Firefox 3 Beta 5 Released

bunratty writes "Firefox 3 Beta 5 was released today. This last beta release sports performance-boosting improved connection parallelism. Not only has 'the memory leak' been fixed: Firefox now uses less memory than other browsers. This is not only according to Mozilla developers, but CyberNet and The Browser World as well. As for the Acid3 test, Firefox 3 Beta 5 scores only 71/100 compared to 75/100 for Safari 3.1 and 79/100 for the latest Opera 9.5 snapshot. The final release of Firefox 3 is expected in June."

Comments

Awesomebar?

[Anonymous Coward]

Did they get rid of the hideous awesombar yet?

Re:Awesomebar?

[lpangelrob]

It's not hideous - sometimes I only remember titles of pages, and other times only the last parts of the URL. The fact that remembering those things counts for something in Firefox (and gets me to my destination faster) makes me far more likely to use it, both here at work on Win2k and at home on my Macs.

Re:Awesomebar?

[jellomizer]

Hint... They are called Bookmarks and History.
Besides anything called Awesombar makes me shiver.

How to Enter Into Firefox.
Click on the RadicalButton view threw the CoolMenu and once the narleyhighlight is set click it and firefox will load and now you can use the Awesombar to browse the web.

Re:Awesomebar?

[Martin Blank]

It's spelled "gnarlyhilite."

Re:Awesomebar?

[lpangelrob]

Yeah, I know. I never used them.

I only use the bookmarks on the bookmark menu. I never open a sidebar or go into the separate bookmarks panel except to organize the bookmarks - a rarity indeed.

Same thing with history. It takes too long. I could have googled for it faster. The interface isn't slow, per se. I've never worked that way, and don't feel like starting anytime soon.

Now if I jump back to wikipedia, I don't have to type "http://en.wikipedia.org/wiki/Ha..." ... I can just type "Ha..." and based on my usage patterns it *knows* I want to go back there. That's smart.

Not perfect. Smart. People like using the Windows CMD+R command bar and launch bars for the exact same reason.

Re:Awesomebar?

[rsborg]

I <3 me some Awesomebar.

Seriously, on OSX, Webkit nightly (Safari) is so much better than FF3B5 (Firefox). Faster, better render, better integration.
Only thing keeping me from Webkit completely is 1) Extensions (Adblock+, Google Gears, Firebug!) and 2) Awesomebar
It's that nice.

All you haters can use a theme that kicks it.

Re:Awesomebar?

[Richard_at_work]

Fine, some people like the 'awesomebar' - a lot, however, don't. A way to turn it off completely would definitely be appreciated, being forced into using it is not.

Re:Awesomebar?

[Anonymous Coward]

It still looks hideous. What do you do when you only remember the title of something on the page but not in the URL or the title? Firefox is of no help to you now. It is far better to set up a full cache of your browsing history and index it. Then you can search on title, url, or anything in the page.

Re:Awesomebar?

[TobyWong]

I'm the same as you. I either flat out remember the url or google for it. I just glanced at my bookmarks now and it's full of junk I put in there "just in case" but never actually used again.

Mind you I usually have 20 - 40 tabs open in firefox all the time and I just resume my session on startup. It's just a different way of browsing and one that I prefer.

Re:Awesomebar?

[SatanicPuppy]

That's fine, but the massive text and the site name stacking crap annoy me to bits, as does the fact that it stores even more useless crap than the old version did.

I think they could make everyone happy by just allowing some damn customization...I seriously don't need site names in my history, and it clutters up the damn dropdown.

Re:Awesomebar?

[MooseMuffin]

I think the behavior of the awesomebar is great, I just don't like how big it is. oldbar takes care of that though.

Re:Awesomebar?

[propanol]

Agreed. I can't help but feel the new algorithm that implements searching bookmarks/page titles/etc. for results when you type in the address bar is aimed at the "I am incompetent when it comes to technical things and don't understand the concept of URLs"-type people; the like to whom the Internet is the blue IE logo on their desktops.

URLs are the key to http IMO - they're the ones to keep in memory as they're unique, unlike page titles and bookmarks. When I type "sla" in the address bar, I want slashdot.org, not some random blog post with the term 'slashdot' in the title I happened to pass by at some point.

At the end, what pisses me off the most about this whole deal is not being able to revert to the old behavior. That kind of forced nurturing is what I'd expect from Microsoft, not Mozilla.

Re:Awesomebar?

[D Ninja]

I know you've been modded insightful, and I'm not going to necessarily disagree with that. The "Awesomebar" (meh on the name) is not for everybody. It's definitely a different way of thinking.

However, I have been using and testing Firefox 3 Betas pretty significantly. Personally, I'm very much enjoying the Awesomebar. I tend not to use bookmarks all that often - it's nicer to just start typing and, based on how I browse, the site I want to go to is usually at the very top of the list. The Awesomebar has also been helpful when I haven't been able to quite remember the site I want to go to. I start typing, and the site is usually listed somewhere near the top.

Either way, it would be cool if there was an option to shut off the Awesomebar (for those people who don't like it) - but a new way to do something does not necessarily make it hideous.

Re:Awesomebar?

[Bogtha]

Same thing with history. It takes too long.

Really? Steps you perform to find something using less-than-awesomebar:

1. Ctrl+L to get to the address bar.
2. Start typing.
3. Pick desired entry from list.

Steps you perform to find something using history:

1. Ctrl+H to get to the history sidebar.
2. Start typing.
3. Pick desired entry from list.

Re:Awesomebar?

[jwo7777777]

I thought the developers were not very interested yet in passing Acid3 tests ... they were in favor of dropping Acid ....

Thank you .... I'll be here for 37 more milliseconds....

Re:Awesomebar?

[Shining Celebi]

URLs are the key to http IMO - they're the ones to keep in memory as they're unique, unlike page titles and bookmarks. When I type "sla" in the address bar, I want slashdot.org, not some random blog post with the term 'slashdot' in the title I happened to pass by at some point.

It learns from past behavior, so once you go to Slashdot a couple of times via that method, you'll get slashdot.org. That said, I agree with you that the old functionality should be there too.

Re:Awesomebar?

[robzon]

Actually, I find the new bar very useful. I understand that it's not perfect for everyone, so an option to turn it off would be great.

Re:Awesomebar?

[RareButSeriousSideEf]

4. Close history sidebar.

I can't stand the name "awesomebar," but IMO it does have better sorting and filtering logic than the history sidebar, and its performance is a bit more nimble, so it's starting to win me over.

Re:Awesomebar?

[diegocgteleline.es]

I can't help but feel the new algorithm that implements searching bookmarks/page titles/etc. for results when you type in the address bar is aimed at the "I am incompetent when it comes to technical things and don't understand the concept of URLs"-type people; the like to whom the Internet is the blue IE logo on their desktops.

Which is why the awesomebar is going to be a big success in the Real World (outside of slashdot). You know, real people don't care about what a URL is, and I can't find a reason why they should.

I'm a geek, and I can't live without the awesomebar. You can remember a domain of a frequently visited page, but the whole URL? When I've to search an article I visited a week ago, I just have to type "slashdot" and some word from the title and the url appears. Typically I'd google to find it, now the awesomebar avoids me that. That alone makes the awesomebar worth of it. When I type "sla", the first item in the list is ALWAYS slashdot, because the awesomebar knows what pages you visit more frequently. Oh, and the favicons make easier to browse at the list of URLs than the old text list, because you can differenciate one domain from other.

Re:Awesomebar?

[MMC Monster]

When I type "sla" in the address bar, I want slashdot.org, not some random blog post with the term 'slashdot' in the title I happened to pass by at some point. The nice thing about awesombar is that the first time you select slashdot from the suggestions, it gets promoted to the first suggestion (if it wasn't already, since you likely visit it pretty often :-) )

Re:Awesomebar?

[oyenstikker]

I've been using the Awesomebar equipped betas for over a month. Sometimes the site I want is at the top of the list, sometimes it is third or fourth. It gets better for a few days, then gets worse for a few days, then gets better for a few days. Not only is it often bad, it is inconsistent.

Re:Awesomebar?

[h2okies]

it is hideous, and its sporadicly accurate at best if doing upgrade. That it works for you is great It like "ribbon" technology should be user optional and configurable.

Re:Awesomebar?

[Anonymous Coward]

who cares about ACID. Does the damn thing even work with private SSL certificates?

Re:Awesomebar?

[Richard_at_work]

Problem with that is, for one example, when you have two 'favourite' websites that get used equally - one being Ebay, and one being your banks website. When I type 'online', I expect to see my banks website URL as the first choice (as it starts with 'online'), and yet the 'awesomebar' persists in putting Ebay as the first choice, because its the 'worlds online market place'.

I have *never* chosen Ebay in that instance, and yet it persists as the top choice in the list. Precisely the sort of behaviour that we are talking about.

Re:Awesomebar?

[Gewalt]

Mind you I usually have 20 - 40 tabs open in firefox all the time and I just resume my session on startup.

Please help me understand... WHY THE FSCK would you EVER want to do that? I don't mean this as a troll, so sorry for offending, I just can't fathom any reason why anyone would ever want to do that to themselves... It takes seconds to open a new tab and navigate to the page. It takes much longer than that to find the tab you want in the 40 some odd tabs you have open at the time. I'm all for tabs, they are awesome. But what do you gain by using them in that way? It just completely /boggles.

Re:Awesomebar?

[FreakinSyco]

Highlight the offending entry and press delete. Issue resolved.

Re:Awesomebar?

[Firehed]

URLs are the key to http IMO - they're the ones to keep in memory as they're unique, unlike page titles and bookmarks. When I type "sla" in the address bar, I want slashdot.org, not some random blog post with the term 'slashdot' in the title I happened to pass by at some point.

Great in concept, poor in practice. While all the blogs out there have pretty, meaningful URLs, most dynamic sites still have all sorts of nonsensical querystring crap that's not too helpful for finding a page. If http://tech.slashdot.org/comments.pl?sid=509004&op=Reply&threshold=1&commentsort=0&mode=thread&pid=22943372 means a whole lot to you (rather than, for example, http://tech.slashdot.org/2008/04/02/firefox-3-beta-5-released/reply/), you're a lucky man.

It does learn quickly, though it's a bit annoying at first. I certainly agree about forcing the new behavior - I expect (read: hope) that this will be a toggle in the app preferences come the final release. You do have to remember that this IS beta software, so it could well be on a to-do list somewhere. The good news is that it takes both urls and titles quite happily, so it doesn't REALLY matter.

Re:Awesomebar?

[D Ninja]

I can't say I have seen that issue, but if you are having that problem, I'd recommend taking note of how you're using it and provide feedback to the developers. My guess is they don't intend for a site that you choose frequently off the Awesomebar to drop farther in the list. They may need to tweak the algorithm.

Re:Awesomebar?

[opus]

Yes, too many entries and two lines each, with the site icon making them look staggered. I simply couldn't see anything useful at a glance.

The oldbar addon gets you back to a clean list: https://addons.mozilla.org/en-US/firefox/addon/6227

I'm able to enjoy the feature now, and I find it useful. This mode should be configurable, as well as reverting to a "dumb" URL text search if that suits your habits. Otherwise, this annoyance has the potential to drive away users, because every time you type a URL the awesomebar will assault you.

Re:Awesomebar?

[saibot834]

Being a Wikipedian myself, I looked for some extension to let me go directly to a Wikipedia article, and I eventually found it: https://addons.mozilla.org/en-US/firefox/addon/443
The way I configured this extension, you can just enter some lemma in the address bar and then Ctrl-Enter takes you to the Wikipedia article. It is quite useful because you don't have to use the mouse to go to the Google/Yahoo/Wikipedia-field. And if the article does not exist, it goes to the site anyway and doesn't redirect to the Wikipedia search (which I find somewhat annoying).

Re:Awesomebar?

[Anonymous Coward]

Mind you I usually have 20 - 40 tabs open in firefox all the time and I just resume my session on startup.

Please help me understand... WHY THE FSCK would you EVER want to do that?

I take it you mean the tab thingie, not the resumed session part of the speech (sessions I do more dislike than tabs, personally).

It takes seconds to open a new tab and navigate to the page. It takes much longer than that to find the tab you want in the 40 some odd tabs you have open at the time.

I never search for a particular tab in my 10-50 opened tabs, they become opened at some time during browsing, then I simply read the web pages one after another and close everything I think I won't need any longer. (Remark: Without extensions like Tab Mix Plus tabs are useless for me, indeed--I need new tabs opened from one of my many tabs being placed directly next to the opening tab.)

I'm all for tabs, they are awesome. But what do you gain by using them in that way?

Sorry for my slow comprehension skills--but what are YOU using tabs for anyway if you DON'T use them in the way I tried to describe? ;)

Re:Awesomebar?

[junner518]

I use the "%s" feature of bookmarks so i can google and search wikipedia, etc. from the address/awesome bar. It really cuts down on the time it takes to navigate: a quick Alt+D and i am there.

Re:Awesomebar?

[Random Destruction]

I use mouse gestures, so right click + mousewheel rolls through all the open tabs.

Now I wouldn't want to run 40+ tabs regularly, but it isn't too inconvenient when I do happen to have an obscene number open.

Re:Awesomebar?

[IdahoEv]

Yep. I used to be able to (for example) reliably use just the letters "en" to bring up a wikipedia url, which I could then edit to go to the page I wanted, even if it was one I had never visited before. That doesn't work at all anymore since "en" matches thousands of URLs mid-word.

And since the new list has two lines per entry, it's MUCH slower to scan through visually. Highly annoying.

Re:Awesomebar?

[0xABADC0DA]

Thank you .... I'll be here for 37 more milliseconds.... It's going to be hard to collect on your 15 minutes of fame that way...

Re:Awesomebar?

[scragz]

Or check out bookmark keywords. I have it setup for "w something" would do a wikipedia search for something, g is google, az is amazon, etc. Also works with delicious and the delicious extension (adds shortcut:w, etc. as a tag in delicious). I use these exclusively and get rid of the search bar.

Re:Awesomebar?

[luserSPAZ]

It's just an informal name, not the actual name for the location bar. Are we not allowed to nickname things?

Re:Awesomebar?

[mattmcm]

Nothing to do with the Awesomebar, but - if you want a quick Wikipedia lookup (or almost any GET-search site, really), you can just do this:

1) Create a new bookmark, name it.
2) In location, put http://en.wikipedia.org/wiki/%25s (/. mangled that - it's a percentage sign followed by s)
3) In keyword, put something short. I use "w" for wikipedia.
4) Save the bookmark, then type "w [something]" into the address bar.
5) ???
6) Profit!

But hey, to each their own. If your way works, that's fine too :)

Re:Awesomebar?

[holloway]

Like 'FreakinSyco' says you can hover your mouse over the drop-down-list of options and press the delete button. This works for google search too, if you want to remove any search terms that, er, people shouldn't see.

Re:Awesomebar?

[luserSPAZ]

I just right click in the search field on the front page, and use "add a keyword for this search". Then I just type "wiki whatever" in my location bar.

Re:Awesomebar?

[luserSPAZ]

The actual appearance got tweaked a bit in beta 5, the fonts are smaller and the colors more muted. As for your second point, are you arguing that because a full-text index of your browsing history is better, that nothing else is an improvement? I think that's bogus. The awesomebar is useful to me every day. Would Google Desktop be more useful? Maybe, but at that point I can just search Google itself and be done with it.

Re:Awesomebar?

[Thugthrasher]

Some people do a LOT of their web-browsing with their mouse. From what I've seen (I did some repair work for a while and would often have people SHOW me what their problem was, which means I saw their habits), this is actually quite probably the majority of 'regular people.' This means that they often only have one hand on the keyboard (and one on the mouse). For those people, the horribly named Awesomebar is MUCH more convenient than browser history. They have to either move their other hand to the keyboard (which adds a step) OR click View->Sidebars->History which is adding more than one step.

Re:Awesomebar?

[Meph0]

I completely understand what you're saying. I have no idea why people want to have so many tabs open all the time. I use tabs to open multiple pages at once, then read them all and close them as I'm done with each of them. Useful with mouse gestures for news articles or porn, but why would you open dozens of pages and leave them open? You'll have to refresh them anyway if you want to use them again. Just bookmark them and open them when needed, I'd say.

Re:Awesomebar?

[c_forq]

I use Webkit almost completely, but I also like the Awesomebar. The things holding me to Webkit are the find interface, keychain integration, and the combined stop/reload button. If someone would make extensions that would enable these in Firefox I would switch in a second.

Re:Awesomebar?

[roca]

The awesomebar has one huge advantage over URL autocomplete: URL autocomplete is very poor for quickly finding one of multiple pages at the same site.

For example, I spend a lot of time in Bugzilla. With the awesomebar I can type "bug 1234" and the awesomebar will find the bug I want if I've visited it before. I can even type "bug image crash" and get a list of bugs with those words in the title that I've visited before.

I think the awesomebar is one of those things that will take people time to get used to, but then they won't be able to live without.

Re:Awesomebar?

[hostyle]

Wow. Seldom these days do I learn handy new things about the Intertubes, but a having never once right clicked on a search field I have to thank you immensely for that. Thank you very much! Its simple and it has the potential to be awesome.

Re:Awesomebar?

[oyenstikker]

I don't really see the point in tweaking the algorithm. The old behaviour was perfect.

Re:Awesomebar?

[not flu]

It takes seconds to open a new tab and navigate to the page. Dozens of seconds at least which feel like forever since you'll be waiting for pages to load for most of that time. Opening everything up in a new tab beforehand will let the browser load the pages for you before you start viewing them. This is very handy if you or the server is on limited bandwidth. Either of the two is so often the case that I open everything up in new tabs automatically these days. Finding the correct tab is rarely a problem since it takes very little time to switch between tabs and you can sort them in multiple windows by various criteria.

I've been using EDGE for internet access a lot lately and I'd often be able to swap through all my 50+ tabs before I even get the first Google results, let alone get to load the correct page. Of course I never need to do that because I more or less know where my tabs are.

Using multiple tabs also simplifies browsing as you'll never need to use the back or forward buttons, and I'd imagine opening up every potentially relevant search result in a new tab is common practice.

Re:Awesomebar?

[Mozk]

To disable Awesomebar read my earlier comment. I'd assume that there is no setting in the main options (yet) because it is not yet a finished/polished feature.

Re:Awesomebar?

[mrraven]

I'll bite... The way I work is I right click on a link while reading an article and it loads in the background while I continue to read, then I tab over to the next loaded article. It's fast and i can compare and contrast numerous articles, what's not to like? BTW I have at least 20 tabs open now, no problem on a 4 year old dual G5 tower or a 4 year old p.c. notebook. just make sure you have at least a gig of ram...

Re:Awesomebar?

[Mozk]

You can turn it off completely. Read my earlier comment. I'm guessing that there is not a setting in the main options (yet) because it is not yet a finished/polished feature.

Re:Awesomebar?

[The MAZZTer]

Lots of the changes in Firefox 3 with regard to bookmarking are in acknowledgment that the current way of bookmarking isn't as efficient as it should be so users DO go and do what you do, just google for their sites.

The star is a one-click bookmark. You can file it later if you want, or just use the "smart" bookmark features.

The awesomebar is basically a search engine for your bookmarks and history. I really don't see why people hate it. If you want to type in a URL without your pr0n sites showing up, clear your history! But seriously... you enter in a key word or key words, and all sites which have some connection with it pop up, with them intelligently ranked based on how often you visit those sites. Even if you just type in URLs you'll find as soon as you type in the "h" of "http" your most frequently typed urls you started typing with "http" in the past will appear! I used to manually type in the address to planet.mozilla.org to go there. Now I just tap h and it's right there by the top for me. The AwesomeBar is designed to make it easier to find your bookmarks and history items.

And if you don't like it... that's why we have extensions.

Re:Awesomebar?

[Anonymous Coward]

The actual appearance got tweaked a bit in beta 5, the fonts are smaller and the colors more muted.

I tried it out. I still say it looks like shit.

are you arguing that because a full-text index of your browsing history is better, that nothing else is an improvement?

No, I'm arguing that the current awesomebar is not much of an improvement, the hideous user interface actually makes it worse, and that the functionality it provides is already better provided by other solutions. Furthermore, there are improvements that could have been made that were not.

You mentioned in another comment that you use bookmark keywords. The awesomebar does not have completion for search terms used with those. That's something they could have added that would have been a real improvment.

Would Google Desktop be more useful? Maybe, but at that point I can just search Google itself and be done with it.

I wasn't talking about Google Desktop. I was talking about being able to search through everything that you have seen on the web. Google itself does not provide that functianality, and I don't think that Google Desktop does either.

Re:Awesomebar?

[prockcore]

Being a Wikipedian myself, I looked for some extension to let me go directly to a Wikipedia article, and I eventually found it

Actually, firefox3 added a new feature that lets you do this for any site.

Go to wikipedia, right click on wikipedia's search box on the left and select "Add a Keyword to this Search..." then choose "w" for your keyword. Now you can go to the address bar and type "w test" and it'll be identical to if you filled out the search box on wikipedia with "test".

Re:Awesomebar?

[John Whitley]

is aimed at the "I am incompetent when it comes to technical things and don't understand the concept of URLs"-type people; the like to whom the Internet is the blue IE logo on their desktops. No, it's aimed at people who understand and can leverage search-based interfaces. I freaking love that I can type *just the different/interesting fragment* of a recent/popular URL and typically have FF3 just dredge it up for me. Yes, there's some culture shock when you first use it... but for my purposes its been fast and rockingly useful. As for "awesomebar"... well, we all roll that "1" on the cool-naming die once and awhile.

Re:Awesomebar?

[prockcore]

I haven't found webkit to be faster. Even on webkit's own Sunspider javascript test, firefox3b5 is 30% faster than webkit nightly. (Tiger on my dual g5)

Re:Awesomebar?

[Vexorian]

I love the new address bar. I also think that people that don't like it have just rushed to their conclussion after the first hour they tried it. Cause I initially didn't see much sense on it.

Hints:

• If you want to force it to just remember a URL type "http://"
• For titles, don't type http:
• Use the star to give priorities to more important pages
• If you just want to type the URL, just do as always, it doesn't prevent you from doing so
• If it is too big for you, there's an addon to make it look like a tiny autocomplete but keep the smartness (I however prefer to be able to see both title and URL, some pages got silly URLs like "http://tech.slashdot.org/comments.pl?sid=509004&cid=22942946"
• Actually use it for a while before judging, it needs to learn your ways some time before it gets really useful

I have personally been inept at using history or bookmarks before, this thing has really made me begin to use that stuff, finding a page you last visited is easier than ever, you just need to remember something about the title or the URL or both. Using the history was always hard until this. This thing can replace bookmarks, history and address bar, so it is minimalistic in design, which means I like it. I just wish there was an icon to remove the entry from history that would then mean it has also replaced autocomplete manage:)

Re:Awesomebar?

[aj50]

The reason we hate it is because we don't use the address bar as a search engine.

We like it to autocomplete the url that we're typing so disabling it completely is a step backwards but the new behaviour seems dumb.

Example: I've typed in web, am I more likely to be looking for "xkcd - A *web*comic of ..." or "GameFAQs... Video games *web* site..", perhaps I want "Lets turn this fucking *web*site yellow" or "Rapidshare: 1-Click *Web*hosting" or maybe, just maybe, I've started typing in webmail.bath.ac.uk like I do reasonably often (but probably not as much as I visit xkcd or GameFAQs).

I admit, web is a very generic word so this is quite an extreme example but I find that when typing in urls into the address bar, the awesome bar is a lot worse at bringing up the rest of the address you're typing.

Side note: I really like the idea of an integrated search for bookmarks and history, it is more useful than I would have thought but it already exists in the history panel (which I have appear in my sidebar). If they wanted to draw attention to it, would it have killed them to integrate it into the search box and make the search box itself more of a central feature? I mean, when I want to search, I use the search bar or hit my google bookmark on the toolbar, I don't type what I'm looking for in the address bar.

Re:Awesomebar?

[PsychoSlashDot]

Like 'FreakinSyco' says you can hover your mouse over the drop-down-list of options and press the delete button. This works for google search too, if you want to remove any search terms that, er, people shouldn't see.

Sorry, but that doesn't cut it because you can't get awesomebar to ignore your bookmarks without deleting them. What this feature needs is a bunch of options all to itself. For instance, you should be able to tell it to completely stop suggesting things from your history, leaving only a} bookmarks and b} things you've explicitly typed into the address bar. Also, you should be able to "tag" your bookmarks with "keystrings". I should be able to make a bookmark for slashdot, and set a keystring of "sla" to it. If I ever type sla into the addressbar, my bookmark absolutely should be the first pick, bar none, because I have a 100% match.

The awesomebar is an interesting idea, but it's very much 1.0 in terms of feature-set and usability. For now, it's annoying enough that I'm using the oldbar add-on, and keeping my history off. Sad that I have to reach that far to make Firefox work the way I want to work. And I'm not a dev, so hitting the source code myself isn't an option, like it isn't for most users.

Re:Awesomebar?

[Loke the Dog]

That's exactly what I thought before I tried it. Really, just try it. Never close anything. I don't think there's really a limit to how many tabs you can have open. I sometimes have around ten tabs in every window and around 10 windows. That's 100 tabs right there.

But I suppose this is why some of us keep all sorts of junk on our computer desktop and others prefer to keep everything tucked away or atleast in folders on the desktop. Some of us just likes living in a mess.

Re:Awesomebar?

[Phoenix Rising]

The search priority of the Awesomebar has been fixed in 3b5, I think. I used to have to type three or four characters to get to some of my favorite sites, and now it's back to searching the URL first. YEAY!

Re:Awesomebar?

[Koiu Lpoi]

Did you know that Alt-Tabing is called "Cool Switch"ing?

Re:Awesomebar?

[JavaBasedOS]

You comment just reminded me of the 'Special' stages from Super Mario World on the SNES.

Re:Awesomebar?

[Tacvek]

Also, you should be able to "tag" your bookmarks with "keystrings". I should be able to make a bookmark for slashdot, and set a keystring of "sla" to it. If I ever type sla into the addressbar, my bookmark absolutely should be the first pick, bar none, because I have a 100% match.
. Well, if you add the keyword "sla" to a slashdot bookmark, then typing "sla" and hitting enter will load slashdot. I would agree though that if a bookmark has a keyword that is an exact match to the current text in the URL bar, it should be at the top of the list, perhaps even visibly separated by a line.

Re:Awesomebar?

[amRadioHed]

With the new address bar whichever page you go to most will be the first suggestion it gives you. If your webmail is the most frequent one then it will be first. If it's not then type one more letter.

I don't get the problem with the new bar. Sure it's bigger than before but it works quite sensibly.

Re:Awesomebar?

[Tacvek]

The awesomebar has one huge advantage over URL autocomplete: URL autocomplete is very poor for quickly finding one of multiple pages at the same site.

For example, I spend a lot of time in Bugzilla. With the awesomebar I can type "bug 1234" and the awesomebar will find the bug I want if I've visited it before. I can even type "bug image crash" and get a list of bugs with those words in the title that I've visited before.

I think the awesomebar is one of those things that will take people time to get used to, but then they won't be able to live without. Would not just a properly constructed quick search bookmark be more useful? I used to have a quick search bookmark such that "bug 87945" would load the bug. It is really easy to create such a bookmark, or an even better one. For the rest I'll assume you are talking about the bugzilla at bugzilla.mozilla.org, although the same procedujre should work for most bugzillas. Step one, create a bookmark of bugzilla.mozilla.org. Then edit its location to be "https://bugzilla.mozilla.org/buglist.cgi?quicksearch=%s". Then add the word "bug" to the keyword field. Now typing "bug 789456" will load that bug even if you have never been there. Further typing "bug image crash" will do a quick search for "image crash". That latter part is admittedly not quite as useful the so-called awsomebar as it does not limit you to only the bugs you have been to, but it is also not mutually exclusive with it either. Typing that phrase but not hitting enter will let awsomebar show you the bug you have been to. If you cant find the bug you want, just hit enter, and bugzilla will search for the bug for you.

Re:Awesomebar?

[CastrTroy]

Seriously though. Awesomebar sounds like something invented by strongbad. I'm sure it's a good feature, but it's hard to talk about it without sounding like a moron. Same thing goes for GIMP.

Re:Awesomebar?

[asa]

Tacvek, it's not the completing the bug number that's the magic here. It's that it will complete against the bug summary (which is the page title) as well. All you have to do is remember a word or two from a bug you've visited before and you can get back to it super-quick. That's the magic, not the bookmark keyword which we've all had for at least 8 or 9 years now.

- A

Re:Awesomebar?

[aj50]

Did you completely fail to read anything I said? I'll put it on one line so it doesn't get lost.

I only type urls in the url bar so it's unhelpful if it matches things which are not in the url.

That's it, that's my only problem with the awesome bar, the way it looks is fine. If it could only match at the start of the domain or sub-domain like Firefox 2 does that would be even better.

Re:Awesomebar?

[CastrTroy]

I never use history. I actually disable it because I never look in there, among other reasons. I think the major problem is that there's just too much stuff in there, and it's too hard to really find, since you can only search by page title and URL, and often times you don't even know what you're looking for. I think history should have a much more hierarchical view. It shows you a tree of how you ended up at a page. So, you search google for Foo, and then from there opened up pages A, B, and C. Those should all be child nodes of the original google search. That way, when you look in your history, you only see the starting points, and you can recall your navigation path in order to find the page.

Re:Awesomebar?

[felipekk]

Stop using online for the bank and try using the bank name...

Re:Awesomebar?

[amRadioHed]

Change your habits. Type ebay when you want to go to ebay and type "bank" or whatever when you want to go to your bank. Problem solved.

Re:Awesomebar?

[hedwards]

And it's essentially OSX only. Unless I missed the memo, the Windows version is still pretty much a waste in all respects. At least FF3 will run properly on more than just one OS.

Most of the other points are completely irrelevant as few people are going to plunk down hundreds of dollars to ditch a free web browser for a different one. Perhaps if somebody were completely split down the middle of Mac v., PC, this would make some sort of difference, but for the vast majority of people, it just isn't a realistic happening.

Re:Awesomebar?

[beav007]

It's a great way to work - I love it. I generally have 10-15 tabs open all the time, which I restore every time I restart firefox.

But I suppose this is why some of us keep all sorts of junk on our computer desktop and others prefer to keep everything tucked away or at least in folders on the desktop. Some of us just likes living in a mess.

I work like this, and I always have. I've always had a messy room, but I've always been able to find what it was I was looking for fairly easily. Every time mum made me clean my room, I couldn't find squat, despite the fact that I was the one who put everything away.

Previously, on slashdot... we discussed why messy desks are more efficient than clean ones. It's like working a stack in a computer program, except rather than being a regimental push/pop of the top item, it's a general "the stuff you use the most is closest to the top".
http://slashdot.org/article.pl?sid=07/03/21/1312258

Re:Awesomebar?

[amRadioHed]

I only type urls in the url bar so it's unhelpful if it matches things which are not in the url. Yes I read what you said, I almost exclusively type urls in the address bar also. So I ask again, what's the problem. Why do you care if a page title is added to the end of the dropdown list? In my experience, as in the example you gave, the desired page will still be the first autocomplete suggestion.

Re:Awesomebar?

[Anonymous Coward]

The awesomebar is basically a search engine for your bookmarks and history. I really don't see why people hate it.

Because bookmarks and history are not the same data.

Bookmarks are a list of URLs, ordered in some sort of way that conveys semantic information: (Geek/Slashdot, Geek/Digg, Humor/Fark, Finances/BankofSlashdot, Finances/TacoBroker, pr0n/Goatse, pr0n/midget), or even just alphabetically.

History is a cache of URLs, ordered by least-recently-used.

And if you don't like it... that's why we have extensions.

There are a lot of good extensions that add functionality that's lacking. But there's also a bunch of people building extensions to revert to old behavior.

For example, there was one called "RetroFind" that backs out FindAsYouType and the Find toolbar. (The logic was that the old Find UI popped up and could be placed outside the browser window entirely, whereas the current Find UI takes up vertical space on the screen.)

Going further back, how about the screaming (from people who were in the habit of opening up 20+ tabs in Firefox 1.x, as opposed the developers, who presumably opened up four or five tabs on a maximized screen, and who seemed bewildered at the outrage) for an about:config setting to have one red "X" to "close current tab", versus a space-consuming red "X" for every single tab on the screen.

Without speaking the merits of RetroFind, the "X" on each tab, and the inevitable TheAwesomeBarSucksBalls extension, isn't this sort of behavior is a sign that something's wrong in the UI department?

Firefox is more extensible than most products, but there sure seems to be a lot of effort going into UI features that people try for five minutes, and then immediately start looking for ways to deactivate.

Re:Awesomebar?

[ozmanjusri]

WHY THE FSCK would you EVER want to do that?

I do something like that too.

I use the "Bookmark all tabs" option to save my open tabs into sets (like [project]_Legal, [project]_Design, etc). The sets are then saved as folders on the toolbar, and when I switch projects, I just right-click the folder and select "Open all in tabs". When that project's done, I can cut and paste the whole folder(s) into an archive.

I also have a few standard sets like "News", "Funny" etc on the toolbar and open them the same way. It's fast and easy to maintain.

Re:Awesomebar?

[Anonymous Coward]

If you want to type in a URL without your pr0n sites showing up, clear your history! Are there still people who don't clear their history by default?

And does this mean that the awesomebar will basically not function if you clear your history regularly?
If so, excellent.

Re:Awesomebar?

[mybecq]

Also, you should be able to "tag" your bookmarks with "keystrings". I should be able to make a bookmark for slashdot, and set a keystring of "sla" to it. If I ever type sla into the addressbar, my bookmark absolutely should be the first pick, bar none, because I have a 100% match.

You mean something like Bookmark Keywords? They are extremely handy -- eg. I have slashdot set to "/.".
In your case you could set your online banking to "online" (or perhaps a bit more obvious "bank").

Re:Awesomebar?

[Anonymous Coward]

Change your habits. Type ebay when you want to go to ebay and type "bank" or whatever when you want to go to your bank. Problem solved.

If I wanted AOL, I wouldn't be running a beta of Firefox 3, would I?

Note to Phishers: Make sure you put the word "bank" in your phish site at 0nline.phishy1337.com, just to make sure that you get this guy who'd rather "go to AwesomeBar Keyword 'Bank'" instead of remembering that his real bank's URL has been online.notphishybank.com and in his most-recent-URL-list for the past 3 years.

Re:Awesomebar?

[mrchaotica]

But "oldbar" should be the default, and the new thing should be the extension!

Re:Awesomebar?

[Anonymous Coward]

Basically:
There are pages I visit routinely, but not directly often.
The URLs for these pages are fairly distinctive in the set of URLs (as an example, webmail.uio.no), but get completely lost when matched to the set of URLs + page titles + bookmarks + whatever.
This means that getting to some of the sites I visit is harder than it used to.
I don't benefit especially from the features of the awesomebar (it just doesn't match how I use a browser), so there's no gain and an annoying loss.

Please tell me why a net-negative feature I can't disable shouldn't annoy me?
Now, if hits where actually added to the *end* of the dropdown, I wouldn't mind it at all. That does however not seem to be the case. (Which is annoyance #2 - I like predictability. Having to check every time I type an URL if the match I want is still the top gets on my nerves.)

Re:Awesomebar?

[mrchaotica]

Or the Firefox developers (and you) could stop being stupid and realize that the address bar is just that: a bar for typing in addresses! If you want a bar to have some fancy searching going on when you type in "bank," use the fucking search bar for it!

Re:Awesomebar?

[AmigaMMC]

I agree with the horrible-bar being hideous, it needs improvement. When I type the initials of a website I expect the bar to show me that website and not something totally unrelated first. At least improve it in the order items are shown. The way it is right now it's useless to me and unfortunately I have to say it'd make the difference for me to choose IE8. Plus, most Add-Ons don't work, including Adblock Plus (shows it works but doesn't get rid of JPG most of the time and never gets rid of Flash). Sorry but if I have to be honest, I don't see a reason to stick with FF3, and I used it since the very first version.

Re:Awesomebar?

[RealGrouchy]

I know!

Every time I type "j", it brings up "jello pudding pops", and every time I type "n" it suggests "New Coke"!

- RG>

Re:Awesomebar?

[amRadioHed]

Ok, so are you prevented from just typing in addresses in the address bar? If that's all you want I fail to see what's stopping you.

Re:Awesomebar?

[Raenex]

Are we not allowed to nickname things? Sure you can, just not in a dorky way.

Re:Awesomebar?

[TaoPhoenix]

I Cleared Private Data a couple times, but those BBC feed entries that I don't want kept appearing.

Re:Awesomebar?

[Anonymous Coward]

Add your online bank url to your bookmarks and tag it with "online". It should come first then.

Re:Awesomebar?

[Wiseman1024]

"Mind you I usually have 20 - 40 tabs open in firefox all the time and I just resume my session on startup. It's just a different way of browsing and one that I prefer."

This is why we need to require licences to use computers. In other industries, you are required to undergo long training and certification processes before you can even get near to a production machine, and none of those machines are remotely as complex and have as many active parts as a computer.

Re:Awesomebar?

[Richard_at_work]

This doesn't solve the problem - deleting 'ebay.co.uk' from the list seems to delete it from all lists, so it doesn't show up when I start typing 'eb'. Not a solution.

Re:Awesomebar?

[Richard_at_work]

I have just set both of those to your recommended settings, and it does partially work - however, it leaves me with only a single suggestion when I type a URL and not all matching URLs. It does not restore the functionality of the address bar from FF2.

Re:Awesomebar?

[lloydchristmas759]

Now I wouldn't want to run 40+ tabs regularly, but it isn't too inconvenient when I do happen to have an obscene page open.
There, fixed it for ya.

Re:Awesomebar?

[Zebedeu]

Your complaint is valid, but your solution isn't (IMHO).

The propper fix for that problem isn't to throw AB out completely, but to improve it so that it learns which results to show on top.

Re:Awesomebar?

[Zebedeu]

At the end, what pisses me off the most about this whole deal is not being able to revert to the old behavior. That kind of forced nurturing is what I'd expect from Microsoft, not Mozilla. I think that keeping old code around just to please users who are afraid of change is more the kind of behaviour I would expect from MS than Mozilla.

If you like the old interface you have a few good options: look for (or code) an extension that mymics the old behaviour; don't upgrade; adapt to the new behaviour; or, if all else fails, use another browser.

Well, you could always complain and try to convince the developers to revert to the old bar, but as you admitted yourself, there are many people who like AB and it's helpful to new users.

Re:Awesomebar?

[Richard_at_work]

The fact that it doesn't behave as it used to - the behaviour has changed significantly and the work-arounds do not solve the issue completely.

Re:Awesomebar?

[TuringTest]

Aza Raskin thinks that Bookmarks and History are not the best way to access previous sites, and the address bar is.

Re:Awesomebar?

[shellbeach]

At the end, what pisses me off the most about this whole deal is not being able to revert to the old behavior. That kind of forced nurturing is what I'd expect from Microsoft, not Mozilla. about:config is your friend -- just set browser.urlbar.matchOnlyTyped to true and you're back with the old site-matching-only algorithm. It's true that about:config isn't the most friendly way for a newbie to do things, but then the newbies are the ones that will love the new urlbar in the first place ...

Personally, I'd been using an extension for FF2 called "my_urlbar.a" or something like that, which did exactly the same thing as the awesomebar does now. The sad thing is, that extension did it *without slowing down my typing*. If only they'd improve the speed of the awesomebar's response, I'd be much happier -- it's a great idea, but I'd happily sacrifice site rankings and whatever other bells and whistles there are in order to make the thing *fast*. After all, it doesn't matter how many usability boxes your interface tweaks if acceptable speed isn't one of them ...

(ps -- speaking of abominations that are ugly and slow, am I the only other person here who finds the new theme of the /. discussion system horribly, horribly wrong?? I'm hoping it's just a very late April fool's joke, 'cause it makes OMGPonies look positively tasteful in comparison ...)

Re:Awesomebar?

[shellbeach]

... and if anyone *was* wondering how to get rid of those ugly grey boxes surrounding the comments, try the following in Stylish:

li[class^="comment contain"] {
          border: 0 !important;
    }

Re:Awesomebar?

[tepples]

Hint... They are called Bookmarks and History. And that's all the awesomebar does: search your history more thoroughly than older versions did.

Re:Awesomebar?

[shellbeach]

BTW I have at least 20 tabs open now, no problem on a 4 year old dual G5 tower or a 4 year old p.c. notebook. just make sure you have at least a gig of ram... Pah, that's nothing -- I've got 55 tabs open now, and I haven't even gotten started yet.

My browser collects tabs like my sink collects dishes. (Now if only there was a "Wash all dishes" menu option ...)

Re:Awesomebar?

[Kashgarinn]

I at first didn't like the "AwesomeBar" because the first thing I did after installing the new version is to try out a web-game a friend of mine had suggested. The webgame meant repeatedly going to pages again and again, and even though I just played it for a few session at the start, it was always.. ALWAYS suggesting that piece of crap for the weeks after.

If you're going to use it, make sure you only visit the pages you use day to day, after a few days you can go check the rest of the crap.

I'd say it's a nice feature, but you need to trim useless pages it pops up with as well.

Re:Awesomebar?

[deadsquid]

So, um, you can turn it off, it's just not in the UI.

Add the pref browser.urlbar.richResults as a boolean, set it to false.

this pref is only checked at startup, so requires a restart. it'll give you the old behaviour back.

Re:Awesomebar?

[deadsquid]

you can turn it off completely.

add the pref browser.urlbar.richResults as a boolean, set it to false

Re:Awesomebar?

[nekokoneko]

Example: I've typed in web, am I more likely to be looking for "xkcd - A *web*comic of ..." or "GameFAQs... Video games *web* site..", perhaps I want "Lets turn this fucking *web*site yellow" or "Rapidshare: 1-Click *Web*hosting" or maybe, just maybe, I've started typing in webmail.bath.ac.uk like I do reasonably often (but probably not as much as I visit xkcd or GameFAQs).

I had the same impression as you did when I first started using Firefox 3, but actually unless you typed "web" when accessing xkcd, Rapidshare or GameFAQs, the address bar will quickly learn that when you type "web", you probably mean webmail.bath.ac.uk. I thought the new behavior was pretty annoying at first too, but the address bar "learns" and after 2 or 3 days using it, I find it pretty handy.

Re:Awesomebar?

[Anonymous Coward]

The propper fix for that problem isn't to throw AB out completely, but to improve it so that it learns which results to show on top.

It does. I suspect very few of the complainers gave it more than five minutes to learn their search patterns before whinging in this thread...

Re:Awesomebar?

[somersault]

I think he just means that most people would have maybe 10-20 tabs at most, having 40 articles to go through is a bit much :P

Re:Awesomebar?

[rsborg]

The things holding me to Webkit are the find interface, keychain integration, and the combined stop/reload button. If someone would make extensions that would enable these in Firefox I would switch in a second.

Touché. Fact is, since I use 1password, I don't miss the keychain integration, as this captures my passwords in a seperate keychain that integrates across browsers.

The find inteface in safari kicks ass, no doubt.

Really, I also run into a few sites that misbehave in Webkit nightly (gmail included).

What I also miss from my windows days is IETab (or even IEView)... it'd be nice if there was something similar for OSX (or Safari!).

Re:Awesomebar?

[luserSPAZ]

I'm not going to argue aesthetics with you, just wanted to reply to your Google Desktop bit. AFAIK, Google Desktop does index your browser cache, so you can search your browsing history. I tried it out once and just didn't find it all that useful.

Re:Awesomebar?

[Anonymous Coward]

The oldbar addon gets you back to a clean list: https://addons.mozilla.org/en-US/firefox/addon/6227

Unfortunately it only fixes the appearance, so you're still left with the needlessly annoying search behaviour. Hopefully soon someone will manage to write a plugin that gives us back the old, useful behaviour.

Re:Awesomebar?

[Haeleth]

How lucky we are that Nanny knows best, and won't let us risk damaging ourselves by giving us optional access to a useful feature that we have been enjoying for many years!

Re:Awesomebar?

[Anonymous Coward]

about:config is your friend -- just set browser.urlbar.matchOnlyTyped to true and you're back with the old site-matching-only algorithm.

Unfortunately not. I type in "per", expecting to get Perlmonks, and instead what pops up is www.thedailywtf.com -- do you see a "per" in that? Because I sure can't see a "per" in that.

Re:Awesomebar?

[TuringTest]

How about providing us with a more efficient feature instead of the (scientificaly proven) suboptimal one that we have been suffering for many years?

Re:Awesomebar?

[shellbeach]

No, you're right -- it's stopped working in Beta 5. (It was working correctly in Beta 4). Nevertheless, you still get less hits with it. I'm not sure how that works ...

Anyway, I agree the awesomebar is considerably less-than-awesome, but why don't you file a bug report?

FIRST POST!111

[Anonymous Coward]

I'm glad that the Acid3 test is just a side mention in this story. The recent Firefox betas look great. It needs to be said though that the WebKit builds that score 100/100 are publicly available. But it also needs to be said that there's a lot more to a web browser than its performance on a single standards test.

Re:FIRST POST!111

[Nushio]

Webkit does 100? That's nothing. The newest Opera beta does 106/106!

Re:FIRST POST!111

[asa]

You certainly didn't see Apple ship Safari 3.1 with 100 on Acid3. WebKit (more accurately Safari) are at the beginning of a development cycle. They just shipped Safari 3.1 after quite a long dev cycle and are beginning Safari 3.next (or 4?) so it makes sense that they tear into their code in a pretty aggressive way. As far as I can tell, Opera 9.5 due sometime soon also won't pass Acid3. All of this work you're seeing on Acid3 is for the _next_ release, not the current release. (where current is Firefox 3, Safari 3.1, and Opera 9.5)

Re:FIRST POST!111

[Anonymous Coward]

You let me know when somebody makes a browser that goes up to a hundred and eleven

Re:FIRST POST!111

[Schnapple]

It needs to be said though that the WebKit builds that score 100/100 are publicly available.

To techies and tinkerers, yes. The average person is doing good to install a web browser at all, much less run webkit on top of it. So until Safari 3.2 or Opera 9.5 comes out and passes Acid3 100/100, I'd say the race is pretty wide open.

Re:FIRST POST!111

[abhi_beckert]

There's nothing "long" about the dev cycle of WebKit, it's in stable development with no big features being worked on and no architectural changes in the works. Just bug fixes, performance improvements and gradual standards compliance additions. Apple will release Safari 3.1.1 or 3.2 whenever they decide there's enough improvements to be worth the hassle of making PR announcements. When they release Safari 4.0, it will almost certainly ship with Mac OS X 10.6, and there will likely be a simultaneous minor version update to Safari 3, bringing it to the same WebKit release as Safari 4.

WeKit is just the rendering engine, it's not tightly coupled with Safari the way Gecko/Firefox is. Major safari updates are always about new GUI features (RSS in 2.0, "webclip" in 3.0, etc). Better standards compliance/performance is a sideline feature for Safari, that's for the WebKit team to work on.

Firefox and IE only just now pass Acid 2 in their *development releases*. They're several months, if not years behind WebKit and Opera.

Re:FIRST POST!111

[asa]

>Firefox and IE only just now pass Acid 2 in their *development releases*.

Ah, maybe you actually investigate and learn something about this before making ridiculous assertions of fact. Firefox passed Acid2 in a "development release" (dbaron's reflow branch builds absolutely were available as "development releases") precisely two years ago and trunk builds were passing in early December of 2006.

I don't know about your definition of "just now" or your definition of "development releases" but it seems to me that you're way off on at least one of those.

- A

Re:FIRST POST!111

[Angostura]

At the risk of splitting hairs, all you have to do to use the latest build of Webkit is download it and use it as your Web browser. No special measures required. Not terribly onerous.

Re:FIRST POST!111

[mrchaotica]

I don't know about your definition of "just now" or your definition of "development releases" but it seems to me that you're way off on at least one of those.

Does any version of Firefox less than or equal to 2.0.0.13 (the current released version) pass Acid2? No? Then any builds of Firefox that do pass are development releases!

Re:FIRST POST!111

[Anonymous Coward]

If the only browser hitting 100 has 5% market share... I wouldn't worry about seeing incompatible websites any time soon. The standards test is important, but being first isn't.

As long as the browser you are using is close to the others then you won't have any problems.

IE7 on the other hand scores 12, doesn't even pass Acid 2, and has the largest market share. It's Certainly much more significant of a problem for Microsoft.

Re:FIRST POST!111

[abhi_beckert]

I did do some investigating before writing my post. According to wikipedia, FireFox 3 is the first release of FireFox to pass Acid 2, and unless I'm miss-reading the title of TFA, FireFox 3 is still in development.

Re:FIRST POST!111

[Ilgaz]

Acid 3 pass reminds me the safe way of developing. Put a perfectly running application first and optimize on it. It is just a number otherwise.

I prefer a complete application and let it get optimized on it. Opera 9.5 will have to get optimized since they have to sell same html rendering code to me, running Symbian S60 V3 Nokia E65 which has 24 MB of free RAM. ;) It runs Opera 8.65 currently even with antivirus/security/antispam (Kaspersky Mobile Security) overhead and a all feature instant messenger/VOIP client (Fring).

I was afraid that Opera guys would get into that pointless "speed race" of 2007/2008 season and forget their real focus. I have nothing against optimization, you can see my various questions about why some parts of browsers aren't using Altivec (PPC) code over the web. I am just saying, a good core first.

Almost there

[mr_da3m0n]

Now if Google could just port Google Browser Sync over...

Re:Almost there

[anaesthetica]

Try Mozilla Weave?

Re:Almost there

[mrvan]

I'm a bit worried about 'giving' google all my history, cookies, and stored passwords, protected by a PIN.

Since the PIN is the only thing you need to set up on a new computer, I don't think the data sent to google is encrypted (using a key unknown by google, ie more than https)?

I guess they don't really want my passwords, but the navigation and form history coupled to my search history... brrr... (I don't even want to imagine using gmail too)

Note: I'm not saying google is evil, I wouldn't trust anyone with that much data, and certainly not a US company with a history of complying to Chinese government demands...

Re:Almost there

[FreakinSyco]

I find syncing histories a tad overboard.

I use FoxMarks for bookmark syncing across multiple FF installs. You can also log on the their website from any internet computer and access your bookmarks without installing anything. Now thats useful.

Re:Almost there

[Anonymous Coward]

Questions 10, 11, and 12 in the Google Browser Sync FAQ imply that the PIN is used to encrypt the data before it's uploaded and that Google doesn't presently have access to the information. This might explain why changing your PIN is not currently possible.

  10. Why do I need to provide a PIN?
  11. Can I change my PIN?
  12. What's the point of encrypting my information?

Of course you still have to trust the companies whose code you are running and allowing to auto-update.

Re:Almost there

[Anonymous Coward]

I am Ron Burgundy?

Re:Almost there

[rumli]

I started using Foxmarks since Google seems to be dragging its heels with its Browser Sync extension. I'm actually finding it a lot better than Browser Sync and now I can access my bookmarks via a webpage so it's accessible from a non-Firefox browser too. It's like integrating Google Browser Sync and Google Bookmarks.

Re:Almost there

[swb]

Do you have any reason or evidence that the encryption isn't usable or that Google has some kind of unencrypted access to tabs, history and windows when they are specified to be encrypted? (Cookies and passwords appear encrypted by default and cannot be unencrypted.)

My guess is that they count on people NOT encrypting history/tabs/windows for metrics, but allow encryption of these items to promote the Google brand, etc.

I like browser sync a lot and am not worried too much about the privacy angle in the current implementation.

I would like to see better control/customization of the sync (including naming which browser wins on conflict, a "trash" folder for bookmarks to be deleted, and some force-sync options), but thusfar it's been great, especially the history sync which lets me get to sites I browsed on whole other computer but lack bookmarks for.

Re:Almost there

[Anonymous Coward]

shit man, I guess we better all stop using it because your a paranoid freak. would that make you feel better?

Re:Almost there

[anaesthetica]

Flawless victory, AC. You win this round.

Re:Almost there

[salmonmoose]

Personally - I just want to sync google bookmarks with the star menu. Browser sync is OK, but if you can't use firefox, (web-cafe or something) you can't get at your bookmarks.

Acid 3 Test

[J_Meller]

I'm glad there isn't an improvement in their Acid3 score with the latest beta. It means that their release procedure is sane and they aren't introducing regressions right before a big release. Kudos to the devs for not pushing patches for the sake of it.

Re:Acid 3 Test

[Anonymous Coward]

Yeah, but are they taking it too far? I can't see any improvements in Beta-5 at all. Despite describing Mac integration as "Improved in Beta 5!", every single one of the old Mac bugs I've found (and reported) that's been around since at least Beta-2 is still there -- and some seem to have gotten slightly worse.

Dear Firefox developers: no, it does not look or act like a Mac app, and it certainly is no better than Beta-4 or Beta-3 was. Please stop advertising in your release notes that it does.

Re:Acid 3 Test

[Millennium]

Perhaps, but shame on the devs for not announcing a 3.1 release to fix Acid3-compliance as soon as possible after 3.0's release. How I long for the days when standards were a priority on that team.

Re:Acid 3 Test

[diegocgteleline.es]

In fact, there are patches implementing ACID3 features that aren't going to be merged in Firefox 3 because they're too intrusive (what, slashdotters want an example? look here: https://bugzilla.mozilla.org/show_bug.cgi?id=421765#c8)

Acid 3, just like acid 2, has been released when the firefox development cycle is focusing on stabilizing...other browsers have focused on passing acid3 like it was the most important thing to do and have done ugly things just to be the first, take for example this: https://bugzilla.mozilla.org/show_bug.cgi?id=410460#c44

And the fact that at least WebKit has introduced a special case for the Acid3
font:
m_allowFontSmoothing = (nameStr != "Ahem");

Re:Acid 3 Test

[p0tat03]

I'm a Mac user, but I haven't tried FF3 at all (WebKit for me right now). What kind of Mac bugs have you found? I want to know if these would be show-stoppers for me. And also, specifically why doesn't FF3 act like a Mac app?

Re:Acid 3 Test

[Hyppy]

BE CAREFUL! Questioning a Mac user's assertions can ruin any karma you thought you had.

Re:Acid 3 Test

[p0tat03]

Well, I just downloaded beta 5, and given it a whirl for a few minutes.

Speed and general responsiveness: Massively improved! Page loading is noticeably faster, generally feels snappier, but that might be sensory bias. I did notice, though, that large config panels (particularly in Preferences) are dog slow on first load. This may be a first-run thing, so maybe it will disappear.

Startup was a bit slow, but that may also be due to the fact that it was starting for the very first time.

The UI: Looks awesome. I have to say that it is VERY true to the Mac way of doing things (at least on Leopard). Heck, the main toolbar looks more Mac than Safari :)

Tabs are also easier to deal with than Safari. When you open a lot of tabs (like I do) Safari stacks the extra ones in a menu. Firefox allows you to scroll to the tab you want. Nice. But this has been around since FF2, so nothing new there.

I like the preferences panel now. Despite the load time it is definitely more true to the Mac way of doing things. The layout stil seems a BIT sloppy, but I'm enjoying it.

Overall verdict from a random insignificant Mac user: thumbs up. Some minor graphical bugs to fix (the search bar icon has a non-transparent background, for example). But overall a big step forward from FF2.

Re:Acid 3 Test

[Anonymous Coward]

There's no hard-and-fast show-stoppers. They're just a mountain of annoyances that remind you "this is a Windows/Linux app that happens to run on the Mac". They all seem to stem from the fact that Mozilla is/has its own GUI toolkit, which has been themed to look like a combination of every Mac OS from 10.2 to 10.5.

Some of the more obvious ones:
- popup menus look like 10.4's menus, even on 10.5
- form controls and toolbars don't look quite the same as native widgets, and don't act quite the same; usually they're just uglier and maybe require an extra click or two to work, but a couple more obscure ones are completely broken (e.g., if you choose a Text toolbar, for example)
- the zoom button in the titlebar doesn't zoom (it maximizes)
- it uses "metal" everywhere, but you can't move a window by dragging it; this means I now have to constantly think about what kind of window I'm dragging: a normal one, or a Firefox one (FF2 wasn't metal, so there was no problem)
- in some dialog boxes, they use 10.5-style tabs in some places, and 10.2-style tabs (!) in others
- lots of windows (including the About box!) have toggle-toolbar buttons which don't do anything
- some Emacs-style keybindings work, but some don't, and in many other places the keybindings are obviously Linux/Windows-style but with the focus-dotted-line invisible

Many of these aren't too bad, aside from being ugly (and reminding you that Firefox devs either don't know or don't care about Macs). For example, I only use the Preferences dialog box rarely, so the 2002-style tabs there don't really hurt anything. The really nasty ones are the places where it looks like a native control, but acts completely different. Non-draggable metal, broken keybindings, and no zooming are probably the worst.

Re:Acid 3 Test

[luserSPAZ]

Chill, Acid3 is not the be-all end-all of web standards. Why should a release be rushed out just to get a perfect score on a test? I'm sure the next (non-security) release of Firefox after 3 will score higher on Acid3. Will it get 100/100? Who knows. Should developers spend their time on things that will gain points in Acid3, but not really improve any web developer or users's lives?

Re:Acid 3 Test

[Bogtha]

And the fact that at least WebKit has introduced a special case for the Acid3 font

That's not the whole story. The Acid3 test assumes specific font-smoothing behaviour (that it doesn't increase the dimensions of the text). This is not always true on OS X and isn't required by any specification. The workaround in Webkit was to guarantee the font-smoothing behaviour that the Acid3 test expected. That font is not a normal font, it's designed specifically for testcases, so both the "bug" and the workaround would not affect normal situations. And the Acid3 test has since been changed to avoid this problem.

Please include this information when mentioning this "ugly thing", because without the pertinent facts, people assume a number of things that simply aren't true.

Re:Acid 3 Test

[luserSPAZ]

Many of these aren't too bad, aside from being ugly (and reminding you that Firefox devs either don't know or don't care about Macs).

That's a bit harsh, I'd say. I think it's more along the lines of "writing good Mac apps is Hard, and doing so without using the native toolkit directly is doubly so." Firefox has not historically had a lot of love on Mac, but that has changed quite a bit in Firefox 3. Is it perfect? Of course not. Is it a lot better? Yes. There was a lot of work to be done, and unfortunately there was only so much time to do it in the development cycle. Some things like the 10.5-style menus were just too big of a change to take at that point in the release cycle. Keep in mind that by the time 10.5 was released, Firefox 3 was just about in beta, so it became hard to make big changes at that point.

In closing, Mac users are very picky about their UI (and that's ok), but it doesn't mean that Firefox developers don't care about Mac.

Re:Acid 3 Test

[leothar]

The thing I find mosy annoying about Firefox 3 is that it's STILL not a native cocoa app. I can't use the Services menu, I can't drag around the window with the status bar and form widgets doesn't look that good. Also, the new back/forward buttons are horrible. It looks just like IE7. :( Edit: hmmm, it only looks like crap with large toolbar buttons. For me, the only valid reason to use Firefox over Safari/WebKit would be extensions.

Re:Acid 3 Test

[treuf]

Maybe you should have checked a little further, this Acid3 test has been removed/changed due to its nature (it was not a 'proper test')
And this commit in the webkit code was then as well removed ...
It's just some sensationalism from the mozilla developpers which didn't liked that race between the wekbit and the opera team.

Re:Acid 3 Test

[Toonol]

I haven't touched a mac since C programming in college (on old mac classics... off a floppy!) I don't really have strong opinions about them, pro or con. But this:

...which has been themed to look like a combination of every Mac OS from 10.2 to 10.5.
Some of the more obvious ones:
- popup menus look like 10.4's menus, even on 10.5

Does OSX change so much that the look of an application can become outdated on a point release? A 10.3 application looks different than a 10.4 application? That seems a little too innovative...

Re:Acid 3 Test

[asa]

Perhaps, but shame on the devs for not announcing a 3.1 release to fix Acid3-compliance as soon as possible after 3.0's release. How I long for the days when standards were a priority on that team.

I think you're confused. The Acid 3 test is not a test for Web standards. It's a test for a particular (and rather small) subset of Web standards. It's not even a representative set of Web standards that would necessarily move the Web forward in meaningful ways if there were compatible implementations across the various browsers.

At Mozilla, we're definitely focused on fixing bugs in our various Web standards feature implementations as well as adding new Web standards capabilities, but we're not going to focus on any one test, especially a test that's designed as much to make browser vendors jump through hoops as much to advance the standards state of the Web.

- A

Re:Acid 3 Test

[Anonymous Coward]

This rather nasty hack was removed from the Webkit tree not long after it was introduced. The reason for it was that the sub-pixel rendering (for which there is no standard) resulted in a different rendering image on Macs then from Windows PCs.

Also the Opera team are right in the middle of a stabilisation cycle (9.5 just around the corner) but have managed to develop internal builds which pass Acid 3.

Re:Acid 3 Test

[Angostura]

I think you're confused.

No, I think you are confused:

The Acid 3 test is not a test for Web standards. It's a test for a particular (and rather small) subset of Web standards

That's right it is a test for compliance for a subset of Web standards.

I think it is great that you have a clear roadmap of things that you want to get fixed before ACID 3. But claiming that compliance with any particular set of standards is not particularly meaningful sets you up for exactly the same type of "they don't take standards seriously" flak as the IE team always used to get.

ACID 3 is symbolic, and it is important to recognize that and not to simply sound grumpy about it.

Re:Acid 3 Test

[BZ]

> ACID 3 is symbolic

Only to the uninformed. The real question shouldn't be "are the right boxes checked?" but "are the standards that I want to use implemented?". ACID3 is only useful in the sense that it gets the latter to happen.

See the posts about Webkits SMIL support (which "fixes" the SMIL part of ACID3), for a good example of the difference.

Re:Acid 3 Test

[asa]

ACID 3 is symbolic, and it is important to recognize that and not to simply sound grumpy about it.

Well, I'd rather Mozilla contributors worked on issues that were real than issues that were "symbolic".

Mozilla has for 10 years, and continues today, to demonstrate a serious commitment to Web standards. For the better part of the last decade, Mozilla has been the only serious standards advocating competitor to Microsoft and Firefox over the last four years has almost single-handedly revived the standards-based Web.

So, if you think that a failure to drop everything else we're working on (to improve the Open standards-based Web) and start tap dancing for Ian Hickson and his Acid3 test erases our credibility on Web standards, then go ahead thinking that and don't expect me to waste further time trying to change your mind.

- A

Re:Acid 3 Test

[coobert]

The specific test for Ahem has been removed, as the test has been changed (it was ridiculous for it to specify how a system should antialias fonts in the first place).

http://ln.hixie.ch/?start=1206756775&count=1

Re:Acid 3 Test

[Anonymous Coward]

Read the code man..

http://ln.hixie.ch/?start=1206756775&count=1 for a full explanation and what's going on

Re:Acid 3 Test

[asa]

Also the Opera team are right in the middle of a stabilisation cycle (9.5 just around the corner) but have managed to develop internal builds which pass Acid 3..

Good for Opera. I publicly congratulated them for that work the day they hit 100%.

But what level of support are they shipping in 9.5 and at what cost of delay to the 9.5 release did they make those Opera 9.next or Opera 10 gains?

Are Opera users and web developers going to have to wait weeks or months longer to get a better Opera experience in 9.5 (which won't 100% on Acid3 but does have other major improvements, presumably usability, standards-support, security, etc.) so that Opera could win the race to 100% Acid3 in a product that users and web developers won't get to see for months or years?

Right now, the most important thing Mozilla can do to improve the Web for developers and for users is to get Firefox 3 shipped. Delaying that by splitting focus, mostly for marketing or fanboy pride reasons, doesn't seem like such a great idea to me.

- A

Re:Acid 3 Test

[Anonymous Coward]

I think it's more along the lines of "writing good Mac apps is Hard, and doing so without using the native toolkit directly is doubly so."

Granted, but I'm not nit-picking every last pixel. I've even admitted that 10.2-style tabs (6 years out of date) are not a big concern. The bigger problems are fundamental usability problems, like "looking metal but not being draggable". This one, for example, got significantly *worse* in FF3.

It would be pretty easy to revert the status-bar to a non-metal appearance. The fact that they haven't tells me they think the Mac UI is about getting the pixels right, instead of getting the interactions right.

Keep in mind that by the time 10.5 was released, Firefox 3 was just about in beta, so it became hard to make big changes at that point.

This has nothing to do with time. I have never seen any program on any release of Mac OS (9 or X) with non-draggable metal. Metal *means* draggable. To use it otherwise is like making something that looks like a button but isn't clickable. It's just plain wrong. If they couldn't make the status-bar draggable, it should never have been made to look that way in the first place.

Re:Acid 3 Test

[p0tat03]

10.3 -> 10.4 is not a point release, despite what the version number may imply. For example, 10.4 to 10.5 is along the scale of XP to Vista. There are certainly some UI differences, some more extreme than others.

Re:Acid 3 Test

[mrchaotica]

The thing I find mosy annoying about Firefox 3 is that it's STILL not a native cocoa app.

Of course it's not a native cocoa app! It's a XUL app, where XUL is Mozilla's own cross-platform widget toolkit. And it has to be a XUL app, because extensions have to be able to modify the UI and extensions are written in XUL. And extensions have to be written in XUL, because they have to be cross-platform. And Firefox has to support extensions, because otherwise it wouldn't be Firefox anymore.

Bottom line: if you want a Gecko browser that's a native cocoa app, use Camino. If you want a browser that supports extensions, use Firefox. You will never, ever be able to have a single app that does both, because XUL and cocoa are different, incompatible technologies.

Re:Acid 3 Test

[TJamieson]

Sorry, that 'special case' was valid, but Hixie got angry about it and confronted Hyatt.

See also http://ln.hixie.ch/?start=1206756775&count=1

Re:Acid 3 Test

[The End Of Days]

Fuck em. Mac users who nitpick about UI consistencies in an honest fashion would never have enough time to do anything but bitch at Apple for all the little ways they violate the expectations they themselves set. Just now they're getting around to fixing that themselves. Apparently their development model is so difficult that even they can't do it right without a decade to catch up.

Re:Acid 3 Test

[DaAdder]

Mozilla has been the only serious standards advocating competitor to Microsoft and Firefox over the last four years has almost single-handedly revived the standards-based Web. Isn't that a slight exaggeration?

Opera has been active for just as long, and these days Safari has a thing or two to contribute as well. There are others on top of that, but the point is that your statement comes off as being somewhat arrogant.

Re:Acid 3 Test

[Angostura]

So, if you think that a failure to drop everything else we're working on (to improve the Open standards-based Web) and start tap dancing for Ian Hickson and his Acid3 test

Yes, that's exactly what I was calling for - for you to drop everything. Not.

Actually, I think it this kind of intemperate posting that does more harm to your credibilit. But as you've said - you've stopped listening.

Re:Acid 3 Test

[Millennium]

I think you're confused. The Acid 3 test is not a test for Web standards. Yes, yes it is. Specs, with their inherent ambiguities and alternative interpretations, can only go so far in establishing a standard. It takes a test suite to remove the ambiguities and clear things up in an implementable fashion. That's something developers can depend on. The platform isn't HTML5/CSS3/JavaScript2: it's whatever subset of those you can be sure will work. Test suites establish those subsets.

It's a test for a particular (and rather small) subset of Web standards. It's not even a representative set of Web standards that would necessarily move the Web forward in meaningful ways if there were compatible implementations across the various browsers.

Yes, yes it would. The designers of the various Acid-series tests have always been careful to test for inconsistencies and missing features that have been annoying site developers.

At Mozilla, we're definitely focused on fixing bugs in our various Web standards feature implementations as well as adding new Web standards capabilities... "Adding new Web standards capabilities" I'll grant you. Mozilla is nothing if not eager to add support for new and relatively obscure standards, even making some of its own on occasion.

But "fixing bugs in our various Web standard feature implementations"? I don't think so. I remember when Acid2 was released, just as you (in the sense of Mozilla, not personally) were going into the final stages of 1.5. That was excusable. But not only was a compliant 1.6 not forthcoming, it didn't even make it into 2.0, released a full eleven months later. And you -again, collective rather than personal- call yourselves focused on fixing standards bugs? Your actions speak otherwise, and they speak very loudly.

but we're not going to focus on any one test, especially a test that's designed as much to make browser vendors jump through hoops as much to advance the standards state of the Web. Then you have lost sight of something very important, and I look forward to the day when you remember it. Until then, I'll be on another browser whose developers haven't forgotten that the most important thing about a browser is that the site must work.

Waiting...

[ServerIrv]

It will come out of beta as soon as Ad Block Plus is updated.

Re:Waiting...

[dbcooper_nz]

With the latest development version of AdBlock Plus it works just fine (in WinXP). Populates the filter dialog box when you select "adblock image" from the context menu. Of course you have to change the max-version in the install.rdf file.

Re:Waiting...

[Anonymous Coward]

I'm waiting for them to fix the memory issues with Ad Block/Ad Block Plus. It works great if you shut it down every two hours, but after a few weeks with hundreds of tabs opened and closed, it takes up tons of memory and won't give it back.

Re:Waiting...

[Not The Real Me]

...but after a few weeks with hundreds of tabs opened and closed, it takes up tons of memory and won't give it back.

Unfortunately, I found that out when I made the mistake of upgrading to FF2 last week. CPU usage went to 100% and stayed there, plus double the memory useage of v1.5. Thankfully, http://www.oldversion.com/ has v1.5.0.12. I was able to downgrade Firefox and all is well again. Apparently, lots of opened tabs and adblock is a highly toxic combo for v2.

Re:Waiting...

[Rogue Pat]

- change the extension from the adblock xpi to .zip
- unzip the file
- find the file called install.rdf, and change maxVersion for Firefox to 3.1 or something and save the file
- zip the files you just extracted
- change the extension of the zip file back to .xpi

Done. Now you made your add-on compatible with Firefox 3.1. This way you still can have compatibility checking on (for maximum stability) but manually change the one extension you can't live without (for maximum convenience).

Re:Waiting...

[Asm-Coder]

I'm not quite sure what you are talking about... could you back up your claim, as I'm currently running Ad Block plus on the 4th beta without any problems that I can see... (and just so it is clear, I'm not using the nightly-builder tool to make it work, it updated on it's own if I recall correctly. Simple little dialog box over a week ago is so incredibly hard to remember.)

Re:Waiting...

[Per Wigren]

AdBlock+ is not updated for beta 5 yet. It works fine if you force-install it but it doesn't officially support beta 5 yet, only up to beta 4.

Re:Waiting...

[QuantumPion]

I did this and it allowed me to install adblock, but the options dialogue does not work! Hitting the options button in the extensions configuration box opens a window of zero-size with no content in it. :(

Warning: This breaks adblock!

[Anonymous Coward]

Now I have to look at ads!

Re:Warning: This breaks adblock!

[dark whole]

what if you turn off the extension checks? is i truly broken, or just not "supported" on the new beta?

Re:Warning: This breaks adblock!

[Anonymous Coward]

I have only tried Adblock up to FF3b4, wherein Adblock works, but loses a lot of functionality. All the new FF3 Betas break it in some way.

If you block an item, when the blacklist pops up the item is no longer pre-populated into the field. And, sometimes, entire websites are blocked even when they're not on the blacklist. I have had to set 3-4 whitelists entries for such sites.

Then again, I love FF3. I finally went back to FF2 because of Adblock and after a bug in FF3 where it would no longer save passwords (if you click "Remember" nothing would happen). But, FF2 is even buggier on my machine (10-20 second 100% CPU spikes when opening a new window). I'm tempted to go back to FF3b5 even with a halfway broken Adblock.

Re:Warning: This breaks adblock!

[Pearlswine]

Works for me...

Just open up about:config and add "extensions.checkCompatibility" as a Boolean set to false.

Re:Warning: This breaks adblock!

[Omestes]

From my experience with beta 4 it works fine when you turn off compatibility checking. The only broken extension I'm run into is Cookie Safe, but CS Lite fixes most of the problems (not all it seems, it still hangs but rarely enough to be hard to isolate). As said this is with beta 4, not beta 5, so your mileage my differ.

So far these betas have been surprisingly good. Once I isolated the Cookie Safe issue, I hardly break 300k of memory usage (6 hours of regular browsing). I still get some odd CPU usage spikes everyonce in a while (a little more often than with Firefox 2), but that isn't too much a deal breaker. The odd address bar has kind of grown on me, as have the IE style navigation buttons.

My only real complaint is the history/bookmarks window. Dragging and dropping between panes is... it sucks. And not having unfiled bookmarks available in a menu is also obnoxious.

Re:Warning: This breaks adblock!

[Anonymous Coward]

For some extensions, it's also necessery to set Extensions.checkUpdateSecurity to false (autohide being one - www.krickelkrackel.de/autohide/ , before suggesting FullerScreen, right click or shift+F11 with autohide with a non-maximized firefox window)

For many extensions, it's currently necessary to use dev or nightly builds, eg. for TabMix plus, adblock plus, downthemall.

What I care about

[microbee]

I want to try beta 5 out (especially after I found Tab Mix Plus is actually supported). But my main worry is how they react to bugs found in the beta. Are they continuously releases security updates for betas the same way as the official released version? Or I'd have to wait patiently for the final release which is more than 2 months away?

Also, every time I uninstall firefox 3, I could no longer click links in outlook unless I reset default browser to IE and switch back. This is very irritating.

Re:What I care about

[Dragonslicer]

Are they continuously releases security updates for betas the same way as the official released version? Or I'd have to wait patiently for the final release which is more than 2 months away? I would guess that there will be at least two release candidates between the last beta version and the final release.

Re:What I care about

[effigiate]

Also, every time I uninstall firefox 3, I could no longer click links in outlook unless I reset default browser to IE and switch back. This is very irritating.

That is how your PC works. It uses certain programs to launch certain file types. With Firefox as your default browser, your PC knows that when you click on a URL to use Firefox to launch it. When you uninstall Firefox, your PC now no longer knows what to do with those URL files. Once you tell it to use IE (by making it the default browser), then it works properly.

That's like using Winamp for MP3s, uninstalling Winamp, and then complaining that Media Player doesn't play your MP3s anymore.

CPU spike bug?

[aredubya74]

I haven't been able to find a bug on Moz Bugzilla on the behavior, but both previous betas would occasionally spike in CPU usage after a few hours' of usage, seemingly at random. Restarting the browser clears the problem. It doesn't seem to be a site-specific problem, as rebrowsing the same pages doesn't immediately trigger the spike. Anyone else seeing this? Otherwise, I've been very happy with the FF3's rendering and feature set.

Re:CPU spike bug?

[Anonymous Coward]

Spike? How long a spike?
1 Second? Don't care.
5 Seconds? Don't mind.
10 seconds? Eww, fix it already.
Until restart? That's no spike, that's being pegged.

Re:CPU spike bug?

[Rurik]

I have that problem with FF2, and it was the reason I went to FF3. I've not had it occur within FF3, but when I temporarily reverted back to 2.0 it was still there.

It would spike for about 10-20 seconds then go back to normal for a few more minutes.

Re:CPU spike bug?

[Shikaku]

Minutes. I've hit this bug before, and on my Sempron (shut up I'm in college) it knocks the computer out cold and hangs everything, and I was forced to reset the computer. It happens randomly, and repeatedly, and if you don't restart firefox it gets worst until the whole computer freezes; at first it just freezes the computer a little and then lets up but then the time it takes to make it free again increases.

Re:CPU spike bug?

[cephah]

Sounds like it could be a resource manager of Firefox that does some work in the background, makes sense since you say it occurs after a few hours use and that it disappears if you restart it (thus removing all of Firefox's allocated resources from RAM).

Re:CPU spike bug?

[Anonymous Coward]

i'll ditto that. i'm using it under os x and, although the issue has become less frequent with beta updates, it still exists. i'll have to force-quit firefox, even, as it will not quit normally. while it doesn't hang the entire operating system, it does slow it and the fans are cranked way up.

Re:CPU spike bug?

[lpangelrob]

Apparently, this was related to an outdated list of phishing sites, causing the browser to try to download updated sites in one bite. See here: http://weblogs.mozillazine.org/asa/archives/2007/11/firefox_3_beta_1.html. It hasn't happened to me in months, so I think it's been fixed.

Re:CPU spike bug?

[WBurton]

I've noticed it jumps between 0% and 50% and back again constantly. It looks like my Task Manager is mapping Earth tremors it jumps so much. It's a fairly consistent cycle, too. It doesn't stay at any one extreme for longer than a second.

Re:CPU spike bug?

[aredubya74]

As the other commenter noted, the CPU spikes to 98-99% and stays there until the browser is shut down (I don't typically have to kill the process - clicking the close button tends to work fine). As I mentioned originally, I've seen this in both b3 and b4, including once today (about an hour before I read that b5 was released). I'm now running b5, and will update if the problem recurs.

Re:CPU spike bug?

[SnEptUne]

It never happens to me. Are you using extensions? Have you tried disabling the extensions?

Re:CPU spike bug?

[obsolete1349]

I'm fairly certain it's a problem with Flash and not Firefox. I have done a pretty crude test for this: I browsed to a lot of flash video sites (pr0n and youtube and etc.) with flashblock disabled. I let the videos play and Firefox open for about two hours. My CPU was on its knees.

I repeated the same thing with flashblock enabled, left it running for several hours and never got the CPU spike.

This was Firefox 3b4, with about 15 tabs open, each one had at least one flash player in it.

Re:CPU spike bug?

[AKAImBatman]

As another poster mentioned, it's caused by Flash movies. (Usually advertisements.) Just find the tab that has a Flash movie running (like Slashdot!) and try navigating away. If you found the right tab, the CPU should drop back to nothing.

In the case of Slashdot, I have found that some of the ads are worse than others. Just refreshing the page is sometimes enough to clear it.

Re:CPU spike bug?

[tepples]

I've hit this bug before, and on my Sempron (shut up I'm in college) it knocks the computer out cold and hangs everything, and I was forced to reset the computer. I thought preventing applications from being able to do that was part of the job of an operating system. Which operating system are you using on your PC with a Sempron CPU?

Re:CPU spike bug?

[aredubya74]

I'm fairly certain it's a problem with Flash and not Firefox. I have done a pretty crude test for this: I browsed to a lot of flash video sites (pr0n and youtube and etc.) with flashblock disabled. I let the videos play and Firefox open for about two hours. My CPU was on its knees.

I repeated the same thing with flashblock enabled, left it running for several hours and never got the CPU spike.

This was Firefox 3b4, with about 15 tabs open, each one had at least one flash player in it. I'll keep an eye out for this, thanks. Wish I could mod this parent up :/

Flashblock

[I'm a banana]

Flashblock is the one, IMO.

Re:Flashblock

[nuzak]

NoScript > Flashblock. I won't even consider running firefox without NoScript.

Unfortunately I think I'll never see a FF3 version of Sage or MouseGestures.

Re:Flashblock

[Thornae]

Unfortunately I think I'll never see a FF3 version of Sage or MouseGestures.

Why is that? Not questioning your assertion, I just am curious.
Mouse gestures are one of those usability enhancements that have become second nature to me (long-time Opera user). The first extension I always install is All-In-One gestures. (And turn off the bloody mouse-trail. I know what gesture I made, I don't want you to waste time drawing it for me.)
Haven't had a look for a while at alternatives - searching just now revealed Mouse Gestures Redox and Firegestures, both of which I might give a test run.

I have to say, though, Opera's strategy of building the gesturing right into the browser engine does make them feel a bit "snappier". If I could code worth a damn, I'd hack up my own version of FireFox that did the same...

Re:Flashblock

[nuzak]

Why is that? Not questioning your assertion, I just am curious. Basically my reasoning goes "if I haven't seen an update for FF3 by now, it's probably been abandoned". Looks like Mouse Gestures Redox is actually the same code, new name. As for Sage, I'll probably just have to live with a much more feature-encrusted, clunkier, slower, and less stable RSS reader.

Got Buttons?

[shogun]

Ok this was amusing, I just upgraded from 3b4 to 3b5 and it decided to replicate the forward/back button control a few times: Screenshot. Easily fixed under customise toolbar though...

Re:Got Buttons?

[mrgavins]

Could you comment in bug 425079 and attach your localStore.rdf, as described in comment 16? It would help a lot - we have a workaround fix, but we're trying to figure out the root cause.

Re:Got Buttons?

[shogun]

Ok no problem, done.

Defaults?

[non-poster]

They changed the default values for some connection settings? What's the big deal? I've had these settings for a really long time now.

roll-eyes.

Re:Defaults?

[anaesthetica]

That's a good point, parent. Because most users know how and why to change default connection settings, especially if they're operating through a proxy, by going into about:config and manually editing variables. Obviously, for the common user, there is absolutely no need to alter default settings.

Re:Defaults?

[non-poster]

The point is that there is nothing new or notable here. Some default connection settings got tweaked that might speed up certain use cases, so... huh?

So obsessed with memory?

[Manip]

Since when did memory usage become such a big deal?

I mean Firefox has had some nasty memory leaks for the longest time and absolutely I would love to see those fixed. But it seems like this is more than just that, it seems like some big epeen contest between browsers.

Memory is perhaps the second cheapest commodity on a modern day PC after disk space. If they get too deep into this then it wouldn't surprise me to see them off-set this reduced usage with increased CPU time or disk seek times (which is destructive on a laptop).

Personally I rate browsers based on something like this:
Responsiveness > Features == Polish > CPU Usage > Memory Usage > Disk Usage

If the Firefox guys want to be No.1 in Memory Usage then perhaps I'll use a browser like Opera which focuses on Features, or one like IE 7 which is more polished than both Firefox and Opera.

Re:So obsessed with memory?

[BZ]

> Since when did memory usage become such a big deal?

Since people started doing more "wep apps" (and memory usage skyrocketed as a result) and since mobile devices started becoming a real browsing platform. RAM on those is not all that plentiful, so far.

Note that the work to reduce memory usage in Firefox has thus far led to performance improvement, most likely due to better cache coherency. There _have_ been some optimizations to reduce memory usage at the cost of more CPU usage (largely to do with how long decoded 4-bytes-per-pixel representations of images are kept in memory), but most of the memory usage improvements have been due to using a better allocator and fixing leaks. There is no "must have the smallest memory usage around" goal; as you note other considerations are at least as important.

Re:So obsessed with memory?

[Digi-John]

Responsiveness tends to be inversely proportional to CPU and memory usage.

Re:So obsessed with memory?

[anaesthetica]

Since when did memory usage become such a big deal?

I'm not sure if you recall reading the comments to any other story about Firefox on Slashdot or Digg or Ars or virtually anywhere else in the past two years, but about 90% of those comments discussed memory usage. The Firefox team is doing a good job responding to its user base. They have not, to my knowledge, had to sacrifice speed or additional features to achieve lower memory usage.

Re:So obsessed with memory?

[MBCook]

Depends on your OS. My Mac has 2 gigs, so I never have to worry about it. Even if I have Parallels open and using 768MB or 1GB, it's fine.

XP is different.

Way back in the 9x days, I used a little program called MemTurbo (by Silicon Prairie Software, I think). It made a HUGE difference in system performance. Windows was just terrible at managing memory. MemTurbo would defrag your memory so that larger chunks were available. You could trigger it yourself, or have it trigger when a certain amount of memory was used. It was especially useful before or after a big task that used tons of memory like playing a game. It made the system feel almost as good as a reboot.

It hasn't been needed, in my experience, since I switched to 2000. They offered a version, but it didn't make much difference. Windows 2000 was much better. But there are still problems.

Windows likes to swap things out to disk. It loves to swap things to disk. So what I've noticed is that I can destroy my box with MySQL pretty easily. I have 2 GB, but it doesn't matter. Running a large query, preferably one that is poorly written, will load up tons of data into memory as it goes through it's process. When the query is done, or if I abort the query, things stay dead. The disks just swap and swap.

Any sufficiently large process will do this. Having Java run on tons of data, or Eclipse (given extra memory) with a plugin with a bad memory leak.

Close things, let them go, force quit them, whatever. I can stop the MySQL server and unload it entirely from memory. Things still page in and out of disk all the time. It can go for at least an hour or two. By that point I almost always reset because it is so bad on productivity.

I've had FireFox trigger this once or twice, when I forgot to close it for a few days. I make sure to completely close FF every night, so it never swells that large and doesn't start causing problems.

Memory still matters for some of us, despite the fact that we have enough it should never matter.

It's nice they have fixed things (I hear the jemalloc, if that was the right name, was a big help). It hasn't been a problem for me, but I know how it can be for others.

Re:So obsessed with memory?

[Hannes2000]

Responsiveness is directly related to memory usage. A Browser just should not eat up a fourth (500MiB) of my memory. And think of those with lesser hardware, you insensitive clod!

Re:So obsessed with memory?

[bunratty]

Since when did memory usage become such a big deal?

Uh, you have noticed all the people crying that Firefox is a bloated memory hog, haven't you? The ones that have been demanding that Firefox use less memory? I guess you should be careful what you wish for. Personally, I agree that it's been made a much bigger deal than is necessary. I wish Firefox developers would spend more time on other fixes, as memory use is far from being an issue for me. On the other hand, it doesn't seem to have caused performance problems or a reduced feature set. It's just being more efficient with memory without much performance impact.

Re:So obsessed with memory?

[pthisis]

The ridiculous memory use is the #1 reason that it's sometimes 30 seconds between when I click and when Firefox responds. Swapping is _slooooow_, and even if you don't hit swap you're trashing the cache and killing responsiveness (which you seem to value).

Re:So obsessed with memory?

[Mattsson]

Memory is a big deal, especially memoryleaks.
For people who throw out their fully functional systems and buy new hardware all the time, it isn't a problem.
My system is a 5 year old laptop. It max out at 1GB memory, of DDR1 type which by the way is starting go become rather expensive, and regularly, firefox use a whole lot of that.
If I want to use a few other memorybloated hogs at the same time, like outlook, word or maybe some java-app, this starts to become a problem.
I totally agree that it shouldn't use less memory at the cost of significantly more disk-io or cpu-usage, but I would much rather see a contest on reducing resource-usage rather than a contest on adding useless features and acid-3 compliance.

Re:So obsessed with memory?

[geminidomino]

Tell me about it...

I've been using my Celeron 512MB box to work on some assignments for a Java course and we just got into the JSF unit.

Between Netbeans and Firefox (since you need a browser to test), I was using 700MB. The joys of constantly swapping. =P

Re:So obsessed with memory?

[vidarh]

I've been whining about memory issues with Firefox at least since 1.5, because FF slowly started getting unusable for me - I had to restart several times a day to prevent it from growing beyond 1GB-2GB and causing massive trashing. Since I installed FF 3 beta 4 a few days ago I haven't restarted it. I'm not celebrating yet, because I'm using it far less (I use Webkit for most of my browsing now almost exclusively because the FF memory leaks), but it is very promising, and if beta 5 is even better I may actually stop complaining about it.

Re:So obsessed with memory?

[youngdev]

I'd like to point out that memory is also extremely important with the use of tabs. When firefox /firebird first came out, I said "tabs are stupid just open another browser window". Nowadays it is not uncommon for me to have eclipse (for dev), tomcat (for testing), outlook (for communication) and firefox (FULL of open tabs) all up at the same time. My eclipse app is loaded with plugins and memory usage is important in all the other apps. Additionally, while researching various tutorials it is not uncommon to have ~100 tabs open at a time (mostly because I don't close anything until I am done working on a particular block of code)

Re:So obsessed with memory?

[nuzak]

I don't particularly mind high memory usage. I do very much mind continously growing memory usage. This isn't an epeen thing, this is a functionality thing. I don't give a damn how cheap memory is, no application should have to force me to buy more because of bugs.

And now that that's licked, let's see if a hanging JS still locks up every tab. Ah, hm. Anyone want to bring back "browse in separate process"?

Re:So obsessed with memory?

[p0tat03]

Memory is perhaps the second cheapest commodity on a modern day PC after disk space.

Perhaps, but keep in mind that the browser is the not the sole consumer of memory on the computer. I have generally... my calendar app, email app, iTunes, some IDE, some text editor, my torrent client, and a few terminal windows open... When I'm hardcore working I'll also have Photoshop and maybe a 3D modeling app open also.

In that case the last thing I want is my browser chewing up 1.5GB of RAM (FF2 has done that, without even me doing that much).

If I had 16GB of memory maybe I'll stop worrying about it, but even at 4GB I would still be very concerned if a single app took up over a gig. I have 2GB in this machine, and when Firefox chews up 500MB by itself it becomes a very real problem. Keep in mind also that more and more people are using laptops as exclusive machines - these are mostly not upgradeable RAM-wise, so minimizing RAM usage is key to legacy support - the expectation that the user can pop more sticks in disappears.

Re:So obsessed with memory?

[99BottlesOfBeerInMyF]

Since when did memory usage become such a big deal?

Memory is a big deal for me because I have a cheap laptop with as much RAM as it can hold. Still RAM can be the limiting factor for two reasons: First, I often have another OS running in a VM, which really puts a dent in your available RAM. Second, My OS and most user software have stabilized and suspend modes have become fast enough. Things just don't crash anymore and as a result I just don't reboot anymore. My uptime on my laptop is generally the same as a few days after the last security patch was released. Sometimes this is months.

Really I don't care if Firefox uses twice as much memory as it does during the first hour I use it. The problem comes when I open a few tabs of reference material, then leave those tabs open in the background until I get back to them. In this instance I've gone back to Firefox to kill it, because it gradually used more and more RAM until other applications were starved. The only other programs I've used that exhibited that behavior are Adobe InDesign (which is even worse than Firefox, 2 days between restarting it, tops) and MS Word (which is about the same as Firefox.

I mean Firefox has had some nasty memory leaks for the longest time and absolutely I would love to see those fixed. But it seems like this is more than just that, it seems like some big epeen contest between browsers.

I know it is uncommon in the computing industry these days, but it is called 'competition'. There are several viable alternatives for your browser of choice and each is striving to outdo the others and gain market share. Sadly, this competition does not extend to Microsoft who is still abusing their monopoly to take most of the market. The end result, however, is that among other browsers a lot of the market is savvy enough to see what memory usage is easily and hence use it as a benchmark of sorts.

...or one like IE 7 which is more polished than both Firefox and Opera.

Heh, your milage many vary, caveat emptor, this statement does not reflect the views of this station or geekdom, etc.

Re:So obsessed with memory?

[RAMMS EIN]

``Since when did memory usage become such a big deal?''

I think it has pretty much always been a big deal. Unless you have plenty of memory, memory is likely to be the limiting factor on the performance of your system. In extreme cases, memory shortage can cause programs to not work at all. Firefox has been a notorious memory hog. So I am _very_ glad to see this addressed. I might actually start using it again.

Re:So obsessed with memory?

[nebulus4]

or one like IE 7 which is more polished than both Firefox and Opera.

IE7?! Are you kidding me? You can "polish" Firefox as much as you want.

Re:So obsessed with memory?

[ergo98]

Since when did memory usage become such a big deal?

***BOGGLE***

Memory usage has been the #1 complaint about Firefox since it was called Firebird. "Wahhhhh, I tried Firefox but holy does it use a lot of memory!". Go to any clutching desperately to their copy of IE board right now -- any Microsoft centric type dev board -- and that's what you'll find recanted ad nausea, despite the fact that most of the memory usage is useful caching, and IE 7 is far worse.

So they're moving forward with the features and reducing a number in the task manager. Everyone is happy, and with more education and evangelism, maybe the morons will have to move onto a different whine to explain their irrational attachment to all things Microsoft.

Re:So obsessed with memory?

[Carnildo]

Since when did memory usage become such a big deal?

Since my desktop system with 4GB of RAM ran out.

Re:So obsessed with memory?

[dlanod]

I found the following paragraph in http://shaver.off.net/diary/2008/03/12/year-of-the-gecko/, linked off the article:

It's a time-honoured programming tradeoff that using more space speeds you up, but that's not what happened here: our memory-reduction regimen actually made us faster in a lot of cases by making us more cache-friendly and by side-effects like using a better allocator. So in this case the memory changes provided significant other performance improvements. I'm a huge fan of the memory changes because it can significantly impact performance when you have Firefox with a large memory usage while trying to use virtual machines or other apps/games that require large amounts of memory as well.

I've switched to FF3 Beta 4 on my home computer and have been amazed at the difference. It used to reach 500 MB in memory usage on FF2, but now tops out at just over 100 MB despite me using many more tabs that I previously would have in order to try and stretch it.

Re:So obsessed with memory?

[Anonymous Coward]

Less memory usage opens the gate for all those things you want. Firstly it'll help a bit with responsiveness as the program will swap less. Secondly features typically use memory, there is room for more features; in FF2 more features would impact performance as it already consumed to much memory to remain responsive from regular use, nevermind any fancy features or extensions.

I'm all for an option to use more memory for faster back/forward operations and holding specific cache items in ram, but lower memory consumption is good and what it allows is even better.

Re:So obsessed with memory?

[WuphonsReach]

Since when did memory usage become such a big deal?

Because the default heuristic in FF 1.x only worked well for systems with 512MB or less of RAM. On systems with 1GB or more, FF would assume that it could triple or quadruple its memory usage without interfering with the end-user. The big problem was that the heuristic looked at total installed RAM and not just the amount of free memory. (Going from memory here... so details are a bit fuzzy.)

All of which meant that on a 256/512MB machine, FF 1.x would stay fairly sleek at around 30-75MB of usage. But on a 1GB machine, it would pig up to 200-300MB of usage.

Those of us who install more RAM then average are doing so in order to run multiple applications. FF 1.x got directly in our way with its over-aggressive land grab on available memory. Just because I have oodles of RAM installed, doesn't mean that I want my web browser to lap it all up.

I'm a 3GB RAM system right now (WinXP) and firefox 2.x is pigging out at 150MB of RAM. And it was up to 367MB earlier today. I want a Firefox that is about 1/3 of that memory footprint.

Re:So obsessed with memory?

[try_anything]

Memory usage has been the #1 complaint about Firefox since it was called Firebird. "Wahhhhh, I tried Firefox but holy does it use a lot of memory!". Go to any clutching desperately to their copy of IE board So, you're one of the people who thought that Firefox's memory problems were overblown? Firefox 1.x memory usage sucked balls for many people, mainly because of the "leak" which caused Firefox to get bigger, and slower, and bigger, and slower, and finally crash. I put "leak" in quotations marks because I remember how strident partisans deflected every mention of the "leak" with arguments over whether it should be described as a "leak" or not, as if imprecise use of language by users was a more important issue than Firefox swelling up and bursting like the guy in Monty Python's Meaning of Life. It seems that a bunch of Firefox fans (and even some developers) decided that all of the reports were just Microsoft FUD, and dedicated themselves to shouting down users who reported problems. Incidentally, when did you develop this little line of patter:

maybe the morons will have to move onto a different whine to explain their irrational attachment to all things Microsoft I've been using Firefox all along because even in the 1.x days it was easier to restart Firefox every two or three days than to use Opera or Konqueror. The lousy period of fanboys insulting everyone who experienced memory problems with Firefox dragged on long enough. Just put a sock in it.

Re:So obsessed with memory?

[ShadowsHawk]

It's a big deal when your laptop only has 512Mb.

Outlook 131Mb
Firefox 110Mb
svchost 117Mb
McAffee 64Mb

422 just in main applications. Memory usage is a BIG deal.

Re:So obsessed with memory?

[ergo98]

So, you're one of the people who thought that Firefox's memory problems were overblown?

Yes, absolutely. Without fail it was grossly overblown. The most remarkable thing about the whole Firefox memory leak issue was how strong the anecdotes and subjective opinions were ,and how stunningly rare any actual empirical metrics were.

Did Firefox use a shitload of memory? Absolutely. As did any competitor doing anything similar.

It seems that a bunch of Firefox fans (and even some developers) decided that all of the reports were just Microsoft FUD

You realize that we have amazing tools like debuggers, or at a higher level simply process performance metrics, right? The reason most complaints were shot down is because they usually had the credibility of someone yelling about how their Q-bracelet makes them feel so much better.

The lousy period of fanboys

When you're faced with countless morons just repeating verbatim the proclamations by someone else, never putting the slightest bit of effort into actually providing credible metrics, then it deserves to be shouted down. If you believe that anyone who is anti-ignorance is a fanboy, you just keep on believing that.

I hear those evolution fanboys keep at it. God created the universe in 7 days, and those assholes should just believe it and quit shooting it down. The anti-Creationism has gone on long enough.

Re:So obsessed with memory?

[try_anything]

The most remarkable thing about the whole Firefox memory leak issue was how strong the anecdotes and subjective opinions were ,and how stunningly rare any actual empirical metrics were. Sorry, I didn't keep statistics about how many times per week Firefox crashed on me before I started preemptively restarting it. I just knew it was about twice a week. And it may not have actually crashed, since I didn't use a crashmeter. It was just my subjective impression.

You realize that we have amazing tools like debuggers, or at a higher level simply process performance metrics, right? You realize that it's the job of developers to use those things, and that the developers who used them found and fixed memory-related bugs, right? In any case, I just googled for a crash-detection utility for Linux that could distinguish between the sudden unrequested disappearance of a process and a true "crash", and I couldn't find one. So why don't you write one, and next time you're calling me deluded I'll use it.

The reason most complaints were shot down

... was that people were asshats, basically. The problem didn't happen for everybody, the bug reports were vague (because it's hard to file a precise bug report that involves opening and closing a couple hundred different web pages in browser tabs), and when somebody did reproduce the problem, they became the enemy as well. And, you know, the actual Firefox developers were (almost uniformly) much more civil than the idiots who simply wished to feel good about themselves by going to bat for an open-source cause. Some of the developers did reproduce specific problems reported by users, and some of the developers simply took users at their word (imagine that!), and they found (and fixed) quite a few real problems... and do you know what the fanboys said? "See? That proves the fucking users were wrong all along! They said 'the memory leak' and it wasn't just a single problem, some of the problems were leaks of other kinds of resources, and some of the problems couldn't properly called 'leaks' at all! Fucking ignorant shills talking about 'the memory leak'!"

Anyway, this should all be in the past. As far as I know, Firefox 2 fixed the last of the completely imaginary memory problems, and by most accounts Firefox 3 is even better than 2.

Beta/nightly vs. regular stable release

[dn15]

As for the Acid3 test, Firefox 3 Beta 5 scores only 71/100 compared to 75/100 for Safari 3.1

If we're comparing a Firefox beta then we may as well look at a newer version of Safari, too. The latest nightly builds of WebKit get 100/100 on Acid3. http://webkit.org/blog/173/

Re:Beta/nightly vs. regular stable release

[Idiot with a gun]

Right, but that's for the next version of Safari, that will use Webkit. Safari 3.1 is currently out in the wild, and Firefox 3 will be soon, so they're comparable. Whereas we won't see the version of Safari that completely passes Acid3 for a long, long time.

Re:Beta/nightly vs. regular stable release

[asa]

>As for the Acid3 test, Firefox 3 Beta 5 scores
>only 71/100 compared to 75/100 for Safari 3.1

>>If we're comparing a Firefox beta then we may
>>as well look at a newer version of Safari, too.
>>The latest nightly builds of WebKit get 100/100
>>on Acid3. http://webkit.org/blog/173/

Actually, that's not quite fair. Firefox 3 beta 5 is the final beta and it's basically done. It will be a shipping browser at the same time as Safari 3.1. Comparing shipping browsers with nearly simultaneous releases (only a few months apart) is an eminently reasonable thing to do.

- A

Re:Beta/nightly vs. regular stable release

[Anonymous Coward]

We're not comparing a Firefox nightly, thus we aren't talking about a WebKit nightly or an Opera super-unstable devel build...

Re:Beta/nightly vs. regular stable release

[Phoenix Rising]

Keyword: Beta. In Firefox terms, that means working toward release, aka stabilizing. Not the time to be rushing in with major changes to the code.

The Firefox 3 Betas have made major steps forward in memory use, rendering, and speed, and they've added significant UI functionality as well (the Awesome Bar is, well, awesome now that they have their search priorities fixed in 3b5...). I'll take a delay on Acid3 compliance in exchange for what FF3 is delivering.

Re:Beta/nightly vs. regular stable release

[Anonymous Coward]

No.

This version of Firefox is actually a beta - it's intended to be released soon with few additional changes, it's been in feature-freeze mode for months (long before Acid3 was released in fact). So Firefox got 71/100 BEFORE THE TEST WAS WRITTEN!

That version of webkit is from a nightly snapshot, in a project which is currently in "lets add as much crap as we can and clean up the mess later" mode, and which isn't intended to be shipped with an actual web browser any time in the next year or so.

Completely fucking different.

Re:Beta/nightly vs. regular stable release

[TaoPhoenix]

"Firefox 3 beta 5 is the final beta and it's basically done."

I kept having (& reporting) trouble with Yahoo Mail on all previous FF3 Betas until this new B5. Because of this, I never bothered to really use the Betas yet.

Is there a plan for a RC before the final ship, and do you know when we might expect to look for it?

3 Beta 5 vs. 2.0.0.13?

[billstewart]

Mozilla wants me to update from 2.0.0.12 to 2.0.0.13. Is there any reason I shouldn't just go to 3.0 Beta 5? I'm assuming it either fixes that security bug or replaces it with some new ones.

Are the critical extensions available? For me, that's Adblock, NoScript, and Flashblock.

Re:3 Beta 5 vs. 2.0.0.13?

[Anonymous Coward]

Adblock, NoScript, and Flashblock. Web browsin: ur doin it wrong.

Re:3 Beta 5 vs. 2.0.0.13?

[Anonymous Coward]

The main reason to go to 2.0.0.13 and not 3beta5 for production is that the latter is a beta, and the former is not.

Re:3 Beta 5 vs. 2.0.0.13?

[Dragonslicer]

Are the critical extensions available? For me, that's Adblock, NoScript, and Flashblock. Flashblock works fine for me on beta3 at home. The install.rdf file says it works with 3.0.*, so you wouldn't even need to disable version checking.

Re:3 Beta 5 vs. 2.0.0.13?

[Anonymous Coward]

No, only Noscript works with it
You could, however, download the extensions and edit the rdf file yourself and install them manually, they'll work fine then.

Re:3 Beta 5 vs. 2.0.0.13?

[o'reor]

I have Firefox 3 pre-beta5 (Minefield) release, nightly build from 12 days ago. Works flawlessly on MEPIS Linux 7.0. Adblock works OK too.

It feels more responsive and faster than Firefox 2.0.0.x. It may be nicknamed "Minefield", but still it hasn't blown my legs off since I installed it. So for me it's pretty stable. Even though it's not a production-grade release, I'd still recommend it over 2.0.0.1x, which in my experience has had a tendency to blow up way more often. Particularly on AJAX-heavy sites such as the new Yahoo!Mail. JavaScript handling is tremendously better in Firefox 3.

Re:3 Beta 5 vs. 2.0.0.13?

[oDDmON oUT]

Adblock, all iterations, are broken ATM.

Re:3 Beta 5 vs. 2.0.0.13?

[the_other_chewey]

Adblock, all iterations, are broken ATM.

Not true. Adblock Plus dev builds work fine in FF3.

Re:3 Beta 5 vs. 2.0.0.13?

[geminidomino]

And Password maker.. That's the biggie for me (with adblock)

Re:3 Beta 5 vs. 2.0.0.13?

[oDDmON oUT]

Dev builds, eh? Using the included update check, and the extensions site says otherwise. Guess I'll have to start hunting for those.

Re:3 Beta 5 vs. 2.0.0.13?

[Huwawa]

I hope it's worth the hour it will take to compile on my Gentoo box. And that's assuming the Gentoo devs decide to put it into the tree...

Re:3 Beta 5 vs. 2.0.0.13?

[the_other_chewey]

Dev builds, eh? Using the included update check, and the extensions site says otherwise.

Of course it does. You wouldn't expect dev builds to show up as updates for release versions,
would you?

Get your dev build here: http://adblockplus.org/development-builds/more-firefox-3-fixes

You might need to disable FF's internal compatibility check - which is no problem, ABP 0.7.5.3+.2008032513 works fine in FF3b5.

Re:3 Beta 5 vs. 2.0.0.13?

[oDDmON oUT]

Found Adblock Plus dev builds and attempted to install. FF3b5 returned an error saying the extension was incompatible. So, I guess I was justified in saying they were broken after all. Thanks anyway, these ads are turning my eyes to mush.

Re:3 Beta 5 vs. 2.0.0.13?

[Dragonslicer]

So I just installed 3.0b5 on my Windows box at work, and it does say that Flashblock is incompatible (I guess whatever code does the check doesn't match 3.0b5 to 3.0.*). Unfortunately, Firefox 3 doesn't install completely independently from Firefox 2 on Windows the way it does on Ubuntu (I'm sure I could hack at it until it does, but it's not really worth the effort).

Re:3 Beta 5 vs. 2.0.0.13?

[the_other_chewey]

Found Adblock Plus dev builds and attempted to install. FF3b5 returned an error saying the extension was incompatible.

The compatibility check is just a string comparison.
Let me reiterate:

You might need to disable FF's internal compatibility check - which is no problem, ABP 0.7.5.3+.2008032513 works fine in FF3b5.

I'm running it right now. Remember, you are using beta software.

Re:3 Beta 5 vs. 2.0.0.13?

[Tumbleweed]

I don't know about FF3b5 yet (just downloaded it), but I can say with some experience on both my home and work machines, and my coworkers experience on his Mac, that FF 2.0.0.13 is an unstable piece of crap, at least when upgraded from a previous version (2.0.0.12 in my case). Installing it fresh may get better results, but I dunno. Beware of the evil .13!

Re:3 Beta 5 vs. 2.0.0.13?

[Rogue Pat]

Mozilla wants me to update from 2.0.0.12 to 2.0.0.13. Is there any reason I shouldn't just go to 3.0 Beta 5?

Yes, there is a reason. It is obviously desirable that people don't get automatically upgraded from a stable release branch (2.x), to a beta of the next stable release.

By the way, in about:config, the behavior can be changed by changing the app.update.channel string. In my case i have it set to "nightly". Have a look at the mozilla knowledgebase for a complete overview of the possible update channels.

Re:3 Beta 5 vs. 2.0.0.13?

[The MAZZTer]

Adblock and NoScript work fine... I've used Adblock Plus since b4 so past b4 I can't say for SURE, but I'm sure Adblock works fine on b5 too. You might have to disable compatibility checks.

I don't use Flashblock... I find between NoScript and Adblock they pretty much block everything, including flash, twice over anyway.

Re:3 Beta 5 vs. 2.0.0.13?

[TheDreadedGMan]

It seems I cannot locate the compatiblity check setting within about:config...

however to get it working:

- download the dev version of adblockplus from http://adblockplus.org/devbuilds/ (right-click "Save Link as")
- then extract the xpi (it's actually a zip file) you'll only need to edit the install.rdf file
- alter the file install.rdf (it is an XML file... if you're using windows, use wordpad, it's easier to read with correct line ends)
- find 3.0b5pre and replace with 3.0b5
- add the altered install.rdf file back into the xpi file, or recompress the whole thing
- type in the address to the altered xpi file in firefox ( e.g. in my case file:///D:/DOWNLOADS/adblockplus-0.7.5.3+.2008032513.xpi )
- install
- ????
- PROFIT!!

Re:3 Beta 5 vs. 2.0.0.13?

[knarf]

If you already use NoScript you don't really need Flashblock as NoScript contains similar functionality. Just tell it to always block flash (also on trusted sites) and you will be presented with a flash-free page. Should you want to see the flash content you just click it - just like you would in Flashblock. The only missing part is that NoScript does not yet have a whitelist for plugin content. As flash (on Linux or also elsewhere?) is currently somewhat broken (version 9.0.115.0 of the flash plugin crashes when playing a second (or third or fourth or ... flash movie, bug has been reported many times, Adobe does not seem to be interested...) this does not bother me in the least. One less plugin means one less part to break the browser...

Re:3 Beta 5 vs. 2.0.0.13?

[jsoderba]

You couldn't find it because it isn't there. Create a new boolean called extensions.checkCompatibility and set it to false.

Re:3 Beta 5 vs. 2.0.0.13?

[oDDmON oUT]

Open about:config

Add booleans extensions.checkCompatibility & extensions.checkUpdateSecurity and set to false

Multiple restarts

Get your dev build here: http://adblockplus.org/development-builds/more-firefox-3-fixes [adblockplus.org]

Adblock Plus still fails to install. /sigh

Any thoughts?

Re:3 Beta 5 vs. 2.0.0.13?

[TheDreadedGMan]

thanks! having since read some comments and several webpages, I now relized it has to be created...

Re:3 Beta 5 vs. 2.0.0.13?

[Anonymous Coward]

adblock didn't work on beta 4 :(

Re:3 Beta 5 vs. 2.0.0.13?

[Hillgiant]

Go to about:config
Insert the name extensions.checkCompatibility
Set it to false
Restart FireFox

Re:3 Beta 5 vs. 2.0.0.13?

[terjeber]

Hey, the fact that Mozilla 3.0 Beta 5 is about as stable as a house of cards would have been during Hurricane Catrina might be a good reason not to go there yet. Mine crashes many, many, many times a day.

Out of date comparison?

[riceboy50]

Firefox 3 Beta 5 scores only 71/100 compared to 75/100 for Safari 3.1 and 79/100 for the latest Opera 9.5 snapshot I thought it was reported a few days ago that Safari was around 95% and Opera had achieved 100%? I can't be bothered to search the /. archive. :P

Re:Out of date comparison?

[The MAZZTer]

Those are probably private builds ATM. The latest public Opera scores 78 for me... I THINK I have the latest public build, at least. The site I get my download news from might not update for each build maybe.

Re:Out of date comparison?

[The MAZZTer]

Now Opera is scoring 79... I think one of the tests they use sometimes fails when it should succeed.

Hmm a few of these tests fail if they don't succeed withing a period of time (click the A for a report). That's probably it.

Re:Out of date comparison?

[hattig]

Opera, on an internal build, got 100/100 (this isn't a percentage, there are two other aspects to Acid3 - pixel perfect placement and animation smoothness).

Safari got 100/100 a day later, but in the process discovered a flaw in the Acid3 test that had to be fixed, making Opera's score 99/100. Safari is at least available in a nightly version. Apparently it also got pixel perfect placement and the animation was arguably smooth.

I don't personally think it counts until it's a full non-beta release.

Acid scores

[Fri13]

How old this info really is? Opera and Webkit has already hit 100/100p on Acid3 and Beta4 already had fixed a lots those memory leaks when comparing it to Firefox 2.x

Re:Acid scores

[compro01]

very early (pre-alpha, i believe) builds of opera and webkit have hit 100/100, and AFAIK, the opera build that does that feat isn't even publicly available. the numbers they're showing are for browsers that are actually available and usable.

Re:Acid scores

[transwarp]

The Opera build is available at http://my.opera.com/desktopteam/blog/2008/03/28/public-acid3-build I agree, though, the Acid-3 stuff won't be in a stable release of either browser for quite a while.

Re:Acid scores

[Fri13]

Yah, and Webkit was available right after text passed, but from SVN what many users, not just developers can easily use, I got webkit from SVN to test and it was my second time i have used SVN.

Re:Who cares about Safari?

[anaesthetica]

Because you care about competition. Once you stop caring about competition, you get sideswiped just like IE has been by Firefox. The whole idea is to have a plural browser environment in which each browser vendor competes to deliver the best standards compliance and the best feature set. If you only care about Firefox, you may be missing the point. We can measure Firefox's progress objectively (against its own past performance), but we also need to assess its progress relative to other browsers so that we can assure it remains competitive, and can (at the very least) hold its ground in market share. No one wants to return to the old days of browser monoculture and stagnation.

How the mighty have fallen.

[lancejjj]

Firefox 3 Beta 5 scores only 71/100 compared to 75/100 for Safari 3.1 and 79/100 for the latest Opera 9.5 snapshot Just last week Opera was at 100/100 and Safari was at 98/100 for ACID3. What happened???

Oh yeah, those were numbers for non-production browsers, in-the-lab builds.

Re:How the mighty have fallen.

[TheRaven64]

The Safari number (100/100, pixel-perfect result) is for the WebKit nightly builds, which can be downloaded from the WebKit site or built by checking the code out of the (public) svn repository. This is directly comparable to a FireFox beta, since both are publicly available but not officially called releases. Since Apple isn't the only one to use WebKit (Nokia do and there are also GTK and Qt bindings now) I wouldn't be surprised if someone other than Safari gets better results by using a newer version of WebKit than the one that Apple officially supports.

Re:How the mighty have fallen.

[jsoderba]

A Webkit nightly is not directly comparable to a Firefox beta. A Webkit nightly is comparable to a Firefox nightly, which hasn't gone trough the testing and triage that betas get.

Re:How the mighty have fallen.

[asa]

A Webkit nightly is not directly comparable to a Firefox beta. A Webkit nightly is comparable to a Firefox nightly, which hasn't gone trough the testing and triage that betas get.

More importantly, a browser, whether nightly or beta, that is in the final days or weeks of stabilization work before it's shipped to the world, cannot be compared to a browser that's at or near the beginning of a development cycle. You cannot expect browser makers in radically different stages of development to be working on the same kinds of things.

- A

Re:How the mighty have fallen.

[asa]

The Safari number (100/100, pixel-perfect result) is for the WebKit nightly builds, which can be downloaded from the WebKit site or built by checking the code out of the (public) svn repository. This is directly comparable to a FireFox beta,

Um, no. Comparing snapshots from a codebase that's in the rampdown to release phase with a shapshot from a codebase that's not is certainly not a reasonable comparison.

Had you said "comparing the final nightly builds of WebKit being used by Safari just prior to shipping Safari 3.1 with the final nightly builds of Firefox just prior to shipping Firefox 3" then you'd have made a reasonable case.

- A

Actually...

[Mitchell Mebane]

Beta 4 only scored 68 / 100, so they have made some core changes. They fixed tests 42, 67, and 69. In addition, the test seems to run about 40% faster in B5 vs. B4, at least on my PC.

Re:Actually...

[barzok]

It's entirely possible that those fixes were not made specifically for ACID3, but instead had been targeted for Fx3 a while ago.

I think someone on the Mozilla team has publicly posted that they are not intentionally going after ACID3 fixes for the sake of making ACID3 fixes, in the interest of a stable & sane release.

Re:Actually...

[asa]

Beta 4 only scored 68 / 100, so they have made some core changes. They fixed tests 42, 67, and 69. In addition, the test seems to run about 40% faster in B5 vs. B4, at least on my PC.

Yes, Firefox does include a few Gecko fixes that increase the Acid3 score, but not because Firefox 3 is chasing the test. We're focused on getting in the right set of changes between now and ship and that's not going to be defined by Acid3.

- A

Re:Actually...

[DKlineburg]

And I thank them heartily. I would rather have a stable V3 than a buggy acid compliant. I think that 70% on roll out is a good score. And with the test only a little baby, that isn't a bad score. It isn't ment to be gotten right, right out of the box. I'm looking forward to installing the update to my V2 soon.

Fair comparison

[bunratty]

Yes, but those are very early development builds of those browsers. They haven't even seen an alpha release, much less a beta. The "Opera" build was actually using the WinGogi interface for Presto, and the Opera developers said not to use those builds for everyday browsing. You would want to compare those browsers to Firefox 4 nightly builds. However, I don't think work has even started on Firefox 4 yet. I opted to compare Firefox 3 to the recently released Safari 3.1 and the soon-to-be released Opera 9.5.

Connection parallelism

[Bogtha]

I'm sure somebody is likely to bring it up, so it may as well be me with some additional relevant facts. The HTTP 1.1 specification, RFC 2616, says that:

Clients that use persistent connections SHOULD limit the number of simultaneous connections that they maintain to a given server. A single-user client SHOULD NOT maintain more than 2 connections with any server or proxy. A proxy SHOULD use up to 2*N connections to another server or proxy, where N is the number of simultaneously active users. These guidelines are intended to improve HTTP response times and avoid congestion.

This "improved connection parallelism" is simply changing Firefox from using the RFC-suggested 2 persistent connections, to 6. Now, SHOULDs and SHOULD NOTs are not set in stone, but they do require careful thought before ignoring.

The Bugzilla entry debating this has a comment that points out that other browsers have also started to ignore this part of the specification:

• Firefox 2: 2 connections
• Opera 9.26: 4 connections
• Opera 9.5 beta: 4 connections
• Safari 3.0.4: 4 connections
• IE 7: 2 connections
• IE 8: 6 connections

Re:Connection parallelism

[garett_spencley]

RFC 2616 was published in 1999.

I agree that specification recommendations should not be ignored without careful consideration. However, I think the jump from 2 to 6 makes a lot of sense after almost 10 years of adhering to the specification and I don't think that it was done without careful consideration. Web servers and bandwidth have both strongly moved forward, and that specific suggestion in the RFC was just that. A suggestion. In the context of 1999.

Re:Connection parallelism

[Bogtha]

I think the jump from 2 to 6 makes a lot of sense after almost 10 years

Not all movement has been in the forward direction. Back them, most web traffic was totally static, even the HTML. These days, it's far more likely that the HTML is generated dynamically from something like mod_php. This, in turn, means that rather than tying up a slim process, a persistent connection ties up a "fat" process with a language runtime embedded in it. Three times as many simultaneous persistent connections means up to three times as much memory usage. Not as much of a problem if it's just a bog-standard 1999 static fileserver, but a big problem if you've got a 2008 dynamic interpreter built in.

Don't get me wrong, I'm not arguing that this is a bad change, I just think that it's got downsides as well as upsides. This will only further the adoption of lightweight reverse proxies like nginx, pound and varnish.

Re:Connection parallelism

[ErikZ]

It was also done over a 56K modem. It didn't matter how many connections you were using, you weren't going to get that web page any faster.

Re:Connection parallelism

[garett_spencley]

Parallelism mostly drives hits to static content. The websites that will benefit from browsers increasing the number of parallel connections are those that offer a large amount of static content on top of a dynamic page.

How many web sites do you know of that drive hits to hundreds of dynamic objects within the page ? Now how many sites do you know of that drive hits to static objects ?

Yes, there are some websites that load up portions of their site in iframes or via XMLHttpRequest in order to hit further dynamic objects. Some images may actually be passed through a dynamic script in order to serve dynamic images etc. but those sites need to consider the scalability of their applications carefully before implementing such features. Most sites have a single dynamic page that calls upon lots of static objects.

I know of many high-traffic sites that force increased parallelism onto browsers in order to speed up load times. They do this by hosting their static content across multiple sub-domains (which are often on the same physical server). So if you're hosting across 4 sub-domains you get 8 persistent connections that grab all of the static content instead of 2.

So I am of the belief that if increased parallelism is going to fry your server then you obviously had scalability issues that you should have addressed in the event of a traffic spike anyway.

Re:Connection parallelism

[Bogtha]

Parallelism mostly drives hits to static content.

That doesn't matter. My point is that typical sites didn't used to have dynamically generated content and now they do. That means that the memory use of each httpd instance rises regardless of whether or not any particular request is dynamic. When your clients triple the number of parallel connections they open, that means the peak number of requests can be significantly higher, meaning you need to run a lot more processes. And because those processes need more memory these days, using lots more processes has a bigger impact these days.

Now you can mitigate this problem by pre-emptively separating out your requests, for example by using a CDN or separate hostname/port for static resources, but if you are doing that, the browser uses more connections anyway.

They do this by hosting their static content across multiple sub-domains (which are often on the same physical server).

Same physical server, maybe, but not the same httpd. It would be silly to use "fat" httpd processes that handle dynamic content to just serve static resources when a tiny threaded/async httpd could do the job quicker using a fraction of the memory.

Re:Connection parallelism

[o'reor]

simply changing Firefox from using the RFC-suggested 2 persistent connections, to 6.

Cooool ! Now it's going to take 3 times fewer Slashdotters to bring a server down !
:-)

Re:Connection parallelism

[bcrowell]

I have my server set up to block access to IPs that do this sort of thing. In extreme cases, I've seen it bring down the server completely.

Re:Connection parallelism

[The MAZZTer]

Fasterfox has allowed you to tweak these settings for a while now, since Firefox 1 I think. You could also skip right to about:config if you know where to look. Of course the main concern is if too many people jack up these settings and download lots of files from web servers and give them a slashdot effect with not nearly as many users needed.

However the reverse of the coin is that if you download lots of small files from a server at once it's really annoying to see them download two. at. a. time. Sometimes you can just open up the floodgates and get it over with quickly and move on.

Re:Connection parallelism

[Bogtha]

Remember that there could be a legitimate, spec-condoned reason for a single IP making more than two connections - proxies with multiple users.

Re:Connection parallelism

[BZ]

> The HTTP 1.1 specification, RFC 2616

Which is currently in the process of being revised. One of the most likely revisions is bumping up the recommended connection limits.

Re:Connection parallelism

[TaoPhoenix]

(This is the one that included how many downloads you could do at once, right?)

Esp. since bandwidth etc. has improved, so has the audience expectation of using the Web. If I go to a utility site and want to download 6 items and then move on, it was absolutely infuriating to have to stare helplessly at the page waiting for the first two items to complete. I would never try to crush a remote server, but by now webmasters publish materials they intend for users to have.

Adblock Plus

[SonicEarth]

Adblock Plus doesn't work.

Re:Adblock Plus

[Incster]

Works for me. Use Nightly Tester Tools.

Re:Adblock Plus

[Anonymous Coward]

Nightly Tester Tools extension FTW.

Re:Adblock Plus

[garbletext]

Disable version checking with extensions.checkCompatibility = false

Re:Adblock Plus

[Chrisq]

Having done that slashdot threads did not display, so I have gone back to beta 4!

How hinged is Firefox development to Gecko?

[pizzach]

I mean, Firefox is just a front end to Gecko, right? Back when the Mozilla suite was the focus of the Mozilla foundation and Firefox was just a side project, Firefox development effected Gecko development very little. Is this still true even with the focus shift from the Mozilla suite to Firefox?

I do know that Firefox nightlies DO NOT equal webkit nightlies. Firefox and Gecko are actually devoloped on separate branches and are only merged at intervals.

Re:How hinged is Firefox development to Gecko?

[luserSPAZ]

I do know that Firefox nightlies DO NOT equal webkit nightlies. Firefox and Gecko are actually devoloped on separate branches and are only merged at intervals. Uh, this is completely wrong. Firefox and Gecko are developed on the same branch. Firefox isn't all of Gecko, but it's the biggest consumer by far, and as such the product cycles are heavily tied together.

Re:How hinged is Firefox development to Gecko?

[Fittysix]

There was recently a long discussion on this:
http://groups.google.com/group/mozilla.dev.platform/browse_thread/thread/4f1595449bf94cd2/

preed had some things to say in his blog as well:
http://weblogs.mozillazine.org/preed/2008/03/what_does_god_need_with_a_star.html

Re:How hinged is Firefox development to Gecko?

[pizzach]

Firefox and Gecko are developed on the same branch. Was this when the Mozilla foundation switched to Firefox from the Mozilla suite? Firefox wasn't official for the longest time. Also, how does XULRunner factor into things?

Re:How hinged is Firefox development to Gecko?

[BZ]

Firefox is the biggest consumer of Gecko, and more importantly the biggest test vehicle. The testing that Gecko gets in the Firefox betas is invaluable in identifying and fixing issues.

People have proposed on and off having a separate release cycle for Gecko, but the testing issue means that those ideas haven't really gone anywhere.

Re:How hinged is Firefox development to Gecko?

[bunratty]

This has been always. Everything (Gecko, Mozilla, SeaMonkey, Camino, Thunderbird, etc.) is all developed on the same branch or all on the trunk. There are different branches for different versions, but each branch contains everything.

Simple Pleasures

[petehead]

Yay, they put the home button back next to the URL.

Re:Simple Pleasures

[randyest]

Put it back? On my firefox2 it never left. I'm pretty sure you can customize all button locations anyway.

Re:Simple Pleasures

[polar red]

yup, just right-click on the toolbar, hit "customize", and you can drag the buttons around

Re:Simple Pleasures

[jbeaupre]

Hey, you must have my home button! Mine just disappeared with the upgrade! Thief!

Re:Simple Pleasures

[Bogtha]

The last Firefox beta didn't have a home button on the main toolbar, and you couldn't put it back via "customise toolbar" unless you knew the workaround of displaying another toolbar and dragging it from that.

Re:Simple Pleasures

[asa]

The last Firefox beta didn't have a home button on the main toolbar, and you couldn't put it back via "customise toolbar" unless you knew the workaround of displaying another toolbar and dragging it from that.

More accurately, IF you had hidden your Bookmarks Toolbar, then you had to restore your Bookmarks Toolbar visibility to its default state in order to move the Home button from there to the Navigation Tooolbar.

So, yes, if you'd previously customized your toolbars, then to accomplish what you wanted, you would have to customize your toolbars again.

- A

Re:Simple Pleasures

[Hillgiant]

I'm not using mine. You want it?

FF & Ubuntu

[jamstar7]

OK, I tried the hardy beta upgrade a few days back, complete with FF3b4. Totally hosed my system, forcing a reinstall. Not sure how much of the problem was with hardy & how much was with FF3b4, but I know for a fact that I won't upgrade til FF3 comes out of beta.

I didn't have back & forth arrows, no home button, and most of the extensions I use on a daily basis didn't work. Neither did the themes. Updates didn't work. And I couldn't edit my bookmarks.

Again, I don't know how much of this was FF3b4's fault and how much was hardy beta's. But I'm not going to upgrade to either til after FF3 comes out of beta. Hopefully, my fave extensions will work then...

Re:FF & Ubuntu

[randyest]

Are you seriously surprised that some firefox2 extensions and themes, which have a max version coded in the rdf file, didn't work with FF3beta? That's amusing. I guess it would be nice if extension developers could magically see into the future and update their extensions in advance for you, but it's unlikely to happen, for betas or final releases. You can, however, try editing the rdf to increase max_version, which often makes an extension work with a new ff.

Re:FF & Ubuntu

[domatic]

The FF that comes with Hardy isn't much different that the FF from Mozilla in the regard. The thing to do was to backup your .mozilla directory. Then if FF3B4 turned out to be rotten for you, just download FF2 from Mozilla and run it from your home directory.

Incidentally, if the problems were just the browser you could have at got back up to a stock configuration just by deleting or renaming the .mozilla directory. Backups in general before doing a thing like that are a good idea. Buy yourself an external USB hard drive. 320GB units can be had for 60 or 70 bucks. Then do something like this

mkdir backup_configs
cd backup_configs
dpkg --get-selections > installed_packages.txt
tar cvfz etc.tar.gz /etc

Plug in the drive then do

cd /
rsync --progress -h -av /home /mnt/disk (or however it shows up)

So you upgrade to a beta distro and things go pear shaped. Your way back is to:

New install of Ubuntu from disk. Just take the defaults. If anything strip it down to a bare minimum for speed. Almost everything will dselect-upgrade when you're done anyway.

Create a user named the same as the one you had in your old install.

Once booted to your new install plug in the backup drive then:

cd /
rsync --progress -h -av /mnt/disk/home/ /home/
cd /home/oldusername/backup_configs
tar xvfz etc.tar.gz
cp etc/apt/sources.list /etc/apt
dpkg --set-selections "less than sign...stinking lameness filter!" installed_packages.txt
apt-get dselect-upgrade

DON'T replace the contents of /etc with your tarball but you can cherry pick anything you customized or worked better in the old install from it. These procedures also work on Debian and other Debian derivatives.

Re:FF & Ubuntu

[jamstar7]

Are you seriously surprised that some firefox2 extensions and themes, which have a max version coded in the rdf file, didn't work with FF3beta? That's amusing. I guess it would be nice if extension developers could magically see into the future and update their extensions in advance for you, but it's unlikely to happen, for betas or final releases. You can, however, try editing the rdf to increase max_version, which often makes an extension work with a new ff.

No, I'm not surprised that FF2 extensions don't work with FF3. Having been around the block a bit, it didn't surprise me at all.

What did surprise me was a lack of updated extensions for FF3. Most of the ones I use on a daily basis just weren't there. FF3 betas have been around for what, 2, 3 months now? Should be a lot more extensions & themes geared up for FF3 already.

Re:FF & Ubuntu

[jamstar7]

Appreciate the advice, but it wasn't needed. I've been round the block a few tiems before, having started out by going totally Linux with Red Hat 3.0.3 back in the day. :D I always keep my /boot, /home & /usr partitions seperate from /. Makes backups & restores a LOT easier. Most of teh 'restore time' is needed to apt-get the packages I usually use. This restore, I think I lost like 2 hours of email (maybe 35 messages; I get a lotta email), and some tarballs I've got backed up on DVD. Oh, and of course, my fonts, also backed up on DVD, my 500 gig USB portable drive, and in my /home. I even got back all my archived old emails (about 7000 of 'em in the mysql database, I use Pronto 3.0.4), and with a 20 second hack, got my mysql back under my command by replacing the deb-admin-user password in mysql.

Like I say, not my first barbeque.

Re:FF & Ubuntu

[Tack]

I have no idea how many other extensions are in the same boat as mine, but I submitted my update for Firefox 3 to addons.mozilla.org in February, and it is still pending review.

Here's what I got the last time.

[Anonymous Coward]

Actually got this with 3.0b2, 3.0b3 and . . .

firefox-3.0b4

luser@3[firefox]# ./firefox ./firefox-bin: error while loading shared libraries: libpangocairo-1.0.so.0: cannot open shared object file: No such file or directory
luser@3[firefox]# uname -a
Linux CJ 2.4.27 #2 SMP Mo Aug 9 00:39:37 CEST 2004 i686 GNU/Linux
Debian

I got newer systems, they've all been the same.

Give me a second I'll try this one.

Re:Here's what I got the last time.

[CajunArson]

I think you are missing some packages Firefox 3 needs. Go to the directory where the firefox binary is located and run this:

ldd ./firefox-bin

You'll get a big dump of *.so files that Firefox expects to find at runtime. For your particular problem, under Ubuntu/Kubuntu the package you need is the libpango1.0-0 package (apt-get install this one). You may find other libraries missing too when you run ldd. Hunting them down can be a PITA, but when 8.04 comes out it will have everything you need and FF3 will be part of the main repository.
    If you aren't using an Ubuntu flavor, the same basic methodology applies, just use your own distro's packaging system to install the pango libs and other dependencies.

Another problem that came up on a friend's install was a library linking problem because a bunch of libraries that FF3 uses are in its distribution directory and for some reason the runtime library loader is not finding them. My friend had an Ubuntu 7.10 install where this was a problem... but at the same time I have Kubuntu 7.10 and it wasn't an issue at all (go figure). That can usually be solved by tweaking the LD_LIBRARY_PATH environment variable to include the FF install directory where its extra libraries reside.

Re:Here's what I got the last time.

[Anonymous Coward]

Linux CJ 2.4.27 #2 SMP Mo Aug 9 00:39:37 CEST 2004 i686 GNU/Linux

Gee, I wonder why you're having trouble running the latest Firefox beta. If you don't have a recent enough gtk2 stack (gtk+, cairo, pango, glib, etc), it won't work...

Re:Here's what I got the last time.

[Anonymous Coward]

with beta 5, I still get ./firefox-bin: error while loading shared libraries: libpangocairo-1.0.so.0: cannot open shared object file: No such file or directory

Running ldd gives this:

[firefox]# ldd ./firefox
                not a dynamic executable

How recent to the pan libraries need to be for it to run? I have a few machines running 2.6.19. . . Firefox2 works on my 2.4 kernel machines.

Re:Here's what I got the last time.

[CajunArson]

No, run

ldd ./firefox-bin

(typed exactly) the important part is the "-bin" after firefox since the file named "firefox" is just a shellscript that invokes the binary firefox-bin executable file. As for libraries... if you are running any modern (probably 2006 onwards) distro and if you install the proper packages it should be OK. If you are on some ancient Linux distro that hasn't been updated then there could be issues. Time marches on.

Bloomin' fast!

[Anonymous Coward]

Looks like it's as fast as shit off a shovel http://blogs.zdnet.com/hardware/?p=1648

FireFox 4

[Starturtle]

It's getting hard to appreciate the forthcoming FireFox 3 when they're talking about the features of the forthcoming FireFox 4 on /.

http://tech.slashdot.org/article.pl?sid=08/03/28/1736255&from=rss

Firefox 3.0 and the spring linux releases

[themildassassin]

I'm wondering how the new releases of distros like Ubuntu and Fedora are going to handle not having a stable version of Firefox 3.0 until June. Currently Ubuntu is using beta 4 for the hardy beta, will the plan be to revert back to FF2 when hardy becomes stable or release with a beta version of FF3?

Re:Firefox 3.0 and the spring linux releases

[asa]

I don't think they're going backwards. Firefox 3 beta 4 is a huge improvement over even the stable shipping version of Firefox 2.

- A

Re:Firefox 3.0 and the spring linux releases

[AncientPC]

Ubuntu 8.04 will most likely revert back to FF 2.0 (and KDE 3) simply because it's LTS.

Re:Firefox 3.0 and the spring linux releases

[martinw89]

Kubuntu Hardy is not LTS, and the KDE 3 and KDE 4 releases are separate entities (the KDE4 release being termed the "remix").

Re:Firefox 3.0 and the spring linux releases

[chasd]

Fedora 9 will ship with Firefox 3. Current Rawhide has the Beta 5 release. If you follow the Fedora development list, you'll see the Firefox maintainer updating Firefox and XULRunner to CVS snpashots almost every day. I believe it is planned to release an update RPM to Firefox 3 final once it is released, even though Fedora 8 ISOs will likely ship with a Firefox Release Candidate.

Not sure about Ubuntu, I don't follow its development.

Re:Firefox 3.0 and the spring linux releases

[Zaiff Urgulbunger]

I hope not. They do have a 8.04.1 scheduled for a few months later... June/July if my memory serves correctly, so I think they can get away with shipping a late beta/release-candidate.

As long as it stops crashing...

[Anonymous Coward]

... like it does every 20 mins (particularly when loading flash content) in my Ubuntu box, I will be very happy.

Get Adblock now

[Anonymous Coward]

I updated as soon as I saw this, and found that adblock was disabled. Download the xpi (automagically when attempting to install) and modify the allowed versions in the file install.rdf inside the xpi. (from mc you can edit directly) Then load the xpi (named tmp.xpi) with something like file:///home/user/tmp.xpi, after you copied it to a... more sane directory than the cache from firefox.

Good luck (well, it worked for me)

Adblock, adblock, adblock...

[TheVelvetFlamebait]

Would it kill you guys to view a page with ads?

Re:Adblock, adblock, adblock...

[billstewart]

No, it wouldn't kill me to see a few ads. But at least under 2.0, it tends to kill my browser performance - one thing I typically do is open tabs for all the interesting news articles in fark.com, and my machine dogs out badly enough if I do have the ads turned off, and is worse with them on.

Back when ads were just non-blinky banners, they were only a problem because my 14.4kb modem took time to download them, but these days they're singing dancing Flash/Javascript/Popup-trying/Popunder-trying/whatever things that get really annoying, in addition to the privacy implications.

Re:Adblock, adblock, adblock...

[TaoPhoenix]

Yep.

After all, net content is usually plain text.

** BUT ADS ARE BOLD!!! THEY WANT YOUR ATTENTION!!!**

The web has come a long way in putting up real news.

** BUT DID YOU KNOW THAT THERE ARE FLASH ADS THAT CAPTURE YOUR MOUSEOVER AND START PLAYING FLASH GAMES!!!????**

Sometimes you don't even make it through a sentence

** FORD REALLY IMPROVED THEIR QUALITY AFTER BEING SHOWN UP BY THE JAPANESE IN THE 1980's. **

before an ad rips your focus away from you.

Re:Adblock, adblock, adblock...

[olman]

No.

If it was done like in /.

However, many pages have shocking number or flashing blinking noisy piece of shit flash gimmicks that hock up 100% CPU and slow the machine to crawl. On top of annoying me to death, that is.

It's like you'd have to watch the most annoying tv spot over and over and over on a 2nd tv which you couldn't mute.

Not to mention that with ads the page layout changes so the actual content is more like 50% of the page with rest being taken up by ads.

So, yeah, I don't mind static image ads or text ads but goddamn animated images or flash ads that play some stupid jingle over and over..

Re:Adblock, adblock, adblock...

[mgblst]

I am a cheap scottish bastard, who lives in a cave, you can be sure I won't be buying anything. So in fact, I am doing them a favour by saving their resources. They should be thanking me.

Re:Adblock, adblock, adblock...

[Hillgiant]

No, but it will keep me from upgrading to FF3. The same way it prevented me from sticking with Safari and sticking with Opera when I tried them out.

I have been browsing with fewer and fewer ads since back in the late 90's with a crude HOSTS file. I really do not want to move backwards.

It actually slows down wikipedia for me

[IdahoEv]

The most common way I used to use the old URL bar is to type just "en" and let it fill out the rest of http://en.wikipedia.org/wiki/, delete the something, and type in whatever noun it is I'm interested in.

When I'm looking up a new wikipedia page - something I haven't visited before - this is MUCH faster than typing out the whole URL or going to the front of wikipedia, typing a name, and hitting "go to" or search. In fact, I periodically visit the wikipedia page for the letter "A" just so there's a short wiki URL near the top of my hit list. This means a new wikipedia page in a tab is only:

• CMD-t for a new tab
• "en", down arrow, end, backspace
• {type name of subject}.

That's only seven keystrokes plus the name of the page for a new wikipedia page I've never visited before, plus no use of the mouse. It takes about half a second total. I do this far more often than I revisit an old wikipedia page. And even when I want to that, I just type "en" and then arrow down through the list of hits.

The awesomebar totally screws this up, because the letters "en" match thousands of other things in my history since they will now match mid-word. Moreover, since it shows two lines per entry with little bolds and underlines everywhere, it's much slower to scan visually, and much slower to draw on the screen on my 2-year-old powermac G5.

I hate the new "awesombar". It's cluttered and slow and much less useful to me. I wish I could turn it off! I'm actually sticking with FF2 for the time being, despite the horrible, annoying and unresolved FF2 Macintosh Window Snapback bug, just because the awesombar slows down my workflow too much.

Re:It actually slows down wikipedia for me

[ChronoReverse]

Have you thought about substituting "en" with "wi" for "wiki"? Do it a couple of times and wikipedia will be the top of the list.

If you ask me, "wi" is better associated with "wikipedia" than "en" is.

Re:It actually slows down wikipedia for me

[Constantine XVI]

R-click the Search box on Wikipedia.
Click "Add a keyword for this search"
Put something in the Name box
In Keyword, put something like "wp"
Click Add

Now, when you type "wp foobar" in the address bar, it runs a Wikipedia search for foobar

Re:It actually slows down wikipedia for me

[ChronoReverse]

That reminds me. Add Wikipedia into the search box and set it as default.

Now you can just press ctrl+k and type a term and press enter.

Re:It actually slows down wikipedia for me

[Mozk]

To disable Awesomebar:
Go to about:config, type urlbar, "promise to be careful" if you haven't already, and either set browser.urlbar.maxRichResults to 0 or set browser.urlbar.matchOnlyTyped to true.

Re:It actually slows down wikipedia for me

[amRadioHed]

Why do that? Everyone knows bitching is easier than change.

Re:It actually slows down wikipedia for me

[Anonymous Coward]

Now, when you type "wp foobar" in the address bar, it runs a Wikipedia search for foobar

Until you go to your friend's computer, or a machine at a training/demo facility, which isn't set up this way, and then you're completely lost because you've forgotten that English-language Wikipedia URLs always begin with "en."

Isn't the thing we hate most about Windows the fact that we have to spend 15-20 minutes de-dumbifying every installation ("yes, I want to see the goddamn file extensions, yes, I want to see file sizes/details instead of big fat icons, yes, I want to see the full path in the fracking file explorer...") on which we're going to be working for more than a few hours? And that we're going to have to put all that crap back to whatever the stupid defaults were if we don't want our friend angry with us when he gets his machine back?

You shouldn't have to have take the time to create/add a bunch of custom settings, one per frequently-accessed web site, every time you want to use a new machine. We know better than to design operating systems that way. Why are we trying so hard to repeat this mistake with web browsers?

Re:It actually slows down wikipedia for me

[aj50]

We thought of that, we called them bookmarks. Make one for every commonly visited site and they get stored in a file which you can take with you. In fact, I believe there some effort going into a project to allow you to store your bookmarks on a server somewhere and grab them from any computer.

I don't use windows much but when using a new linux computer I copy across my .vimrc and .bash_profile and I'm all set (for important stuff at least).

Re:It actually slows down wikipedia for me

[nekokoneko]

The most common way I used to use the old URL bar is to type just "en" and let it fill out the rest of http://en.wikipedia.org/wiki/, delete the something, and type in whatever noun it is I'm interested in.

I access Wikipedia in the exact same manner as you do: a quick workaround to this problem is accessing "en.wikipedia.org/wiki/" a couple of times so it stays on top of the address bar suggestion list. It's actually quicker than accessing "en.wikipedia.org/wiki/A" because you don't have to delete the A.

Re:It actually slows down wikipedia for me

[phyrz]

seriously, type wiki then your term in the address bar in firefox 2 and it goes to the wikipedia article via google feeling lucky - 99 times out of 100 (possible more - its never failed for me)

Re: New Profile When Upgrading

[ChronoReverse]

Have you through about creating a new profile for Firefox and checking which of your addons are causing the leaks?

A truly clean copy of FF2 doesn't leak that much and FF3 even less. At least, whenever I followed the instructions of people complaining about leaks on a clean copy that's the case. The leaks has thus far been minimal.



Post the tabs that you use at once and the way you use them. I've always wondered which sites caused that much leaking.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Konqueror

[RAMMS EIN]

How does Konqueror stack up? I've been using Konqueror ever since I discovered it uses much less memory and (thus?) runs faster than Firefox. Now with the improvements in those areas for Firefox, this may no longer be true. Does anyone have numbers?

Re:Konqueror

[bunratty]

I would be very surprised if a full-featured, modern browser used significantly less memory or was significantly faster than Firefox 3. Maybe Dillo and Links, but not Konqueror.

Re:Konqueror

[pherthyl]

Well, I just did a simple test. Start Firefox 3 beta 5, no extensions loaded and open the parent story on slashdot. Then start konqueror (4.0.2) to view the exact same page and scroll to the bottom on both browsers.

Memory usage measured with exmap (the only decent way to measure memory usage on linux http://ktown.kde.org/~seli/memory/analysis.html )
Effective resident (K)
firefox-bin 71262
konqueror 30868

So on this simple test, konqueror uses less than half the RAM as Firefox 3b5. Obviously this test is simple minded, as I only load one page. And Konq in KDE 4.0.2 still has some nasty bugs. But because I'm running KDE, konqueror actually shares a lot of its memory with other apps, while Firefox is one monolithic beast that doesn't share libraries with any other app on my machine.

Re:Konqueror

[pherthyl]

Sorry, my mistake. I noticed I did have an extension still active in FF. So firefox 3 actually uses 45735k of effective resident. So not nearly as bad as I first posted, but still 50% more than konq on the same page.

Re:Konqueror

[bunratty]

Konqueror is using 33% less memory than Firefox after one page is loaded. I think you'll see that advantage drop as you continue to load pages. Be sure to load the same pages in both browsers for a fair comparison. It's only after you actually start using the browsers that you notice the memory improvements in Firefox. And that's the memory use that matters, not memory use after loading just one page.

Re:Konqueror

[pherthyl]

Yep. After loading slashdot, then opening the first 5 stories in tabs, then switching to them and scrolling to the bottom, then closing them, then going to qtcentre.org and opening the first 6 forum posts in tabs, memory usage is 68mb for firefox and 64mb for konqueror. Closing all those tabs and going to about:blank puts memory usage at 64MB for both.

So basically memory usage for normal browsing is the same.

Re:Konqueror

[pppppppman]

I wouldn't be running to Firefox any time soon. I got 3b5, and after looking at 2 websites, I have closed all tabs, closed download window and navigated to about:blank.

Firefox: 72MB
Opera: 63MB (I have 6 tabs open, and have been using it muchly)

I for one call BS on the fixing of memory leaks.

side by side install with FF2

[Tumbleweed]

I haven't installed it yet, but FF Portable has a FF3b5 version available:

Firefox Portable

Yay!

Re:side by side install with FF2

[jsoderba]

You don't need to download a new portablefirefox package every time there is an update. Just use the normal Firefox updater from the Help menu.

Re:side by side install with FF2

[klui]

Probably not a good idea. http://portableapps.com/forums/support/firefox_portable/faq http://portableapps.com/node/10338

Still not right

[rikkards]

I still get a tab "hanging" on loading and any subsequent attempt at opening any link in another window hangs as well.
Sigh...

Separate processes! separate processes!

[manifoldronin]

Separate processes!

When are we going to be able to start firefox in a process separated from one already running? I know it can be achieved by starting it under a different profile every time, but that is not convenient, and sometimes doesn't work - I'd lose access to all the bookmarks and cookies in the main profile.

I mean, don't mozilla people also hate having 31 Firefox windows all crash and burn just because one of them tripped over a badly implemented web site? Or do they just browse in one window, one tab, ever?

Re:Separate processes! separate processes!

[bunratty]

Firefox will not crash due to a badly implemented web site. Firefox will crash only due to a bug in Firefox or software that is running inside the Firefox process that you have installed on your computer, such as an extension, plugin, driver, or the operating system itself, or in some circumstances, a hardware problem. Is Firefox crashing often for you? If so, follow this advice.

I don't need no browser

[Britz]

What I badly need is a replacement for that awful Flash player. There is so much Flash content on the web now, that unfortunately I need a viewer for this. Firefox 2 is fine. The need for better/faster viewing or more features is not very big.

So please Mozilla foundation: If you want to do something to improve my web exprerience just put some effort into Swfdec or Gnash or do something from scratch and put it into Firefox.

http://swfdec.freedesktop.org/wiki/
http://en.wikipedia.org/wiki/Swfdec

http://en.wikipedia.org/wiki/Gnash

Re:I don't need no browser

[maxume]

Use Flashblock:

https://addons.mozilla.org/en-US/firefox/addon/433

It's only a 94% solution, but that ends up going pretty far. It doesn't do anything about the flash plugin being a proprietary binary though.

Re:I don't need no browser

[kurtb149]

xul, xbl, and svg are intrinsic to firefox and should replace all flash applications.

Re:I don't need no browser

[Anonymous Coward]

I'll get right on that!

Memory abuse = Poor responsiveness on XP

[billstewart]

I agree with the parent article - my work laptop only has 512MB, and it's only been recently that memory's cheap enough for me to bother paying my own cash to upgrade it with DDR1 just to support one over-bloated application. Ok, that one over-bloated application _is_ my browser, and I use it even more than email, and it's not like I'm going to switch to IE for regular use unless I'm really desperate, but Firefox 2.0 is usually using far more RAM on windows I've already closed than Microsoft Office is currently using for the 17 Powerpoint windows, 17 Word documents, 2 Excel spreadsheets, a Visio diagram, and 4 IE windows that I've currently got open now. (Those add up to 13 MB, Outlook is bloating along at 49, and Firefox is currently at 55 only because I just killed the previous 167MB that was swapping to death and reopened 4 tabs of Slashdot. But Firefox is normally well over 100 except when I've just restarted it, and usually turns my machine into a total dog somewhere around 200-300 MB.)

Re:Memory abuse = Poor responsiveness on XP

[dedazo]

Interesting how it's supposedly Microsoft that creates "bloated" and "inefficient" software, isn't it?

Re:Memory abuse = Poor responsiveness on XP

[pthisis]

You're implying a logical fallacy. "Microsoft creates bloated software" does not imply "Non-Microsoft does not create bloated software".

Re:Memory abuse = Poor responsiveness on XP

[dedazo]

I will bow to your superior grasp of logic as soon as you tell me where I implied that, beyond pointing out that the anectode presented by the OP is good empirical evidence of my point.

Three Times The Fun!!!!

[Slimee]

I don't know what happened, but upon installation and opening I now have three forward/back buttons....I just don't know which one to choose when I want to click back!! there's so many choices! I imagine this problem will fix itself if I close out, but I don't know if I want it to go away just yet hah.

Re:Three Times The Fun!!!!

[asa]

I don't know what happened, but upon installation and opening I now have three forward/back buttons....

This is a known issue and it's being worked on (bug 425079.) In the mean time, you can fix it by simply going to View -> Toolbars -> Customize and removing the extraneous buttons.

- A

Downloading .exe files?

[klui]

Does anyone have trouble downloading exe files? I get a "Blocked: Download may contain a virus or spyware -- sourceforge.net". I'm trying to download the portable version, and 3b4 actually had trouble downloading it as it had trouble with some downloads where it would remove the file as soon as it's done with the download. 3b5 is worse in that it appears all exe files are blocked. Doesn't seem to be fixed if I toggle browser.download.manager.scanWhenDone to false. The message comes up immediately and nothing gets transferred. Perhaps it's not working quite right with Symantec AntiVirus Corporate?

Re:Downloading .exe files?

[klui]

Looks like my problem is solved by https://bugzilla.mozilla.org/show_bug.cgi?id=416683#c8 under Windows. Tying the ability to download files to Internet Options settings on Windows is new behavior for beta5.

How to make it go away

[Anonymous Coward]

Right-click, Customize, Use Small Icons.

If I hadn't been able to get rid of the new "Star Trek communicator meets Amiga 3-D effect" forward/back buttons, I would have stayed with Firefox 2.

Greasemonkey

[BrookHarty]

Greasemonkey broke again, /sigh

Every update it seems to break, what keeps changing that this addon breaks every time?

Re:Greasemonkey

[anaesthetica]

The extension doesn't break. Extensions have an inbuilt version number specifying which versions of Firefox they are compatible with. When you download an extension, it has an internal specification saying, "I work on Firefox 1.5 - 3b4". When you upgrade to Firefox 3b5, your extension, like a good citizen, disables itself, because it has not been tested on your new version of Firefox, and could potentially cause massive problems in unforeseen ways if there was a core change in how Firefox behaves. It is up to the extension developer to do due diligence in checking his/her own extension, and then if no problems crop up, bump the version compatibility number on their extension and re-release, often with no other changes to the core code of the extension itself. This system exists to protect your data. If you do not like it, you can turn this protection off, see here.

I'd like to see a version of FF3 for the OLPC XO!

[timothy]

I put Opera on my XO a few days ago, which turns it from a really *neat* device to a really *useful* one.

I would rather run software freer than Opera currently is, but I also (esp. on a device like the XO, where switching tasks is notably slow) want to have tabs in my browser. Right now, as far as I know, Opera's the only way to get that -- but I'd rather it be Firefox, because of the extensions, muscle memory for shortcuts and menus, etc.

timothy

512 MB + heavy use = Poor responsiveness on XP

[bunratty]

512 MB of RAM is simply not enough to comfortably run Windows XP, Outlook, Powerpoint, Word, Excel, Visio, and two different browsers. 512 MB of RAM is the minimum needed to run Windows XP and a few light applications. For a heavy user such as yourself, 1 GB is needed if you don't want it swapping to death.

Re:512 MB + heavy use = Poor responsiveness on XP

[billstewart]

Works just fine running XP, Outlook, Powerpoint, Word, Excel, Visio, and *one* of the two different browsers (IE), though of course the other browser (Moz2.0.0.12) is the one I do most of my browsing on.... Looks like I should try installing Mozilla 3.0b5.

So...

[bunratty]

So... it's faster for the vast majority of users that don't tweak those settings. That is new for beta 5.

More about the memory tuning...

[imtheguru]

http://blog.pavlov.net/2008/03/11/firefox-3-memory-usage/
http://blog.pavlov.net/2007/11/15/less-fragmentation-coming-in-firefox-3/
http://blog.pavlov.net/2007/11/14/leaks-memory-we-never-forgot-about-you/


Cheers.

Interface niggle - smooth scrolling tabs

[Thornae]

Okay, trying it out, and immediately found an annoyance:
They've implemented Smooth Scrolling in the tab bar. I really dislike smooth scrolling, and turn it off whenever possible, but in this case, you can't.

Unless you dig into the discussion of the patch to find that you have to add

user_pref("browser.tabs.useLegacyScroll", true);

to your prefs.js file. Unfortunately, this also disables the good bits of this patch, one of which is that triple-clicking on the arrows at either end of the tab bar will take you to that respective end of your tabs list. That's not very well worded, but if you use lots of tabs, you'll know what I'm talking about and why it's a good thing.

However, I doubt that they'll add an option to turn the smooth scroll off after all the hard work they went to to implement it. As they say: "animation is something we definitely want to do more of" (blatantly out of context quote, but it does seem to be the majority opinion...)

"Redirect Notice" for any google search result

[Anonymous Custard]

http://bugzilla.mozilla.org/show_bug.cgi?id=426679

After visiting google.com and searching for any term, such as "digg", clicking on the search result takes you to a page with the following text:
================
*Redirect Notice*

  The previous page is sending you to _http://digg.com/_.

  If you do not want to visit that page, you can _return to the previous page_.
================

I suppose this is intended to fight Phishing redirection attacks, but as of Firefox 3 Beta 5, the Redirect Notice is shown for all search results, including clearly non-phishing results such as digg.com and yahoo.com.

This does not occur in:
Firefox 2.0.0.13
Firefox 3 Beta 4
Internet Explorer 7

This does not occur for the following google-owned sites: blogger, orkut, youtube.
It DOES occur for similar but not google-owned sites: wordpress, linkedin, hulu.

Quickly Uninstalled

[Vampyre_Dark]

I found this new version clunky and annoying. What the hell is up with the address bar? I haven't been following the development of FF recently, but that bugs the hell out of me. I didn't see a way to turn it off, so I nuked the whole program.

Also, my must have extensions don't work, (which was a given).

Re: Killing RichResults

[TaoPhoenix]

I got grumpy with all these BBC feed entries appearing in my dropdown. For the moment, I set richresults to 0. But do you know how to get FF3-B5 to simply report real histories only without any of Mozilla's or BBC's entries?

Re: Backing out Fancy stuff

[TaoPhoenix]

Yep.

I just butchered my Bar in an attempt to do this, though when the official FF3 comes out I'll have to look for Back-Out add-ons.

x64 build

[Anonymous Coward]

Here http://www.mozilla-x86-64.com/download.html

Still slow as fuck

[Wiseman1024]

It's still slow as fuck because the Mozilla guys are obsessed with that Cairo crap. That's it, I'm tired of Firefox getting uglier, slower and more bloated with every new version, and just for what? The release of Firefox 3 will mark the day I switch to a better browser. I'm undergoing evaluation of Firefox alternatives as I post this.

Don't want to quibble

[Chrisq]

But I fust downloaded 5b4 for windows and scored 70 rather than 71 on acid 3

Re:Who cares about Safari?

[sanyacid]

Last time I tried Safari on Windows - it was terrible. But when comparing OS X versions of Firefox and Safari, I actually prefer the letter one. The only thing that's missing is extensions. So, obviously, many people who use OS X do care about development of both Firefox and Safari.

Smarter parallelism

[Anonymous Coward]

When I load a session in Firefox 2 with 20 or more tabs it seems none of them load. If I load 10 at a time its all fine and dandy. Some way to limit the total number of simultaneous connections Firefox 3 makes would be much appreciated. Even moreso when I have to use the 64k connection, not everyone can get superfast broadband. Likewise when I'm only viewing one page I'd like it to grab all the external images simultaneously to all the same-host stuff.

Re:I'd like to see a version of FF3 for the OLPC X

[RAMMS EIN]

I've been impressed by Konqueror. I don't know how it compares to the new Firefox, but compared to Firefox 2.0.0.12, Konqueror 3.5.5 eats up a _lot_ less memory and runs noticeably faster. And, contrary to Opera, it is open source. I've run 3.x versions of Konqueror on machines with 64 MB memory and 200 MHz Pentium MMX CPUs...I don't know how that compares to the OLPC, but it may work. If you give it a try, let me know how it goes.

Don't like awesomebar? Turn it off.

[deadsquid]

Like a lot of things in Firefox, you can turn it off if you really don't like it. To disable the awesomebar and just have the old URL autocomplete, you'll need to add a pref. I confess, I didn't much like it at first, but the behaviour learning and improved search introduced in 3b4 has sold me on it. Any who... to turn it off:

Add the boolean pref browser.urlbar.richResults and set it to false.

This pref is only checked on startup, so youâ(TM)ll need to restart Firefox for it to take effect. More information about this pref can be found here. More info on setting prefs through about:config starts here

Make a new RFC or make an X- header

[tepples]

But there exists no later version of the RFC than the one from 1999. If a web server wants web browsers to use behavior that violates the latest RFC, the server should indicate this to the user agent.

If you want mail, type mail.

[tepples]

Example: I've typed in web, am I more likely to be looking for "xkcd - A *web*comic of ..." or "GameFAQs... Video games *web* site..", perhaps I want "Lets turn this fucking *web*site yellow" or "Rapidshare: 1-Click *Web*hosting" or maybe, just maybe, I've started typing in webmail.bath.ac.uk like I do reasonably often (but probably not as much as I visit xkcd or GameFAQs). What you want is mail, not something generic on the web. Have you tried typing mail, or possibly webmail?

Re:If you want mail, type mail.

[aj50]

Yet another person who has missed my point.

I don't want 'mail', I want 'webmail.bath.ac.uk' which is what I'm typing in.

I am typing a url in the url bar because a url is a unique identifier for a web page, it's the way that page is identified and isn't ambiguous.

Why should I want to type in 'mail' let my computer guess that I really want webmail.bath.ac.uk, and have me confirm that when I know I where I want to go?

If you want your bank, type your bank.

[tepples]

When I type 'online', I expect to see my banks website URL as the first choice (as it starts with 'online') But why should other people expect starts-with to be the criterion, rather than contains? You could just type the name of your bank.

Smartphones

[tepples]

Memory is perhaps the second cheapest commodity on a modern day PC after disk space.

Modern-day PCs cannot be carried in your pocket. These are designed to target Pocket IE and the smartphone version of Opera[1].

[1] The smartphone version of Opera is a full web browser. It is not Opera Mini, which uses a proxy to distill HTML.

dolsen

[danolsen]

I have updated to Beta 5 and already twice I have had Firefox go crazy with my memory. It got up to 1.5GB before I forced it to quit. All I had up was Google Reader, CNN and the Prototype documentation open. So is this another memory leak that has been introduced or is it just going crazy for me?

incorrect comparison

[gerardrj]

"As for the Acid3 test, Firefox 3 Beta 5 scores only 71/100 compared to 75/100 for Safari 3.1 and 79/100 for the latest Opera 9.5 snapshot. "

Why are you comparing two betas to a shipping product? The "beta" Safari, known as webkit (www.webkit.org), has been scoring 100% on Acid3 for about a week now.

Was it ignorance, shame or malice that causes people to use inappropriate comparisons to bolster their positions?

Re:incorrect comparison

[LiquidFire_HK]

Webkit isn't the "beta" of Safari, it's just the name of the rendering engine. As for your concern, see this comment: http://tech.slashdot.org/comments.pl?sid=509004&cid=22943286. And I don't see how this is bolstering anyway.

but it only uses half my CPU!

[Harik]

Still no 64-bit release, but at least they're building it on the minefield nightlies.

Which, BTW, seem to indicate that b5 is the last before final. Nightlies are now 3.0pre not 3.0b6pre.

Well. Scooped again

[BattyMan]

Little Green Footballs ran this yesterday.

Imagine my shock when I showed up at my brother's place after work, and he's all enthused about the FF3 he's just installed on his iLaptop(tm)(R)(c). He heard about it on his political (not even techie!) blog. It does inertial scrolling! At least on iMach(tm)(r)(c).

"What? Firefox 3!? No way. I woulda seen it on /.!"

At least I was able to take some wind outta his sails by running Acid3 on it. He was pissed to see it score 'only' 71%: "WTF?! It doesn't even _work_!"

So, while you guys still _really_ rule when it comes to Anime & TV Sci-Fi, you're falling behind the curve in the tech deployment stories.

And if you want the dirty details of OffalXML (wait- I guess that's now ISO DIS29500(tm)(r)(c)), Groklaw is the place for that.

But nevermind that. What do we need serious stuff for, anyway?

Who's already seen BSG 4.1, (out Friday on Sci-Fi)?
WTF happens?

And who's ready for Summer Glau as a ballet Terminator? Yow!

My Impression

[DollyTheSheep]

Not really faster than 2.x with FasterFox with respect to loading content. But Beta 5 is really quick with executing JavaScript, as can be most prominently experienced with GMail.

Gecko still fails on Mac OS X

[owndao]

Please note that regardless of their rendering abilities both Opera 9.27 (that uses the Opera Widget Engine, I believe) and Firefox 3.0b5 (which uses the Gecko engine), still cannot retain styles and links when copying from them to any style-capable editor (just about everything) under Mac OS X (any version). Safari, OMNIWeb, and other WebKit-based browsers have and still handle this with no trouble. This is a bug in Gecko that has been in Bugzilla, and ignored, for years. Please appeal to Firefox and Opera development for fixes in this area. Those of us who must copy and paste styled sections of webpages must use Safari or other WebKit apps as constantly switching back and forth is, in my opinion, simply not worth the effort.

use enigma instead

[Anonymous Coward]

I suggest people look for enigma browser and download it from http://store.democratz.org/

You can find the link just above the products and you can download it for free.

Engima runs faster than Firefox and you can easily run 32 tabs and not slow down the machine.