conclusion is that IE's absolute user base must fall more than 25% on weekends.

But, it's even a bit more complex than this. If you look at browser _versions_, you'll see that IE 7 and IE 8, along with all versions of Firefox, increase share on weekends.

So, what's actually happening is that IE 6 is falling very very steeply in absolute terms on the weekends and so all other browsers, as a percentage of total usage, are rising.

The big conclusion from all of this is that IE 6 is either more heavily used during weekdays (at work) or much less heavily used on weekends (from home).

If you could pull IE 6 out of the entire system, you'd see that the usage share of all of the other browsers was actually pretty consistent without dips or rises on weekdays or weekends. It's the one strange outlier, IE 6, which holds enough share and absolute users, to have a rather dramatic influence on the entire system.

People aren't choosing Safari. Safari, like IE, gets all of its share by shipping as the bundled browser with its operating system. Growth of Safari users trends precisely with the growth of Mac. Actually, Firefox is slowly increasing its share on Mac so Safari is actually bleeding users like IE, thought not quite as quickly.

Hey Dave, thanks for posting this. I've digged it.

http://digg.com/software/Delicious_Fix_for_the_...

- A

I can see from the screenshot that you have at least three extensions installed. I suspect AdBlock is the one causing you problems at Flickr and Facebook (I have seen dozens of reports of Adblock being the problem at Facebook with the solution being to whitelist facebook.com in AdBlock Plus)

I'd recommend disabling all three add-ons, checking out Flickr and Facebook, then enabling them one at a time with testing in between. When the pages go wonky, you've identified the culprit extension (again, I suspect AdBlock Plus, but it's worth actually testing).

Once you've identified the offending extension you can try to fix things by either seeking an update to the extension, configuring the extension to not mess with those pages, or removing the extension.

It's very rare that Firefox itself has problems with such major sites as Facebook and Flickr because the people developing those sites are Firefox fans and users. So, if you see something that obviously broken at a smart site like Flickr or Facebook, then extensions are the first place to look for the problem.

Good luck. I hope the fix is as easy as a slight mod to the AdBlock Plus whitelist.

- A

Billy, you de-legitimize this site and the other people commenting here by turning to personal insults.

>I’m not sure I’d trust an open source platform to sync my passwords.

Um, 170+ million users trust Firefox with their passwords and it's open source. The beauty of high bit encryption is that the source code doesn't help you break the encryption.

As for alternatives, Mozilla's Weave, still in early development but quite functional for me, is a pretty good alternative and does quite a bit more than just bookmarks. If you just need bookmarks, give Foxmarks a try.

Deb Richardson had a great post on color profile support in Firefox a little while ago. See http://www.dria.org/wordpress/archives/2008/04/... for more on this very cool improvement to Firefox 3.

Josh Pyles, I disagree. The passport officer speaks specifically to identification. The lock speaks to some general idea of safety. Accuracy here is important and the lock is not just imprecise, it's misleading.

You may have found the simplest way to visually explain security, but when simple isn't meaningful or accurate, then it's probably not a great idea.

You can probably restore them by following these instructions:
http://support.mozilla.com/en-US/kb/Lost+Bookmarks

- A

SSL just tells you that the connection between you and that website will be encrypted. It doesn't tell you much about the actual website you're connected to. EvilSite.com can get an SSL certificate for less than $100 with little or not background check. With EV Certs (extended validation certificates) the site must go through a more substantial audit and that audit should give you more confidence that you're connecting to the real PayPal, for example, and not PayyPall (a fictional bad guy site.)

For EV Certs, the issuers must pass an independent audit and they must all follow the same guidelines when issuing an EV Cert:

* Establish the legal identity as well as the operational and physical presence of website owner;
* Establish that the applicant is the domain name owner or has exclusive control over the domain name; and
* Confirm the identity and authority of the individuals acting for the website owner, and that documents pertaining to legal obligations are signed by an authorised officer.

This should make it more difficult for the bad guys and give users more information about those who do get issued an EV Cert (their physical address, for example.)

The other issue that's got PayPal concerned (and many others, including Mozilla) is phishing. Firefox has a built in Phishing Protection feature that warns you when you've ended up on a site known to be a phishing site. This is another way that you can know you're at the real PayPal and not PayyPall. IE 7 has a some protection against phishing too.

Safari has many great attributes, but helping users stay safe on the Web of 2008 isn't at the top of that list and I hope they release an update soon that has both EV Certs and some form of phishing protection. They're the third most popular browser and with a user base in the millions, they've got a real responsibility to stay competitive with the leading browsers.

Firefox and IE have both stepped up on this and so should Safari.

- A

SSL just tells you that the connection between you and that website will be encrypted. It doesn't tell you much about the actual website you're connected to. EvilSite.com can get an SSL certificate for less than $100 with little or not background check. With EV Certs (extended validation certificates) the site must go through a more substantial audit and that audit should give you more confidence that you're connecting to the real PayPal, for example, and not PayyPall (a fictional bad guy site.)

For EV Certs, the issuers must pass an independent audit and they must all follow the same guidelines when issuing an EV Cert:

* Establish the legal identity as well as the operational and physical presence of website owner;
* Establish that the applicant is the domain name owner or has exclusive control over the domain name; and
* Confirm the identity and authority of the individuals acting for the website owner, and that documents pertaining to legal obligations are signed by an authorised officer.

This should make it more difficult for the bad guys and give users more information about those who do get issued an EV Cert (their physical address, for example.)

The other issue that's got PayPal concerned (and many others, including Mozilla) is phishing. Firefox has a built in Phishing Protection feature that warns you when you've ended up on a site known to be a phishing site. This is another way that you can know you're at the real PayPal and not PayyPall. IE 7 has a some protection against phishing too.

Safari has many great attributes, but helping users stay safe on the Web of 2008 isn't at the top of that list and I hope they release an update soon that has both EV Certs and some form of phishing protection. They're the third most popular browser and with a user base in the millions, they've got a real responsibility to stay competitive with the leading browsers.

Firefox and IE have both stepped up on this and so should Safari.

- A

I think this feature is so awesomely awesome that it alone would warrant a full version rev. I can't wait for two hundred million people to experience the awesomeness of the AwesomeBar this year.

Can you please update the Firefox logo you're using here to the one that Firefox actually shipped with? The logo you're using was a draft version that was replaced almost four years ago, well before even Firefox 1.0 shipped.

You can find the current logo at the Mozilla website or you can get it from my weblog where I've posted several versions at different formats and dimensions.

Thanks,

- A

http://weblogs.mozillazine.org/asa/archives/200...

We're on it.

https://bugzilla.mozilla.org/show_bug.cgi?id=42...

It's not hitting most users but it's definitely odd feeling when it does.

You can work around the problem by doing View -> Toolbars -> Customize and clicking "restore default set"

That's a mac system wide setting you must adjust if you want to tab through _all_ elements on the page.

the style applied to focused and unfocused windows is going to be fixed for FIrefox 3.

https://bugzilla.mozilla.org/show_bug.cgi?id=40...

@havoc said
>If Firefox popped up an update dialog giving users the
>option to download Thunderbird, nobody on this blog
>would say boo.

Havoc, Mozilla would never include an offer for a new Mozilla product in a security and stability update for Firefox. It's that simple. Our update mechanism is for, you know, updates. New software offers are not updates. Software update mechanisms should be for, um, updating Software, not advertising new software offers.

If Apple had a second tab in their utility or had otherwise made a very clear distinction been "these are critical security updates for your existing installed programs." and "these are some other programs we offer that you might be interested in" then no one would be calling foul.

It's the dilution of *update*, which to date has meant "you really need this because without it, your installed software is not safe," that makes this such a problem (and not just to Mozilla but to every software vendor trying to keep their users safe through software updating mechanisms.)

Even un-checking the safari box wouldn't be sufficient to correct their misstep. Apple needs to move it's new program offers to a separate space, one that users will understand is fully distinct from updates to existing installed programs.

When my neighbor, who happens to spend a lot less time reading tech news and blogs than you and me, and so doesn't have a clue what Safari is or whether or not it's already on his system, when he gets a prompt to install security updates to QuickTime and the update also comes with this Safari thing, he's going to OK it because he's worried that not doing so will affect the security of his computer.

This blurring of the line between updates and new software offers is just wrong, and it's dangerous. That other vendors like Java and Adobe have decided that their software update systems are a wonderful revenue opportunity does not make it right for Apple to do the same. All of these vendors that are abusing user trust and making security an even more difficult concept for users to manage should be called out and shamed into better behavior.

If Apple had used its installed software to launch giant "Have we got something for you you're gonna just love" new software installation offers, that would be something entirely different (and probably quite a bit less effective) and something I would probably just laugh off. But they didn't. They intentionally slipped a new program into the security and stability update mechanism for existing programs. That's wrong and they should change.

- A

Oh, I forgot to say that Firefox pre-releases developer builds have been passing* the ACID2 test since April 11, 2006 (yes, 2006). The first evidence that we have of IE passing ACID2 came less than 3 months ago**.


* http://www.flickr.com/photos/dbaron/126886608/
** http://blogs.msdn.com/ie/archive/2007/12/19/int...

Comparing a beta version of IE with a version of Firefox that's been shipping for a long time seems a bit unfair. Compared to Firefox's beta, you won't find any IE lead on Acid2 or Acid3 (where Firefox's beta crushes IE's beta.)

- A

https://bugzilla.mozilla.org/show_bug.cgi?id=39...

- A

While I share some of your thinking on this, there are several places where I'd definitely like today's online apps to work offline. Reading webmail offline, composing a blog post offline, doing some quick tabulations in a spreadsheet offline, all examples where having the tool itself, not necessarily all of my data, offline would be really nice.