Firefox 18 Launches With Faster IonMonkey-Enabled JavaScript, Built-In PDF Viewe

An anonymous reader writes "Mozilla on Tuesday officially launched Firefox 18 for Windows, Mac, Linux, and Android. The improvements include a new JavaScript compiler, a built-in PDF viewer, as well as Retina and touch support. The release notes are available, as is a list of changes for devs."

Comments

Lawlz

[Anonymous Coward]

Psh, I just upgraded to Firefox 22 just 5 minutes ago. Firfox 18 is so 30 minutes ago.

Re:Lawlz

[MightyMartian]

I'm waiting until next week. I hear version 50 is going to be awesome!

Re:Lawlz

[jonadab]

Yeah, but you still haven't caught up with me. I'm using Emacs 23.2.1. Since 23 is more than 22, clearly I'm better than you. Though I really should upgrade to version 24 one of these days.

Re:Lawlz

[jonadab]

Oh, and my Gecko version is 20081217. That's WAY bigger than your puny little Firefox version number. So there.

Re:Lawlz

[hobarrera]

Are we really going to have this kind of joke every time a firefox version is released? I'm mean, it's getting a bit repetitive. I've seen about about 10 times in the last 10 minutes!

Re:Lawlz

[Anonymous Coward]

Yes. And ONLY for Firefox, because they're the only ones who get these kinds of posts on Slashdot, because other browser's users would rather pick on Firefox than post updates about their own favorite browsers. Haven't you heard? It's cool to pick on Firefox now!

Re:Lawlz

[Anonymous Coward]

I'll stop when Mozilla stops with the silliness of marketing minor versions as major ones.

Sent from my Firefox 21.0a1.

Re:Lawlz

[Anonymous Coward]

Are we really going to have this kind of joke every time a firefox version is released? I'm mean, it's getting a bit repetitive. I've seen about about 10 times in the last 10 minutes!

And I've seen about 10 supposed major version upgrades of Firefox in the past 10 minutes, so it evens out.

Re:Honestly?

[Spy Handler]

Google Chrome is on version 23, so I suspect Firefox won't rest until they've at least achieved parity with Chrome in the "Browser Version Number Wars".

Re:Honestly?

[Shikaku]

http://www.mozilla.org/en-US/firefox/organizations/all.html

ESR versions are yearly if you care so much about fast releases.

Re:Too many revisions chased me away

[Anonymous Coward]

Try Firefox ESR. I'm still running 10.0.11

Re:Too many revisions chased me away

[Anonymous Coward]

So did you switch to ice weasel or what?

Re:Honestly?

[Malc]

And still a tired old monolithic app. I switched to Chrome eight months ago, and although it uses a lot of memory it does give me the ability to properly manage its memory and CPU usage: it's so much easier to identify pages to kill when they're running in their own process space. Not only does this allow me to selectively reduce the app's memory footprint, but I can conserve battery life on my laptop by easily culling busy pages.

Stupid

[Anonymous Coward]

These numbers are meaningless now. It's just stupid. It would almost be better to start calling releases stuff like "firefox 20130108" or something, at least you'd have a logical indicator of how recent the version is without having to look it up on wikipedia or something.

Re:Honestly?

[Tumbleweed]

Feel free to use the LTR and only upgrade once a year if you like. Nobody's forcing you to upgrade with every release.

I miss classic tabs in Android

[rockerito]

Firefox for Android lacks tabs. I mean, you have them in a sidebar as big screenshots of the website, and there is no option to use classic tabs at the top of the window. I hope they fixed it with Firefox 18, will try it later today. Otherwise, it is a great browser and the only one in which I can use flash for android jelly bean (though I had to install flash by hand). Chrome and Dolphin disabled flash, even if it is installed, for jelly bean.

Quit whining

[mcrbids]

Honestly, your whining is counterproductive.

Firefox is following a standard open-source style policy of release early, release often and as a vendor following this exact mantra, I see that although I do hear a lot of whining from some of our (typically more backward) customers, we are able to evolve to meet new needs better than our competitors which has allowed us to grow at a sustained rate better than 50% per year for years on end.

Many of our meetings with clients start with whines about how they have trouble keeping up with all the changes, followed up by hours of specifying new changes and additions that they'd like, closing with my pointing out that all the changes that they requested will be released as developed and them having to keep up with them as they are made available.

Perhaps it's necessary for some people to see improvements in a bad light, but if you really don't like it... leave! Go use some product that doesn't update at all if you want. I hear you can still find Firefox 3.6 binaries if you look hard enough. Even Chrome updates constantly.

Re:Quit whining

[girlintraining]

Honestly, your whining is counterproductive.

And your narrow-mindedness is annoying. Do you know why Microsoft only releases patches once a month for its operating systems? Because corporate environments can become violently ill when something is updated without it being tested first. You call it whining and say your customers are backward, but I see customers that prefer to not jump in head first into a pool they can't see the bottom of. That's why Chrome, Firefox, etc., aren't the most-used browsers.

Hell, even Linux has a stable and development branch. With Firefox, the stable branch might as well be the development branch. Plugins become outdated with every new release because they changed something in the API, or the way this feature works, or what that variable contains.

And for the record, a product that doesn't need to be updated is something some programmers strive for: It means they've made something that does its job so well there's no need to change it. That's not whining: That's damn good engineering.

Re:Quit whining

[Bill_the_Engineer]

And your narrow-mindedness is annoying. Do you know why Microsoft only releases patches once a month for its operating systems?

And your complaining about the mainstream version of Firefox while ignoring the existence of the enterprise version of Firefox makes your argument disingenuous.

Here let me get you started: http://www.mozilla.org/en-US/firefox/organizations/

Re:Quit whining

[Lucky75]

Firefox has a version that releases less often for corporate users. Also, Chrome does the exact same thing, sans the alternate version.

Re:Quit whining

[Anonymous Coward]

I can understand your reasoning.

However, it seems to me - I am Firefox user myself - that Mozilla folks executed this paradigim in a somewhat faulty way regarding extensions API. Extensions are constantly broken and as far as I know, current policy to avoid it is to set 'maxver' metadata to 99 or something like that. This seems wrong and Mozilla fails to see and solve the problem.

Re:Quit whining

[girlintraining]

And your complaining about the mainstream version of Firefox while ignoring the existence of the enterprise version of Firefox makes your argument disingenuous.

I'm well aware of it. It has support for all of... 1 year. Also, to quote directly from the same page: "Backports of any functional enhancements and/or stability fixes are not in scope."

So, who's being more disingenuous here... the person who makes the argument that "release early, release often" may not be suitable for all applications, or the guy that handwaves the argument, claims a lie of omission, and then makes a lie of omission of his own? Stupid facts, getting in the way of a good internet roasting...

Re:Quit whining

[Anonymous Coward]

Because corporate environments can become violently ill when something is updated without it being tested first.

That is why Firefox ESR releases exists.

Plugins become outdated with every new release because they changed something in the API

You meant extensions, not plug-ins. Your information is outdated. Except in rare cases, most extensions are compatible by default now.

And for the record, a product that doesn't need to be updated is something some programmers strive for: It means they've made something that does its job so well there's no need to change it. That's not whining: That's damn good engineering.

That may be true for Solitaire or TextEdit, but the web is constantly evolving, new standards mean new requirements, new customer expectations mean new requirements. Imagine trying to use Mosaic today.

Re:Quit whining

[Vaphell]

That's the reality of the web. People want to use css3, html5, svg, faster javascript and what not now, not in 1 year, maybe.
I don't really pay too much attention to what companies want, if they had their way we'd be still using IE6.0

Re:Quit whining

[mcrbids]

Do you know why Microsoft only releases patches once a month for its operating systems? Because corporate environments can become violently ill when something is updated without it being tested first.

We even offered to have a "stable" release version with updates only every 1 to 6 months, and have every released version have a 30 day trial period so that they could preview changes. We asked a 5% premium for this service. We thought as much as half of our client base would go for it based on the loud verbal feedback. But as soon as our clients found that they were choosing between having last year's product, totally stable with no updates or getting the new one with all the latest new features, bells, and whistles, guess how popular this option was? How many clients do you think signed that contract?

Not one.

My "narrow-mindedness" comes from my past experience... so now we listen to the whining carefully, and try to identify ways to better disseminate our change logs.

And for the record, a product that doesn't need to be updated is something some programmers strive for: It means they've made something that does its job so well there's no need to change it.

It's also a sign of a stagnant industry/marketplace. Needs change as circumstances change, and if the software doesn't change with the customer, it tends to disappear.

Re:Quit whining

[mlts]

Web browsers are the front line in the security fight too. It used to be that the biggest point of attack was the firewalls. Now with packet inspection on even the inexpensive home routers, as well as IDS/IPS appliances being relatively inexpensive, going in via the Internet is fairly difficult.

However, Web browsers and add-ons are the gold rush for malware authors. Add-ons until recently tended to be written at best to support functionality; security was an afterthought, if there was any thought to it at all.

Having a Web browser constantly updated, especially if it is response to potential new attacks against it or add-ons is a good thing. It is one of the few software programs on client machines that is always in contact with untrusted and potentially malicious code on a constant basis. Even trustworthy sites can have ad servers which expose visitors to malicious code.

It is understandable to have a stable browser in the enterprise. However, even IE is updated monthly on average, perhaps more if an out of band patch is required.

The best compromise would be something like FirefoxADM, where on the internal WSUS patch release, push out Firefox as well.

Until the Web browser stops becoming the main staging ground for intrusions, having constant updates is a good thing.

Re:Quit whining

[Bill_the_Engineer]

If you are going to accuse me of making a lie of omission, you need to not base that accusation on another "lie of omission". Let me quote the whole paragraph below with the omitted portion in bold:

"Maintenance of each ESR, through point releases, is limited to high-risk/high-impact security vulnerabilities and in rare cases may also include off-schedule releases that address live security vulnerabilities. Backports of any functional enhancements and/or stability fixes are not in scope."

The above policy seems reasonable and addresses the concerns of the enterprise users.

I still don't understand how pointing out that your argument completely ignores the existence of the ESR is a "lie of omission". I'll throw it in the "grasping at straws" bin.

Re:Quit whining

[Anonymous Coward]

Do you know why Microsoft only releases patches once a month for its operating systems?

Do you know that Microsoft releases patches for its operating systems once a week, specifically every Tuesday? Yes, the Wikipedia article implies once a month, except that really isn't the case any more -- they're deployed weekly, except for critical/uber-important security holes, which are pushed out ASAP. Things like "general security issues", even for older OSes like XP, are still pushed out weekly.

This is the 2nd thread I've seen from you talking about Microsoft as if you know the place. I get the impression you may have worked there or knew someone who worked there who gave you a lot of misinformation. I worked there for a total of 7 years (non-linearly; 2 years as a contractor, then many years later, 5 years as a FTE) and it disappoints me to see so much misinformation being spread. :(

As for your opinions on Firefox release schedules, for corporate environments your opinion is spot on - the Firefox release model does not work well in corporate environments (at my past job we used Firefox 3.x exclusively, despite Firefox double-digit builds being available, because upgrading would have caused more issues than sticking with what worked), and their "Enterprise" edition is not realistic/pragmatic in the least. But that has absolutely nothing to do with which browser is "used the most".

As for product which lack updates indicating stability or reliability - this is often nonsense spewed forth from mouths of people who are managerial and not actual engineers. Real engineers know that deep down inside there are always bugs, and that updates to fix those are always warranted, no matter how long it has been since the previous update. Here's a good example of a bug that lasted almost 20 years; does this indicate stability to you? It shouldn't.

Re:Quit whining

[Cinder6]

Do you know why Microsoft only releases patches once a month for its operating systems? Because corporate environments can become violently ill when something is updated without it being tested first.

I don't use Firefox, and I know it's popular to bash its rapid release schedule for some reason, but...

Firefox 14.0: June 26, 2012
Firefox 15.0: August 28, 2012
Firefox 16.0: October 9, 2012
Firefox 17.0: November 20, 2012
Firefox 18.0: January 8, 2013
Firefox 19.0: February 19, 2013
Firefox 20.0: March 26, 2013

Average is well over a month for each major version number. Granted, 13.0 came out less than a month before 14.0, but that was several versions ago, and is the only version in 2012 that had such a short lifetime. Mozilla doesn't ship new major versions that fast. There are point releases, but not as many as you'd think (only 1 or 2 per release).

I don't see why people are upset about Firefox's release schedule. It's similar to Chrome's, which seems to largely get a pass. Then again, Chrome does silent updates. I assume (hope) Firefox does the same--is this not the case? If not, then that could get annoying.

Re:Quit whining

[Bobfrankly1]

Chrome can't even handle roaming profiles without violent crashes and bugs. Please don't suggest that it's corporate/enterprise friendly. There are many threads which will testify otherwise and regardless how much Google claims the issue is fixed, it's not.

Re:Quit whining

[Anonymous Coward]

If you are going to accuse me of making a lie of omission, you need to not base that accusation on another "lie of omission". Let me quote the whole paragraph below with the omitted portion in bold:

"Maintenance of each ESR, through point releases, is limited to high-risk/high-impact security vulnerabilities and in rare cases may also include off-schedule releases that address live security vulnerabilities. Backports of any functional enhancements and/or stability fixes are not in scope."

The above policy seems reasonable and addresses the concerns of the enterprise users.

I hate jumping into someone else's argument but as to what you pointed out: Does it really? Say someone compiled in a check that if firefox had been running for more than 5 minutes, then it crashes. That isn't a security issue any any real sense of the word, which means that, under that policy, it would not be fixed. However, that would be a HUGE concern for enterprise users. Now I know that example is a little extreme, but the point remains that any issue that causes the program to crash will not be fixed. So the choice becomes to use either a version of the program with known crashes that were serious or common enough to be fixed OR to use the constantly moving target that could change UI elements, how pages appear or internal apps work. In addition, once the ESR is up for the year, then everyone will suddenly lurch to whatever version of firefox was current at the time and the whole cycle starts over only with new and different set of known crashes.

Re:Quit whining

[Billly Gates]

Still not corporate friendly. Release schedule not the enemy anymore more than the lack of AD and deployment tools.

I think an ancient version that never changes is last decades news for IT. Why can't things change often? During Netscape this was never a problem. As programmers support standards they will just work with whatever browser at whatever version. Only IE 6 - 8 do not do this. IE 9 and 10 so far are breaking the pattern. Just a few things off in IE 9 that are fixed in IE 10, nothing like hacking custom CSS code with hack after hack and hidden features which wont work in a future release of IE.

The newest apps that are certified with Firefox 3.6 will run on pretty much anything now. But until there is a way to push Firefox on +30,000 workstations where an admin writes a group policy once and is done with after still far away.

Until then IE is a better bet for the corps. I do not care about the longer release unless the users use special plugins that can break or if the app is old and only for IE 6 or 7.

Re:Quit whining

[Anonymous Coward]

I believe they implemented changes that were supposed to fix this issue a while ago, I think it was in version 10, but may have been a few versions after that.

Re:Quit whining

[betterprimate]

That's why Chrome, Firefox, etc., aren't the most-used browsers.

You may not have noticed, but they *are* the most-used browsers. http://en.wikipedia.org/wiki/Usage_share_of_web_browsers

Re:Quit whining

[Bill_the_Engineer]

Except that the ESR was the consumer version at some point of its lifetime. If the consumer version had the issue that you described it would have been fixed at the next point release. Basically the ESR just extends the major version lifetime, it is not a separate product.

When the next major version of firefox is released, only the browsers on the consumer release channel will get the update. The ESR versions will not see the update. Later on if a major security vulnerability is discovered in the ESR branch, Mozilla will issue an update to the ESR branch. Also if that update "broke" the ESR branch I'm pretty damn sure Mozilla will push out a fix. I believe the number of updates to the ESR branch once the consumer release channel jumps to the next major version will drop to a very low number... (if any).

The ESR allows the IT department to manage the update process which is what the GP was alleging was missing therefore less enterprise worthy than windows.

Re:Quit whining

[Anonymous Coward]

You should arrogant, rude, and hateful.

Re:Quit whining

[DragonWriter]

Firefox is following a standard open-source style policy

Its more an Agile or Lean approach than an Open Source one; its development methodology, not licensing structure, that is involved.

Re:Quit whining

[LordLimecat]

Firefox has admitted basically forever that "Corporate is not our target". If it were, we would have all sorts of things: ADM templates / GPO support, MSI installers, longer support terms, support for the OS certificate store (AKA an easy way to seed trusted root certs), support for smartcard readers, etc.

Incidentally, Chrome has all of that, and I believe you can even pin a certain version if you need to.

If both of those are unacceptable, use IE. Your environment is not one that Firefox has ever stated a commitment to supporting; their goal is to make a good general purpose browser, and if thats no good to you then oh well. These changes theyre making are generally a REALLY good thing for the average internet user.

Re:Quit whining

[Anonymous Coward]

V17 couldn't even properly run Selenium IDE on my machine so call release early release often what it is: Mozilla Firefox is perpetually in beta.

Re:Quit whining

[mrbester]

postMessage working across tabs / child windows would have been nice for IE10. IE 8+ supports just iframe communication and only strings can be passed, (W3C spec says *anything* can be passed) and IE9 silently fails if you try to polyfill it to support simple objects.

Firefox first supported it - fully - in 3.6...

Re:Quit whining

[LordLimecat]

Chrome has full blown official MSIs, ADM templates, and full GPO support. It supports pulling in the system proxy and certificate store. It supports smart-card readers.

It is, aside from IE, the most corporate friendly browser out there. That you think otherwise makes me think you havent done much research on the matter; certainly its lightyears ahead of firefox in a corporate environment. Its only big failing is that it utterly dies when used in a TS scenario, and I have a feeling it is an architectural problem that cannot be resolved without a rewrite.

It appears that there are workarounds for the roaming profiles issue, but dont say "cant even handle roaming profiles" as if thats a gross failing-- a lot of apps fail horribly when run from a network share. You ever hear Microsoft's stance on using Outlook with network-stored PSTs?

Re:Quit whining

[LordLimecat]

Needs change as circumstances change, and if the software doesn't change with the customer, it tends to disappear.

Or become entrenched, irreplaceable, and utterly loathed by all of its users.

Re:Quit whining

[yurtinus]

I don't care about the release schedule, I just miss meaningful minor and subminor version identifiers. Going from version 2 to version 3 used to indicate there was some significant change, rather than the latest batch of bug fixes. How are you supposed to get excited about the new features they're cooking up for Firefox 22? I hear Firefox 27 is going to be *groundbreaking*

I know it's just a preference thing and we've all gotten used to it - but I much prefer to see subminor version increments for bug fixes, minor for noticeable new features, and major for significant changes. Look at it this way - I remember some of the changes between Firefoxes 1 and 2 (or 2 and 3) - but can't for the life of me tell you what's changed between say, versions 7 and 18.

OK, my rant's done. Thanks for listening.

Re:Quit whining

[sjames]

Plugins become outdated with every new release because they changed something in the API, or the way this feature works, or what that variable contains.

I keep reading that over and over, but I also keep NOT seeing it happen when I update Firefox.

You must either have a metric assload of plugins installed, or you use a very poorly written one that keeps breaking.

Re:Quit whining

[Billly Gates]

IE 10 is pretty darn compliant. MS used to put in technology in IE 6 that the W3C changed which made the browser incompatible. As a result MS wont put it in IE until it is approved or at least recommended which is why is has a slightly lower score than Firefox in html5test.

Put standards do not count. It is the apps that the corps need and practically of deploying to thousands of computers that many are in different continents. Until Firefox supports this it wont matter. IE is still the best corporate browser until Firefox and Chrome come up with better tools.

At least you can run other browsers now at work if your boss is not cheap and has upgraded in the last 2 years or so.

Re:Quit whining

[girlintraining]

It's also a sign of a stagnant industry/marketplace. Needs change as circumstances change, and if the software doesn't change with the customer, it tends to disappear.

IBM is still selling mainframes. Your argument is invalid.

Re:Quit whining

[Anonymous Coward]

Which is a perfect example of a stagnant industry, especially compared to web software.

Re:Quit whining

[Anonymous Coward]

Are you *really* implying that IBM mainframes don't change, or don't change much?
Do you realize that in recent years there has been a complete new release of z/OS *every year*? And that new hardware models are appearing constantly?

Even missing alternate releases (which is the most you are allowed to do for supported upgrade paths) we** are on a constant upgrade cycle alternating between z/OS and CICS/TS upgrades. And yes, some upgrades break applications (although typically IBM give good notice/warning for this).

**Major outsourcer running IBM mainframes for multiple customers.

Re:Quit whining

[Anonymous Coward]

Firefox is following a standard open-source style policy of release early, release often

Linux versions include 3.6.11, 3.4.24, 3.2.36, and 3.0.57. If Firefox's version numbering scheme was a standard open-source policy, Linux's major version number would be 3 + 11 + 24 + 36 + 57 = Linux 131.0 or higher.

Re:Quit whining

[Anonymous Coward]

The mainframe IBM is selling today is a much different beast than what you probably think of as a mainframe.
 
But I know... I know... you're an expert on everything so you must be right.

Re:Quit whining

[Bobfrankly1]

It appears that there are workarounds for the roaming profiles issue, but dont say "cant even handle roaming profiles" as if thats a gross failing-- a lot of apps fail horribly when run from a network share. You ever hear Microsoft's stance on using Outlook with network-stored PSTs?

and yet Outlook runs with network-stored PSTs without skipping a beat. I've tried resolving the issue through Google AND their forums and most of their support hasn't a clue. As I understand it, is an architectural issue with the multitude of processes trying to perform concurrent operations on the same files, although this is a rough approximation from the only engineer there who seemed to have a bead on the issue.

That you think otherwise makes me think you havent done much research on the matter

I've done much research on the issue. I wanted it to work and be reliable. I used their group policys and written some of my own in attempts to resolve the issues. Chrome is great on a single computer and I use it often, but it's failings with network shares (as in TS as you mentioned) are loud and spectacular, and suggesting users use different tools in different environments (local machine vs ts) is a recipe for being replaced. :/

I wonder...

[jfengel]

I set it up to use Chrome as my PDF viewer. (Which wasn't easy, since the nonstandard way Chrome installs itself under Windows meant that it didn't show up on the list of programs.)

I wonder if it's going to override that setting when it updates itself. I don't really care, as long as it works. I just liked keeping my system clean and didn't want to download Adobe if I didn't have to.

Re:I wonder...

[ShaunC]

I wonder if it's going to override that setting when it updates itself.

Doesn't seem that way. I had the Foxit Reader plugin installed, and after upgrading, PDFs still opened in Foxit. Quite frankly I can't figure out where the built-in PDF reader even is; I uninstalled Foxit and if I try to load a PDF, Firefox now just prompts me to save the file.

Re:I wonder...

[Cyberax]

You need to set pdfjs.disabled to 'false' in about:config. PDF reader seems to be disabled by default for old installations.

Re:I wonder...

[Anonymous Coward]

Posting ANON to avoid undoing moderation (+1)

"""Update: The built-in PDF viewer is not included in this release, although it was in the Firefox 18 beta. We have updated the article to reflect that."""

So, no PdfReader included yet

Foxit Rocks

Re:I wonder...

[Anonymous Coward]

Apparently it's not enabled by default.

I had to set pdfjs.disabled to false in about:config and then PDF to "Preview in Firefox" in Preferences -> Applications to get it working.

Re:Too many revisions chased me away

[Spy Handler]

Not any more. My Firefox just updated itself to ESR 10.0.12 while reading your post!

Wasn't this supposed to happen silently?

[viniosity]

I recall reading that the updates would start happening silently in the background. I still appreciate the announcement, but my guess is that next week I'll go to the About Firefox section and find that it waited until just then to download the update to version 18. I much prefer the way Chrome handles it where I go to check and find out it happened while I wasn't paying attention. Where's the hold up on making that happen? Did it happen and I just haven't flipped a setting?

Re:Wasn't this supposed to happen silently?

[ChronoReverse]

It already happened. Check your settings to see if it's turned on.

Re:Wasn't this supposed to happen silently?

[Anonymous Coward]

For reference, it's under options > advanced > update

On Windows, if you've got the 'use separate updater' box checked, you might also want to make sure that the Mozilla Maintenance Service is set to run automatically in services.msc.

Re:Wasn't this supposed to happen silently?

[viniosity]

It already happened. Check your settings to see if it's turned on.

As suspected, my settings are correct. When I went to About Firefox I got the dialog.. downloading update 0 of 24.2MB downloaded. Seems like they still have work to do.

Re:Wasn't this supposed to happen silently?

[gQuigs]

Something isn't quite right there. For whatever reason you are getting a full update, when it should have tried a binary diff first (~3 to 6MB).

It's gotten much better for me over the last couple releases. I used to have to do what you do in checking manually but haven't for a while. (That's when I maintain Windows, Ubuntu's Firefox obviously updates with the system)

Re:Wasn't this supposed to happen silently?

[ChronoReverse]

Might just be staggered release then.


The update was just released and it's entirely reasonable it could stagger you a couple days even before auto-downloading.

Re:Wasn't this supposed to happen silently?

[Anonymous Coward]

Might just be staggered release then.
The update was just released and it's entirely reasonable it could stagger you a couple days even before auto-downloading.

Exactly

Re:Wasn't this supposed to happen silently?

[ProbablyJoe]

Actually, it does seem to update itself on Ubuntu, outside of the package manager. Mine updated overnight on my work computer, but I have no root/sudo access to install programs normally. Not entirely sure how it does that.

IonMonkey

[girlintraining]

Seriously, were y'all drunk when you came up with that name? It conjures up images of some kind of celestial primate flinging high energy particles about. Firefox at least sounds like something that could be found frolicing about in heavily wooded areas.

Re:IonMonkey

[Zontar The Mindless]

It conjures up images of some kind of celestial primate flinging high energy particles about.

Sounds to me like something that'd be hanging out with a dragon.

'Firefox' just makes me think of that lame Eastwood film.

Re:IonMonkey

[cruff]

It conjures up images of some kind of celestial primate flinging high energy particles about.

At least that's better than the stuff they usually fling around...

Re:IonMonkey

[A10Mechanic]

"Firefox" , . Remember, you must post in Russian.

Re:IonMonkey

[Anonymous Coward]

Firefox at least sounds like something that could be found frolicing about in heavily wooded areas.

I guess it's currently frolicking in Australia.

Re:IonMonkey

[caspy7]

A browser's JS JIT compiler isn't exactly a user-facing component.

Re:IonMonkey

[Anonymous Coward]

LOL, that takes me back.... Firefox the movie with Clint Eastwood in it as a pilot.

Yes but Nightly can drink beer now.. legally

[Lawrence_Bird]

Happy 21.0a1!

PDF viewer?

[Anonymous Coward]

Sounds like good functionality for a plug-in.

Re:PDF viewer?

[jones_supa]

Yep, and reading PDFs inside Firefox has already worked using the plugin which ships with Adobe Reader.

Re:Too many revisions chased me away

[Anonymous Coward]

They've sapped all the fun out of major version changes. "Ooh, new version 5.0! I can't wait to see the huge amount of major improvements they've made!" Now it's like they assign a new major version number for every little fucking thing. Whores.

There is no PDF viewer, yet

[ShaunC]

From another article:

One feature that didn't make it into this release, by the way, is Mozilla's new built-in PDF reader. While the organization has been working on this for a while, it will only make it into the beta release that's expected to arrive on Thursday.

Re:There is no PDF viewer, yet

[MarcoAtWork]

there sure is, although it wasn't on by default for me: enable pdfjs inside about:config and set the pdf in content to 'preview in firefox'

Re:There is no PDF viewer, yet

[Anonymous Coward]

I prefer to open PDFs and other non-HTML documents with dedicated viewers/applications. I never understood the attraction of embedding everything in the browser window, dedicated applications usually have better user interfaces for the file type.

Re:Too many revisions chased me away

[Anonymous Coward]

Mozilla chased me away a long time ago due to add-ons continuously breaking on updates and Mozilla blaming the add-on developers.

Still crashy?

[Anonymous Coward]

Does the entire browser still lunch itself when a plug-in fails on some page?

Chrome showed us that the browser must be robust and generally survive glitches caused by individual sites. Please learn from Chrome and adopt that model.

Re:Still crashy?

[Anonymous Coward]

Here's a spectacular crash.

Re:Still crashy?

[Lucky75]

Nope, plugins are in a separate process. Have been for quite a while now....

Yawn

[sunking2]

I stopped really caring years ago about what happens to be on a computer. They all pretty much work at this point. Just like at a restaurant when I ask for a Coke and they say, 'Pepsi ok?' and I just say sure and go happily on with life.

Re:Yawn

[Anonymous Coward]

If you stopped really caring years ago about what happens to be on a computer then why are cluttering up this thread with retarded posts about how you don't care? Why do you even visit slashdot at all since it's very much for people who *do* care about these things?

PDF.js

[oever]

The PDF viewer in Firefox, PDF.js is an amazing piece of software. It is written entirely in JavaScript and runs in the same sandbox in which a webpage runs. So it is very safe. The layout accuracy and speed of PDF.js are simply amazing. Text selection happens just like it does in the browser. Some PDF viewers only allow you to draw a rectangle on which to do OCR. PDF.js simply lets you select the glyphs.

This viewer has been available as an add-on for a while already.

Re:PDF.js

[Anonymous Coward]

With PDF.js Firefox will "eat" a lot of memory hundred of megabytes or even one GB. A solution cold be a special (firefox) profile only for PDF.js.

Re:PDF.js

[Anonymous Coward]

Firefox will "eat" a lot of memory regardless of what you do with it. Even without extensions, any memory usage improvements mozilla may or may not have implemented do not play any role in long term usage scenarios (that is, leave the browser open for a couple of weeks), it'll end up using 1-2 GB of ram.

Re:PDF.js

[DirePickle]

I can not comprehend of anything worse than a PDF viewer written in javascript. The worst of both worlds, together in one package!

Re:PDF.js

[Anonymous Coward]

It may be built in but for now it is disabled...

Reference here: https://wiki.mozilla.org/Show_PDF_Inline/

Cheers

Re:PDF.js

[Anonymous Coward]

You can always use Adobe Reader - the best of its kind ;)

Re:PDF.js

[Anonymous Coward]

I can not comprehend of anything worse than a PDF viewer written in javascript.

Clearly, you have never used Adobe's PDF viewer.

Re:PDF.js

[Billly Gates]

Have you looked at recent benchmarks?

I posted a story last July and Firefox handled the most amount of tabs with the least ram. IE 9 surprising wins too if you have just 1 - 2 tabs. Chrome now is the new pig. My, have things changed in just 1 year.

Re:PDF.js

[Tailhook]

I can not comprehend of anything worse

That's not much of an imagination. You can embed Javascript in PDF...

It's turtles all the way down.

Re:PDF.js

[Pascal Sartoretti]

It looks great, even if not 100% ready (I guess this is why it is still an add-on and not a part of the standard installation).

What I like most about it : no more PDFs cluttering my "Downloads" folder. And if I really want to save the PDF, it is only one click away.

Great job, guys.

Re:PDF.js

[DuckDodgers]

They did that because an awful lot of PCs hacked by a website were hacked through security flaws in the PDF viewer. Writing the PDF viewer in Javascript means that the Mozilla developers only need to make Javascript in Firefox secure to protect the machine from intentionally badly formed PDFs, and of course they already needed to secure Javascript so that's no extra security work.

As a scripting language, Javascript is still slow compared to something like well-written C++. But Firefox 18 is pretty close to the latest version of Chrome for Javascript performance (e.g. arewefastyet.com ), so I bet the PDF viewer in Javascript works quickly enough.

Re:PDF.js

[Anonymous Coward]

The fact firefox performs better now then other browsers (which I've never used) does not automatically make Firefox's excessive ram consumption justified.

Re:PDF.js

[Anonymous Coward]

Kernel can handle 20-30-40 tasks better than a single large task with a big workload, from point of view of CPU time, RAM and swap memory. Unfortunely Firefox Electrolysis (e10s) was postponed indefinitely.

Re:PDF.js

[xtracto]

http://blog.kowalczyk.info/software/sumatrapdf/free-pdf-reader.html

You are welcome.

Re:PDF.js

[Dwedit]

PDF viewer written in Javascript, and it's a file format that can contain embedded Javascript...
Yo dawg.

Re:PDF.js

[Anonymous Coward]

PDFs don't just download for me; Firefox asks if I want to view them or download them.
It's been that way for many years.

Re:PDF.js

[Vegemeister]

speed of PDF.js are simply amazing.

With the handicap of being an interpreter for a layout-oriented scripting language written in an interpreted scripting language, I suppose one might choose to call the performance amazing.

Re:PDF.js

[Anonymous Coward]

Very safe? Are you saying it doesn't support JavaScript?

Re:PDF.js

[LordLimecat]

I can: Unpatched acrobat installations infecting all my family / friends' computers with malware because PDF plugins are the bane of all security.

Auto-updating flash and pdf handlers is one of the reasons ive been suggesting Chrome to everyone, and why it has reduced the number of viruses I deal with enormously. This is wonderful that firefox is jumping on the wagon. Now that both Chrome and Firefox also block vulnerable java, we just need a long-term fix to the flash question in firefox.

Re:PDF.js

[LordLimecat]

Correcting link: it is case sensitive...
https://wiki.mozilla.org/Show_PDF_inline

Re:PDF.js

[Anonymous Coward]

What? Are Mozilla paying you for advertising PDF.js?
It is a great idea to sandbox the pdf-viewer, and the speed is decent for an application written in JavaScript, but I have yet to find a single pdf beyond some simple sample text that it displays correctly.
I hope they'll make some major improvements before actually enabling it by default.

Re:PDF.js

[oever]

No, I'm not being paid. I work on a similar project: WebODF. I wrote the post so someone (thank you) would ask me what is in it for me and I could plug this project.

Seriously: my experience is that PDF.js works acceptably for most PDFs I threw at it. That included large PDFs with designer layout and scientific papers. Granted, poppler (okular, evince) is still way faster in rendering, but I enjoy PDF.js because it is good enough and I know the work it took to make it and can see the improvements they are making still. Trying to write a desktop type application in the browser makes one appreciate a good one like PDF.js.

Re:PDF.js

[tyrione]

I'd rather they combine cairographics and poppler in a XUL UI [if necessary] to manage the inlined PDFs. Much better code source, tested even with Adobe and far more extensible support for unicode, etc.

Re:PDF.js

[AmiMoJo]

But Firefox 18 is pretty close to the latest version of Chrome for Javascript performance

In my experience FF does reasonably well in benchmarks but chugs like hell on real sites that Chrome runs smoothly. Clearly the raw performance of the JS engine is not all there is to it.

Re:PDF.js

[AmiMoJo]

Does it support PDF DRM though? All the best PDF viewers ignore the stupid "may not print" and "may not copy text" flags.

Here's a hint Adobe: the PrtScn key still works and I can re-type anything I can read, or just open with Ghostscript and bypass your stupid restrictions on automation.

Re:PDF.js

[DuckDodgers]

The only time I notice a performance difference between the two is when I use a bookmarks folder to open six or eight sites at the same time. Firefox takes a while to finish - and I'm assuming the problem is related to writing the files it's receiving to the cache folder, because the problem is much worse if I'm copying large files when I do it. Chrome seems to do better.

At this point I think Chrome is still the superior browser, but I'm desperately hoping Firefox catches up completely or almost completely. I don't want a world where just one browser, even an open source one, is the undisputed king.

Re:PDF.js

[squiggleslash]

OK, I don't get it. PDF is a moderately good system for digitally representing paged media allowing the same content to be distributed formatted for printing (or scanned from paper with the layout intact) while providing a clean, digital, representation for viewing on a computer monitor. And Javascript is a fairly decent scripting language, logical, clean, and increasingly efficient. If there's a "problem" with "javascript" it's usually that most developers have to use it in a web browser, the limitations and poor APIs of which are generally what developers find annoying rather than the language itself.

So what's the problem with this exactly? And, even with the ugly browser API, how is it worse than loading an ugly, insure, C++ binary blob into memory to handle PDFs instead?

HTML5 and H264

[Anonymous Coward]

Question for anyone running the Windows nightlies with the native H264 support: is it stable? Having to use Flash as an HTML5 H264 decoder is one of the most pointless things I can think of.

OTOH

[dogsbreath]

Went to Chrome for a while to see what the buzz was about. Supposedly faster, cleaner, etc.

Got po'd when I couldn't configure it to operate the way I wanted it to. Just personal taste and not a criticism; to each their own, as they say. However, I did not see any improvement in responsiveness and, for me there was a genuine loss of functionality. Went back to Firefox and have been very happy. Sure it would be nice to have some process options but Mozilla seems to be doing a bang up job of dealing with the various issues that caused process hangs and memory leaks. I can't remember the last time I had to kill an unresponding FF process. Used to happen weekly, even daily. Kudos to the FF team.

For the most part the Firefox version changes have been transparent to me (well, except for tabs - grrrr - but I have been able to customize them to work the way I want). The update cycle is more or less the same with Chrome and IE. If they changed the numbering scheme so it went from, say, 10.17 to 10.18 instead of 17 to 18, there would be less reaction. Or maybe not. Anyways, it is not a huge issue.

Firefox is easily competitive with any other popular browser and is well supported. Don't think I will bother trying a change again for a while unless something truly game changing comes along.

Re:OTOH

[LordLimecat]

Once youve used chrome in a corporate environment, you never go back. Supports a lot of corporate control-- awesome GPOs, pulls proxy in from IE, MSI installers, and no admin-for-update required. Plus it tends to "just work" in a way firefox doesnt-- all of our corporate apps work flawlessly in chrome, including those which require a smartcard login (which Firefox doesnt know how to deal with).

And the "web application" feature is really cool-- works great with OWA and whatever other apps you have.

Firefox is better with extensions undeniably, but theres so much else going for chrome (like vastly better security, and STILL better updating) that Firefox remains a distant second for me. Though, Ill admit that their new html inspector is really slick, at least to this webdev-layman.

Re:OTOH

[MarcoAtWork]

I think 'vastly better security' is being able to run noscript + adblock + requestpolicy, I don't think you can do that in chrome, can you?

Re:OTOH

[LordLimecat]

I have "NotScripts", Adblock, and use Chrome's "click-to-play" functionality for plugins. I imagine theres something for requestpolicy.

Generally extensions ARE better on firefox than chrome, but I dont use many of them, and Chrome is better at what it does support-- being able to dynamically load, unload, install, and remove extensions is REALLY nice.

Security-wise, their implementation of "private browsing" is also the best-- other browsers tend to hit the disk cache at least temporarily, and can be sniffed with an undelete tool (like photorec). Chrome's private browsing operates completely out of RAM. Its also a lot better with plugin security and in-general sandboxing.

Both tend to be decent browsers, but unless you have really specialized extension needs, id give Chrome another whirl. Theres a lot to love about it.

Re:OTOH

[Anonymous Coward]

I imagine theres something for requestpolicy.

There isn't, because that works at the network level, and only Firefox has its own networking stack that extensions have access to.

Generally extensions ARE better on firefox than chrome, but I dont use many of them, and Chrome is better at what it does support-- being able to dynamically load, unload, install, and remove extensions is REALLY nice.

You're talking about restartless extensions, which have been in Firefox since version 4. Developers can choose either type of extension, and as the Add-on SDK allows more functionality, more developers are doing it the restartless way.

I was already skipping your repetitive posts waxing lyrical about all the extra corporate features in Chrome, but when you make a patently outdated and false assertion about extension installations, I know I can safely ignore the rest of your ignorant posts.

Re:OTOH

[dogsbreath]

Once youve used chrome in a corporate environment, you never go back.

Your experience, obviously, but not mine.

My company supports only IE for corp. apps/sites/etc. Frankly, it works just fine for the corp stuff and there is no compelling need to go to Chrome. Again, I have compared user experience and although Chrome did not give me any problems, it did not do anything for me either so why bother?

For anything else, I use FF. Not hard to have multiple browsers available.

Chrome is OK; I just find it to be a bit of a PITA with things that I want to configure that I can't.

Re:OTOH

[metrix007]

Being able to update without admin is not an advantage in a corporate environment. Sharing IEs proxy is also not an advantage.

The only real advantage Chrome has is with the better private browsing mode, and the better security architecture. Still, as long as Firefox is kept updated and a workstation is sufficiently locked down, it isn't that much of an issue.

Maybe when Chromes extension framework catches up to Firefox's, I'll give it another look.

Re:OTOH

[LordLimecat]

There isn't, because that works at the network level, and only Firefox has its own networking stack that extensions have access to.

This really isnt true. NotScript does basically everything that NoScript does, the only thing it doesnt do is cross-site whitelists like RequestPolicy. This isnt because of an API deficiency AFAIK, its because they didnt create that functionality. Chrome certainly has its own network stack, and extensions can certainly access part of it, which is how NotScript and Adblock work to begin with.

I was already skipping your repetitive posts waxing lyrical about all the extra corporate features in Chrome

Youre free to consider me ignorant if you want, and if youre feeling generous you might point me to a page that informs me of how to use all of Firefox's corporate features. Last I checked the only MSI packages were third party, and the only GPOs were both sparse AND third party and I believe came under a restricted license. Firefox also completely ignores system certificate stores, which has made for "interesting" troubleshooting across a number of platforms as I have had to add certs first for the the OS (OSX, Windows), and then for Firefox's own special cert store.

So forgive me if I prefer the browser that is manageable without unofficial unsupported GPOs and dropping managing prefs.js files.

I am aware that it is possible to do restartless extensions in firefox, but in practice very few of them that I have seen actually use it. NoScript doesnt, nor does RequestPolicy.

Re:OTOH

[LordLimecat]

Being able to update without admin is not an advantage in a corporate environment.

It is when its trivial to disable through a GPO or pushed registry setting.

Sharing IEs proxy is also not an advantage

I fail to see why properly reading and using the system proxy setting is worse than Firefox's model of "screw the system setting and GPOs, we have our own way of doing things". I guess its kind of nice that I can bypass our proxy, but it doesnt recommend itself as a good candidate for roll-out.

Re:OTOH

[metrix007]

It's not the System proxy setting, it's the IE proxy setting. Name any other software aside from IE that uses it?

Re:Too many revisions chased me away

[Lucky75]

Pretty sure that doesn't happen any more with the new way they write extensions.

Bug Fix/New Feature 12 years in the making

[Tailhook]

They've landed the solution to this issue, first submitted in 2000. Clinton was still president.

Re:Bug Fix/New Feature 12 years in the making

[Anonymous Coward]

And I'm still waiting for them to provide better user interface controls for HTTP AUTH which would allow a user to see their "logged in" status and be able to log out of a site (expire their cached credentials for the site). This should have been done back in the nineties. Meanwhile, every web app in the world has had to implement their own (hopefully secure) login system. Oh well.

Re:Bug Fix/New Feature 12 years in the making

[Anonymous Coward]

That's nothing, they're still yet to implement Select/Copy of ::before or ::after text, from 1999.

Re:Honestly?

[gregmac]

I've been using Chrome for well over a year, and have had this discussion many times. Yes, Chrome uses more ram. But I can close a bunch of tabs, and it frees it up. Firefox, every time I try it and despite that it's memory management is "getting better", still eventually uses several GB of ram and requires that I completely exit and restart before it's freed.

My browser is one of the first things I start up when I turn on my PC, and generally stays open until my PC has to reboot for some reason (which may be anywhere from a week to a month). This is really only possible now because I use Chrome.

Old bug

[jones_supa]

Heh, they surely fixed an old bug in this release.

Re:Honestly?

[Anonymous Coward]

ESR, not LTR. Not that it matters all that much.

And the ESR does change with the regular line as they backport any security fixes.

I did just grab ftp://ftp.mozilla.org/pub/firefox/releases/17.0.2esr and ftp://ftp.mozilla.org/pub/thunderbird/releases/17.0.2esr.

Re: Firefox is not sandboxed!

[Billly Gates]

This is a major security risk if you ask me. Chrome and IE are and Mozilla is still behind. Flash luckily is now sandboxed which is a huge improvement but PDFs can contain nasty javascript exploits and without a sandbox could be a SECURITY NIGHTMARE.

I am sticking with Firefox ESR 17.01. It will be supported for a year and and want to see if my suspicions are right.

If my information is outdated feel free to correct as I am in the process of not recommending Firefox anymore unless the corporate system is still on XP. IE is much secure now in Windows 7.

a plugin full or security issue

[JcMorin]

Yeah and Adobe have a long reputation of having seriously security with their PDF reader. Wonder why they want to make it run without the plugin...

Re:a plugin full or security issue

[robmv]

Exactly, and they are not embedding a C based PDF library like Chrome, this is a new implementation (that needs a lot of testing, printing is still awful) made in Javascript, sandboxed by default

Re:Honestly?

[Anonymous Coward]

I do the same with Firefox with no problem.

Use whatever you like, but no modern browser has significantly poor performance with general use.

Re:Honestly?

[GNious]

My solution to the "restart firefox" - use the Aurora build: There is a new build every day, so you're suggested to restart daily :) ..and yes - my mac becomes noticable faster every time I restart FF.

Re:Honestly?

[dogsbreath]

My browser is one of the first things I start up when I turn on my PC, and generally stays open until my PC has to reboot for some reason (which may be anywhere from a week to a month). This is really only possible now because I use Chrome.

I call shenanigans. "[rebooting monthly] is ... only possible now [because of Chrome]" is just not true.

I'm running Win7-64bit on a laptop with 6G ram and I use Firefox. FF is always running and I very very seldom kill the process. Like almost never. I reboot about once a month and usually because of something non-related to Windows or FF crashing/hanging. Usually just a Win security update.

I run some heavy memory usage video editing apps and usually have a LOT of terminal windows open, along with multiple desktops. FF has not been an issue.

eh, your mileage may vary but that is my experience.

Re: Firefox is not sandboxed!

[Anonymous Coward]

Why do you need to embed the JavaScript in a PDF when you can just run the JavaScript exploit directly? Do you think that running a PDF viewer written in JavaScript is going to give the embedded JavaScript in a PDF some sort of special powers?

Problem is the update dialogs

[Anonymous Coward]

Ever since I started using Google Chrome I've gotten used to fact it doesn't give you annoying update / addon update dialogs. I tried firefox few version numbers ago, it still threw the dialogs! That is insane with this release schedule.

Please Mozilla, make dialogs by default off, and create option to enable them.

Where's my x64 Windows support dammit?

[Anonymous Coward]

Come on, x64 OS's were out in like 2004... it's now almost 10 years later and we're still clinging to 32-bit versions just for shitty plugins to keep working
Flash has x64
Java has x64
Nothing else is required.

I personally welcome any attempt...

[Mister Liberty]

...to bring about the demise of the dud that is called Evince;
they finally broke its last functionality under Linux Mint.

Re:Honestly?

[Anonymous Coward]

I do the same with Firefox with no problem.

That's nice for you, but it does not invalidate the point of the person you're replying to. I have the same problem he does: Firefox (and ONLY firefox, out of all the things I run) eventually grows to several GB and starts my system into swapping paralysis. Killing firefox instantly makes the system come back to its senses again. None of the other apps I run show this runaway-ram-usage behavior. My only add-on is adblock plus.

Why is that every time in Linux land that someone reports a bug, a million other people jump all over him and say, "but ... but.. but ***I*** don't see it! It must not exist!" Sheesh - there are hundreds upon hundreds of people reporting this same behavior. YOU are not the whole world.

Just switched to v18

[cvtan]

Now it runs but will not show any web pages. I just get spinning circular arrows. All pages are blank. :( I want v17 back!

Re:Just switched to v18

[mishikal]

I've noticed two bugs so far with firefox 18 -- It is extremely slow to start up, and pages take about 30 seconds to come up after starting FF. The other issue is that foxyproxy no longer works.

Re:Just switched to v18

[LordLimecat]

Sounds like a corrupt profile or installation. See if moving your firefox profile elsewhere for testing fixes the issue-- ive run into that a few times over the years.

Re:Just switched to v18

[Anonymous Coward]

Go to Help -> Troubleshooting Information, then click the "Reset Firefox" button.

gave up on Firefox a year ago

[CosaNostra Pizza Inc]

I use Chrome on my Windows machines and Chromium on Linux.

Re:gave up on Firefox a year ago

[Anonymous Coward]

I have both installed.

Re:gave up on Firefox a year ago

[Anonymous Coward]

You have no interest in the article and nothing to add to the thread, so I am curious. Why bother posting?

Seriously, I'm not trolling or trying to be mean. I am just sincerely curious as to the mentality behind this behavior. You are not alone, I see it a lot, but I was just hoping you could explain it to me.

Re:gave up on Firefox a year ago

[squiggleslash]

Cool story bro?

Re:gave up on Firefox a year ago

[CosaNostra Pizza Inc]

AC, I'm simply curious as to why people pick one browser over another. It usually comes down to personal preference. I use Chromium on Linux because I prefer the look and feel of Chrome to Firefox but Chrome uses its own proprietary flash player, which doesn't support DRM in linux. This is a problem for me when accessing Amazon instant video. Therefor, I use Chromium, which uses Adobe's Flash player.

Re: Firefox is not sandboxed!

[Billly Gates]

Why do you need to embed the JavaScript in a PDF when you can just run the JavaScript exploit directly? Do you think that running a PDF viewer written in JavaScript is going to give the embedded JavaScript in a PDF some sort of special powers?

The power to execute data? Yes. If Mozilla took it out people would whine their PDFs wont work. It was a moronic thing for Adobe to include. I use Foxit PDF because it allows limited javascript functions, is sandboxed, and on top of that will display a warning and run in safemode with javascript disabled by default. Mozilla does have many holes in Javascript fixed, but it is not fully sandboxed like the other browsers to save ram.

Whoa

[engineer_uhg]

No kidding about the faster javascript. Now Gmail is practically instantaneous.

Just switched to Firefox 17 ESR

[Dwedit]

I just switched to Firefox 17 ESR.
CacheViewer got broken by the upgrade from 17 to 18, and I don't want any more automatic updates that will break extensions, so I'll just stop automatic updates, and keep a browser that works, and will get updates only for security fixes.

Re:Just switched to Firefox 17 ESR

[Shikaku]

08 January 2013: This addon is not compatible with Firefox 18 (yet). Mozilla removed a key method in their API that this addon was using to retrieve entries from the cache. Implementing the new method for getting cache entries will not be a trivial matter. I will release an update as soon as I can, but it could be a while.

Re:Too many revisions chased me away

[nanoflower]

Agreed. I haven't seen any issues with add-ons since around version 4. I've been able to upgrade without any worries and have been on the beta channel since v15 due to the memory enhancements it had.

Re:Honestly?

[Anonymous Coward]

My Linux just updates and I leave it alone and don't try to be picky or find some LTR.

This eats a lot of time and some bandwidth (not free) updating 5 machines and FF & TB being over 20MB each nowadays.

Re:Honestly?

[nabsltd]

Firefox, every time I try it and despite that it's memory management is "getting better", still eventually uses several GB of ram and requires that I completely exit and restart before it's freed.

This was a big problem for me as well, but around version 15 or so, it looks like Mozilla finally fixed it.

I can now do my normal browsing and after a week, Firefox is using about 500-700MB and works fine. Before, it would be up to 1GB in a day or so, and would become sluggish and even completely non-responsive at times. I do run with NoScript and AdBlock Plus, so Flash doesn't start unless I explicitly click on the object, so YMMV if you have different browsing habits.

Re:Honestly?

[K. S. Kyosuke]

They still have to catch up on Emacs 24.2, and Emacs can do anything!

Re:IF web developers did their jobs ..

[Billly Gates]

Then we wouldn't have obsolete browsers in the office. IE 9 is the first good MS browser in 12 years! It is standards compliant and does not support the proprietary jscript shit that IE 8 still ties that corps can not leave from which is why corps are not upgrading.

    do not recall seeing corps violently opposed to upgrading Netscape or IE (pre IE 6) did you? Infact, the managers DEMANDED up to date browsers! IT workers who did not upgrade all the time were incompetent as the world moved on. IE 6 changed all of that. Even IE 8 has proprietary apis in jscript and VBscript that developers loving using to tie to other products. Even if the rendering engine is not atrocious like IE 6 is.

IE 10 is almost done for Windows 7 and scores to what Firefox 7 does HTML 5 and CSS 3 wise.

The world is moving on and it is time app developers STOP SUPPORTING LOCKIN. If your office choices to stay with IE 8 until 2020 then do not expect the web to support you. It is ridiculous and I wont wait until the 2020s to enjoy todays technology. Lets hope in the coming years as IE is standards compliant that the code will magically work with all other browsers as well.

Macs are in offices more than in any other time since the 1980s. A compliant intranet app that can run with Safari is a plus. The world is changing.

In Russian it is then...

[fallen1]

Well damn, the Cyrillic won't post into Slashdot so go here to see Zontar's post in Russian. ;-)

Last crash: 1 hour ago.

[Futurepower(R)]

Firefox is the most unstable software in common use. One percent of the time it crashes: Crashes per 100 Active Daily Users

I use it only because it has add-ons I need.

Re:Last crash: 1 hour ago.

[Vaphell]

no shit the browsers are not rock solid. They are ungodly complicated, probably 2nd only to games (which crash orders of magnitude more often) and they have to deal with copious amount of shoddy html and javascript all day long.

Re:Last crash: 1 hour ago.

[Anonymous Coward]

I would be upset but I'm too enamoured by the awesome crash stats page.

Re:Last crash: 1 hour ago.

[egranlund]

no shit the browsers are not rock solid. They are ungodly complicated, probably 2nd only to games (which crash orders of magnitude more often) and they have to deal with copious amount of shoddy html and javascript all day long.

I understand.

However out of all the browsers I've switched between using on a daily basis Firefox is by far the most crashy. Hell, even IE 10 is more stable.

Re:Honestly?

[UltraZelda64]

Nothing is as advanced as Windows Server 2008. NOTHING!
Everything else might as well just give up... or append the current year to their program's name before everyone else does.
Hey, it's still early in the year... there's still time!

Re:Update

[Anonymous Coward]

In the following 3 hours they released Firefox 19 and then Firefox 20, because marketing thought 20 sounded better.

Eventually people are going to realize these jokes about the rapidly increasing version numbers of Chrome and Firefox are far past their prime. Well, okay, most people have already realized this, but eventually all people will come to a similar realization. I hope...

Really looking forward to ESR 17 version!

[billstewart]

At $DAYJOB, the IT department policy used to be "IE6 Only", which everybody ignored and installed Firefox. Eventually they finally decided to support IE7 (and now support IE8, at least on Win7), and they installed Firefox on our machines the middle of this year. Unfortunately, it's the FF10 ESR, which broke my working environment (FF13 really did do a much better job of memory management, and since IT only supports 32-bit Win7, I can't just fix the problem by installing more RAM.) So I'm hoping they'll get moving and let us upgrade to 17 ESR real soon. (And given the latest IE bugs, I'm hoping they'll let us upgrade to IE9 or IE10 soon?)

My lab machines are mostly running Linux, where this is of course not a problem. And the Linux virtual machines on my desktop run relatively current FF, but there's not really enough room for a big enough VM. One of my coworkers installed native Linux on his laptop with a VMware Windows machine on top that's running the IT department official versions, which let him max out the hardware RAM and lets him do most of his work from Linux, which was at least somewhat helpful.

Re:Really looking forward to ESR 17 version!

[ls671]

One of my coworkers installed native Linux on his laptop with a VMware Windows machine on top that's running the IT department official versions, which let him max out the hardware RAM and lets him do most of his work from Linux, which was at least somewhat helpful.

I used to do the same in 2002. Funnily enough, IT support guys would come to my desk to install stuff and I had win NT running in a VM fullscreen and the IT guys never realized I was running linux as the native host.

Re:Really looking forward to ESR 17 version!

[viperidaenz]

IE8 will be supported by MS until 2020. Don't hold your breath

Re:Really looking forward to ESR 17 version!

[nobodie]

I managed to find a neckbeard with admin rights to get fedora on my desktop and win7 in a virt, couldn't get him to agree to KVM, but Oracle VB was accepted (while a neckbeard, he also drinks the green kool-aid and has to admin win servers and desktops mostly, so he is being nice to me overall). Once a week I go to win7 to use filemaker, and other win7 stuff (updates and av updates, etc). that part of life is good.

Re:Honestly?

[Chrutil]

Python 3000

Re:version numbers

[LordLimecat]

Theyre only getting publicity because slashdot bothers posting the update stories, which honestly is the only way Id know there was an update.

They switched models because its a BETTER MODEL. They can actually get useful features out more quickly than the old 1-year dev time. I dont know if anyone remembers, but the upgrade from 1.5 to 2.0 took like a year, and came with like 4 features-- a new-tab button, a completely messed up (still bitter) options GUI, and tab-close-undo.

Now we get about that many features in an 8 week dev period, and incremental speed increases to keep pace with chrome. Im failing to see how this is a bad thing; its keeping Firefox remotely relevant to Chrome who was kicking their hiney in features and speed for about 2 years.

Re: Firefox is not sandboxed!

[LordLimecat]

Unless Im mistaken, as pdf.js is written in JS, it IS sandboxed, and in a way that Acrobat is not / cannot be. Remember that Firefox's JS implementation has to be secure enough to interpret code from all over the internet, and has had a LOT more "battle-hardening" as well as a better security record than Acrobat.

I imagine the only downside is rendering speed, but it seems pretty quick.

Re: Firefox is not sandboxed!

[LordLimecat]

Relying on a third-party for security in what is seen as "standard functionality" seems like a really bad idea, especially given Acrobat's (and even Foxit's) security record.

Re:Honestly?

[ls671]

Python 3.0 (a.k.a. "Python 3000" or "Py3k") is a new version of the language that is incompatible with the 2.x line of releases....

http://www.python.org/download/releases/3.0/

Is Phyton a fork of php? it seems like it has some similarities anyway...

Re:Honestly?

[LordLimecat]

....You mean Server 2012, right?

Re: Firefox is not sandboxed!

[Anonymous Coward]

You missed the point.
If there's a security hole, it's in their JS interpreter. That hole would be openly exposed to the web anyway, so the PDF reader doesn't provide any further attack surface. Any random website could already exploit that JS exploits.

I get it

[Anonymous Coward]

It's so unstable that it lost its connection while typing the subject?

(Someone accidentally an R.)

Re:Honestly?

[UltraZelda64]

Well, if there is one. I don't exactly use it myself or keep up-to-date on the latest news.

Wow, more excuses.

[Futurepower(R)]

Why not just fix the bugs?

Here is a Slashdot comment from 7 years ago: There is a HUGE, well-known bug in Firefox 1.5, the CPU and Memory Hogging bug.

Re:Wow, more excuses.

[Anonymous Coward]

Why not just fix the bugs?

They have been.

Re:Honestly?

[ls671]

You can enable the "save tab prompt" when quitting. I saves all open tabs and you get re-logged automatically into the sites you were logged in when FF restarts. I close random tabs to leave only the tabs I need to work open when I restart because FF takes to much memory. about:config, preference browser.tabs.warnOnClose, browser.warnOnQuit, browser.warnOnRestart

http://support.mozilla.org/en-US/questions/796107

http://support.mozilla.org/en-US/questions/935532

http://support.mozilla.org/en-US/kb/how-do-i-restore-my-tabs-last-time#w_restore-the-previous-session-every-time-you-open-firefox

This road seems familiar.

[VortexCortex]

Element & Style Editing, PDF rendering, scripting, embeddable animation, sound, video, client side storage, 3rd party plugins...

Am I talking about Flash? Java? A web browser? MS Word? A bloated do-everything "Kitchen Sink" Business Solution? WHO CAN TELL?

Fuck that. I already know the best way to engineer things: Each thing does one thing and does it well, and provides an interface so it can be used in conjunction with other things to perform a task. Some call this "The Unix Way", but really it's the only sane way to do anything. See also: Object Oriented Programming (same shit), or Model View Controller (same shit), etc, etc...

A browser should be a text markup system that relies on a content provider layer to provide services for rendering embedded elements: images, videos, animations, games, whatever. It should be a glue program, and I should be able to select what content / mime-types map to what data decoding providers in the content providing layer. That's the way to make a minimal "do everything" engine. That way I can roll my own graphic format and renderer service (maybe for 3D point cloud data), and register it with any browser and create my documents with new elements, maybe submit an element to the W3C, and get it standardized so others can also implement it, then "image/cloud" mime type and my rotate&zoom capable 3D data display can be adopted. W3C needs a content provider / plugin standard for browsers, this way plugins or applications can work across all browsers.

Look, we can't have our cake and eat it too. Either browsers are a lean and mean text and image display system, for static documents, or it's a platform for making any kind of program you want -- If it's the latter then YOU DON'T PUT EVERYTHING YOU CAN THINK OF IN THE PLATFORM. You kick all that shit out of the platform, reduce it's role and complexity, and offload PDF, video, sound -- hell even scripting -- to plugins. Everyone's already seen the include everything approach. It's dumb. You end up with Java. We should be making browsers that can be integral parts of our OSs, standardized so we can swap them out without losing functionality. Make "browser" instances isolatable (optional non shared DLL/.SO) so I can fire up a client side only instance and stay isolated from the web and its exploits.

Gods damn, I know this is just a pipe dream, but this shit WILL happen eventually, why not start doing it now? It's so frustrating to watch all this effort wasted time and again on the do-everything approach, when we could do shit the right way, "the Unix way". WE HAVE THE TECHNOLOGY.

Re:This road seems familiar.

[Anonymous Coward]

maybe submit an element to the W3C, and get it standardized so others can also implement it, then "image/cloud" mime type and my rotate&zoom capable 3D data display can be adopted.

Since there's already WebGL in <canvas> I predict your submission will be rejected.

Re:This road seems familiar.

[MadMaverick9]

YOU DON'T PUT EVERYTHING YOU CAN THINK OF IN THE PLATFORM. WE HAVE THE TECHNOLOGY.

Gentlemen, we can rebuild it. We have the technology. We have the capability to make the world's first web browser. Mozilla Firefox will be that web browser. We can make it better than it was before. Better...stronger...faster.

We will, simply because we can ...

Re:This road seems familiar.

[freeze128]

I, sir, would like to buy you a beer!

Re: Firefox is not sandboxed!

[snadrus]

Exploits need impl. holes or local trust, while Firefox provides neither.
Firefox patches exploits fast. IE sandboxing mitigates damage post-exploit when they have a slow security response: browser data is still at risk.
Fast-patching is the better bet for me, but I'd like both.

Re:Honestly?

[viperidaenz]

Would you rather use Adobe software to read PDFs?

no good

[Anonymous Coward]

Pdf.js is simply not up to standards. It doesn't seem to be able to render documents from InDesign using relatively simple transparency features.

Re:Honestly?

[ButchDeLoria]

lol sysadmins using Metro

Re:Honestly?

[LordLimecat]

Hyper-V metro app, whats not to love, amirite

Re: Firefox is not sandboxed!

[DuckDodgers]

Actually, PDFs can have exploits that have nothing to do with Javascript. If the problem is Javascript, they can deliver it directly to Firefox and don't need to embed them in the PDF. But the real risks are buffer overflows and similar exploits that attack the PDF rendering program directly, which is typically Adobe PDF Reader.

The Firefox developers are aware that they're behind Chrome and IE on this, and they're definitely working on it. Starting at some point last year, Firefox automatically warns users when their browser plugins are out of date. PDF.js in Firefox, once it becomes the default for showing PDFs, will eliminate Adobe PDF Reader as a source of error

With IonMonkey, Firefox has closed the performance gap with Chrome well enough that I would no longer consider Chrome's speed as a deciding factor when choosing a browser. But now Firefox has three remaining Chrome features to tackle:
1. Sandboxing plugins, as you said.
2. Multi-process, so that errors in individual tabs don't crash the entire browser.
3. (Least important.) Updates in Firefox are automatic, but not transparent. Making them transparent would help.

I hope they manage all three.

Re:Honestly?

[DoktorSeven]

There are other PDF readers, you know. All much better than Adobe's bloated viewer.

Re:Honestly?

[viperidaenz]

Standalone PDF readers with plugins tend to try and support as much of the latest PDF features as possible. I'd rather my browser PDF viewer didn't have the capability to run embedded PDF scripts and load additional content.

Re:Honestly?

[hobarrera]

No, I'd rather use zathura. Windows users can use SumatraPDF.
Why do people keep assuming that Adobe is the only PDF reader there is, there's dozens out there.

I prefer firefox not to have a PDF reader, so when I click on a link to a PDF I'm prompted to download it, instead of having to wait for it to load and be rendered with JS before downloading it.

Re:Honestly?

[viperidaenz]

I don't, I use foxit. Most people use Adobe, because they're told to download Acrobat reader if they're unable to view the document.

I hope so.

[Futurepower(R)]

Version 17.0.1 crashed a few hours ago; it is considerably more crashy than earlier versions. There was a 3. version that was much more stable. Maybe version 18.0 will be better.

The crashes occur when I am doing a lot of research, and have many windows and tabs open, and then go in an out of hibernation or sleep mode.

Re:Honestly?

[hobarrera]

So if you're perfectly aware that the are alternatives to Adobe Reader, why do you suggest what, by discarting firefox's builting PDF, I'd be using Adobe software?

Re:Honestly?

[viperidaenz]

I stereotyped Anonymous Coward as being "average internet user who clicks and install what ever website xyz says"

One bug has NEVER been fixed after 18 versions

[hack slash]

When you click on a link of a file that isn't a usual image/link/txt type it asks you if you want to Open or Save the link to disk, but even though I've got the checkbox ticked for "Do this automatically for files like this from now on." it ALWAYS brings up that fucking popup! They've NEVER fixed it!

Re:Honestly?

[GNious]

interesting :)

When restarting due to update it usually starts up with the same tabs. Some pages looks slightly different due to web-apps not having states saved.

Dear Sir.

[TheRealHocusLocus]

Dear Sir,
I am firmly opposed to the spread of PDF files either to the home or to the office, We have more than enough of them foisted upon us in public places. They are a disgusting Americanism, and can only result in the farmers being forced to grow smaller potatoes, which in turn will cause massive un-employment in the already severely depressed agricultural industry.

Yours faithfully,
Capt. Quinton D'Arcy, J.P.
Sevenoaks
-- Letters To The Editor, The Times of London
___

p.s. embedded html5 video widgets suck, most flash apparatus work better. It's getting so that I have to postpone Firefox updates (including security ones) until some day in which I have enough time to spend disabling all the new features also included in the release.

p.p.s. I am a 21st century luddite. My car has window cranks, my cell phone has 10 separate buttons, one for each numerical digit.

Re:Honestly?

[RaceProUK]

Well, if there is one. I don't exactly use it myself or keep up-to-date on the latest news.

There is ;)

Re:Honestly?

[RaceProUK]

lol sysadmins using Metro

Unless you install it in Server Core mode - no Metro then.

Does it still leak?

[Anonymous Coward]

I have the latest v17 browser and, after a week of being open, it is consuming 3.2GB of memory. No addons. No plugins.

Firefox for iPad

[Anonymous Coward]

When do u have a plan to designed it for ipad??????????

Re:Honestly?

[async5]

No, here is why http://web.nvd.nist.gov/view/vuln/search-results?query=Adobe+PDF&search_type=all&cves=on

Re:Honestly?

[async5]

Here is news for you too http://web.nvd.nist.gov/view/vuln/search-results?query=Foxit&search_type=all&cves=on SumatraPDF will be in this list too once it gains user base

TLS 1.2

[petteyg359]

Some day, Mozilla will drop their craptastic and archaic NSS in favor of some modern SSL library that supports modern protocols.