The actual download itself is available on the "Systems & Languages" page:
> https://www.mozilla.org/en-US/firefox/all/
This kind of download is important, when you want to try out a new release, without committing yourself to it. For example, let's say you want to load it up in a VM, without contaminating your normal environment.
Googling for things like "firefox standalone redistributable offline install" are a road to nowhere.
You sort of just have to "know" that "Download Firefox in your language" means "Download the standalone installer, and not the stub installer". They don't explain this, or make it obvious that this is how you can get a copy of a static binary for predictable results.
It's a complication, but it eliminates the need for the user to come back after waiting for some dozens of megabytes to download and then interact. You can download a stub near-instantaneously, do things like pick your install directory, and then forget about the install process entirely as it completes without your supervision. That's a good thing, and it isn't possible with offline installers.
Browsers like Firefox and Chrome already need network access after installation anyway. They update on a regular basis, and pull down optional packages like Chrome's software WebGL rasterizer, malware blacklists, and Firefox's GPU blacklist. So, in practice, it is already impossible to do a true offline install of either browser; you just may not be aware of the things that are left out of the installer.
> Large downloads dramatically decrease the number of
> users who get from a download page to actually running
> an application
> they would be foolish to make things worse for the
> majority of users
> just to make them slightly easier for theoretical users
> ...do things like pick your install directory, and then
> forget about the install process entirely as it
> completes without your supervision. That's a good thing,
> and it isn't possible with offline installers.
Firefox's installer is dead simple, and believe me, I am HUGELY grateful for that. The only thing easier than the Windows installer is unzipping the Linux tarball into whatever path I choose. I hate Adobe's Flash and PDF installers, and I fucking DESPISE the Java installer, with its damnable bundling of that bloody Ask.com toolbar.
Seeing Firefox add these additional layers of background updater services and stub installers deeply worries me, and fills me with concern that as a "salary paying organization" (albeit, ostensibly non-profit), they might veer down an ugly path, go the way of the Sith, and start engaging in questionable behavior that is inappropriate for an open source project.
Consider the example of Ubuntu's desktop file search bundling Amazon ads in the results. What if one day, Mozilla decides that in order to pay the bills, it needs to negotiate a deal, whereby all those users with automatic updating enabled should get railroaded with some kind of optional-but-defaulted-to-enabled third party feature suggesting helpful reminders to buy more burritos from Jimbo's Refried Beans Emporium. Just sayin'...
> Browsers like Firefox and Chrome already need network
> access after installation anyway.
A browser only "needs" to to exactly what I ask it to do, and not much else. I tell the browser what to do, not the other way around.
> They update on a regular basis, and pull down optional
> packages like Chrome's software WebGL rasterizer,
> malware blacklists, and Firefox's GPU blacklist.
Malware blacklists are another concept that I tend to reject as mostly ineffective in achieving their stated goal. We could go round and around with that argument for days. Let's not get started on THAT can of worms.
> So, in practice, it is already impossible to do a true
> offline install of either browser; you just may not be
> aware of the things that are left out of the installer.
Some people have lousy Internet connections. Those people use things like download managers to help. Download managers don't work with stub installers.
A simple link to a true offline installer would help those people. (Torrent is fine too, I guess.)
You've just excluded >95% of Firefox users. Those that do use a VM will know how to get hold of the full installer.
If they'd stop making full installs, I would be annoyed. But they're just optimising for the most common use cases.
My complaint is simply that it's hard to find, and for those who want it, there's no real road map or sign postings for where to get it.
The intention is to provide a better download experience: there was a gap between the number of people who begin downloading Firefox (counted via hits on download.m.o), and the ones who open it for the first time (counted via hits on the firstrun page). The stub installer is intended to address this. It also checksums the full file to make sure you're getting what you think you're getting.
(Disclaimers: I work for Mozilla, and am responsible for download.mozilla.org among other behind the scenes web things. I don't work on the installers, but we had to change the download scripts to support the stub. I do not speak for Mozilla.)
(I wonder if they push the stub so as to collect more (system) information without having to say that it's Firefox itself that's collecting that information.)
My suspicion is that it's essentially a "security through obscurity" tactic, so that the majority of users are left with no choice but the "smart" installer, which offers greater control for load balancing heavy duty traffic during peak download periods.
That sounds rather unlikely.
Also, here's a genuine usecase for a truly offline installer... What if you are the "resident software/pc maintainer" of your extended family.... In such a case, would it not make better sense to have one complete download, then copy it into a pen drive, and install it on every machine, without spending 30 mins to an hour on every machine, doing a download+install for every single one!
The rationale of an online installer might be justified as some of the replies indeed seem to do, but, I think that there is something fundamentally flawed/wrong with having to first "download an installer", only to have it "download the application" again! I mean, what the hell!?
EDIT: Also, the speed of these release cycles (not just Mozilla, but other software too!) are getting old and tired to keep up with, TBH...
Especially for consumer PCs, there doesn't need to be anything to keep up with.
That said, they do make the full installers easily available (google “Firefox offline installer”) should you need to update a ton of computers behind a slow connection.
You're talking about a very small number of people who will update the way you described.
Most Firefox users will update by having Firefox automatically update in the background and then be prompted to restart.
I just downloaded the latest version (in my language) by going to 'About Firefox' and hitting the update button.
It makes sense though. The modern browser is epitome of "the user is always online, and should always receive and run the latest version of code on the fly" execution model. Why not apply those assumptions to the execution environment too?
That site does not readily provide hashes for their binaries (MD5, SHA512, whatever), and the page is smeared with ads. There's no way to know if the site is trustworthy.
And, anyway, where would I get the trustworthy hashes? They'd need to be communicated to me, directly from the horse's mouth. I'd need to visit ftp.firefox.com, to find out what the correct hashes are in the first place, and then check them against the binary download I receive from your site.
The link you should have provided was:
http://sourceforge.net/projects/portableapps/files/Mozilla%2...
Version 25 isn't listed there yet.
All of this is even less convenient than digging for and deciphering the meaning of Firefox's official downloads.
The concept of PortableApps has been simply amazing! (FWIW, I donate to this project too, it's been really worthwhile having the platform).
I have begun to adopt this mechanism for some of my software maintenance strategy. Nowadays, I test the latest version of the updated "PortableApp" on my local copy on my machine first, and if all works well post-update, I ask others to choose the update. Else I simply tell them to uncheck the update until I can confirm that existing functionality remains unbroken. :)
Users of VMs are more than likely power users at the very least, and if you want to test different versions then you should probably also be testing beta and even nightly. I don't think a full exe just to install in your VM really needs to be "homepage" visible.
Edit: It's probably just that I'm using Linux.
However, I haven't used it in years, except for testing. Philosophically, I agree more with Mozilla than Google, but I just can't seem to get past three things:
1) Chrome gets out of my way. The UI is unobtrusive. Firefox is still bulkier here. 2) For whatever reason, more websites render oddly in Firefox/Gecko than in Chrome/Webkit. 3) Chrome, being a part of the Google ecosystem, plays very nicely with my Android phone.
This is not to say I'm not open to switching back to Firefox, but I'm going to need a compelling reason to uproot my workflow and abandon my apps and extensions.
Firefox 25 doesn't appear to give me that compelling reason.
1) I really want to like Chrome, but I find the UI pretty bad. Yes, it's fine when you have 5/10 tabs open, but try to open more and you'll see (Firefox without addons is just slightly better than Chrome here, but at least I can customize Firefox). That said, there's a planned restyling of the Firefox UI (which you might already know), called Australis, which might fit more your liking (but I'm not sure when it will land).
2) I'm going to admit that I've never had this problem with Firefox. Firefox is a pretty reliable browser (for what is worth, check the Tom's HW Browser GP), and should render pages just fine. The only time I see problems is when developers use webkit- specific features, without including the moz- too (some times it's something not supported by Firefox, but many times it's just lazy developers that test their sites only with Chrome/Safari)
3) I'm in love with the Firefox mobile UI, and I find mobile Chrome just terrible. I'm not sure what do you mean by "plays very nicely", you mean tab sync?
To be clear, I'm not trying to say that Firefox is better/worse/equal/similar to Chrome, and I'm not trying to convince you to switch back. I just wanted to reply to your points.
Firefox's rendering issues have been largely, I imagine, the case of developers designing only for Chrome, or only for IE. It's not Firefox's fault, in that case, but it's still a reason for me to be wary of Firefox.
As for tabs in Chrome... I never have more than eight open at a time. I'm religious about closing tabs I haven't looked at in over five minutes. My own personal quirk, that!
Just yes. God yes.
It's absolutely must-try if you are the can't live without TabMixPlus type. Mission critical if you do any kind of ticket management/research at work. I frequently convert coworkers with the combo.
Bonus: https://addons.mozilla.org/EN-us/firefox/addon/context-searc...
I wonder if this is because Firefox has been optimized, or whether Chrome has simply stagnated. I suspect it's the latter; when I got the Retina MBP I immediately noticed that graphics performance was laggy compared to what I was used to on my previous (slower but non-Retina) MBP.
Is there any way to set a preference for free formats by the way? I.e. if the page specified several sources for the audio or video tags, but puts MP4 first let's say, is there way to override that and say "always prefer WebM if present" or something like that?
This is on my to-do list. Should be in Nightly builds in a few months' time.
This is coming soon.
But youtube encodes popular videos into the native firefox formats.
http://www.opengl.org/registry/specs/ARB/vertex_array_object...
http://www.extremetech.com/gaming/154994-epic-mozilla-releas...
You can't reliably detect the audio formats Web Audio can decode, and they vary both across hardware and software configurations (because the spec doesn't specify them), so your only choice is to download an entire audio file, try to decode it, and if you fail, download another and try that. Worse still, the <audio> mechanism of handing the browser a list of sources (so it can try to aggressively reduce wasted bandwidth) doesn't work because Web Audio requires you to download a full file before it can be decoded. Oops!
Shipped demos out there rely on Chrome-only Web Audio features that never made it into the spec, because Web Audio was written first as a non-portable implementation and then haphazardly turned into a spec while people were already building code against that implementation.
The spec changed after Google shipped it to the public (without putting it behind a flag or otherwise versioning it) so there are corner cases (albeit mostly small ones) where two spec-compliant implementations (Firefox and Chrome's for example) will produce different output. The only way to address some of these is update the application.
On the bright side, devs from both Mozilla and Google (among others) are working hard on addressing the issues, so it'll probably be fine in a year or so.
I try to do my part by submitting pull requests that fix it, when possible, but most of those pull requests have been sitting ignored for months.
Clearly more people need to be educated that user-agent sniffing = bad.
Feature detection simply doesn't work it many cases, and there's no alternative. Especially with new stuff like audio, sometimes function calls need to be called in different orders, have subtly different behaviors, etc., and user-agent sniffing is literally the only thing you can do.
When I programmed an HTML5 music player, calling .load() before .play() on one platform was necessary, and on another it would crash the Android browser IIRC. No choice but to use user-agent sniffing -- I probably had 20 different things that depended not just on the browser, but on the browser version.
I'd love if we could use just feature detection. Unfortunately, we have to program in the real world.
1) Signing into Chrome is easier and more intuitive then syncing with a code. Also having your extensions is nice.
2) The Inspect element and console are easier to use in Chrome. (Copy as HTML, etc)
3) I seem to have to click on a YouTube video twice for it to pause in Firefox.
4) Typing search queries in chrome is easier. For example in Firefox, you can't just type define: word to search Google for the definition of 'word', as it complains it is not a known command. I don't often use URL commands and think you should access about:config from somewhere else.
5) In Firefox when you scroll with the mouse you scroll too far per mouse tick.
ad 4: you can create a bookmark, edit its "location" and "keyword" accordingly. "%s" is the placeholder for the text you type. E.g.
Location: http://www.google.com/search?hl=en&safe=off&q=%s Keyword: g
Usage: Type "g whatever" in the URL bar to launch google search for "whatever".
You just need to extract the minimal URL and place "%s" accordingly. That way you can create multiple ad-hoc search engines, like "so foo" -> google search "site:stackoverflow.com foo" etc.
re 4: I do have the keyword 'g' set up but it is still easier to not type g[space] everytime. And yes there is ctrl +k but it is still easier in chrome with just one location.
I am willing to help you get started, if you are interested.
They couldn't let a few rogue cowboy-developers with an agenda and complete user disconnect overnight just ruin what Mozilla have spent decades building up.
Merely stating that Java is a security problem hasn't really sped up a solution to the problem. It's like the talk about how terrible PHP and JavaScript are, but it isn't really helping, is it?
So while my personal experience would not have been ruined by the initial proposal from Mozilla (as I do not use any plugins in my Firefox, I leave that for my Chromium browser), I do agree that they made the right call in the end. But it did at least spark a debate on the matter, and an interesting one at that. And it'll probably come up again, when another Java security issue appears.
CHANGED Resetting Firefox no longer clears your browsing session
Previously, when the "Reset Firefox" feature was used, all of your open tabs would get lost in the shuffle. This has now been fixed, and you won't lose your open tabs and current browsing session when Firefox restarts with a fresh and clean profile.
IMO It's a rather poorly described change.
If you have a pet bug that badly needs fixing, though, I can help you get started, if you are interested.
Images that aren't visible aren't decoded as they are downloaded, or something like that. It reduces peak memory usage on image-heavy pages. It should be coming in Firefox 26.
(Also that there are definitely still people working on improvements solely to desktop, see the upcoming redesign)
You do realize that you are commenting in a thread about a short-release-cycle release that adds a good number of improvements while you're also able to check on the improvements coming up in the next releases?
I wish Chrome would start adding them, then I could start using them for personal/particularly nerdy projects.
Firefox 26 will introduce Array.of, Math.fround, Generators and Structs.
It was about time!
Aren't minor updates not a big deal when a vendor is doing rolling releases?
The idea is that you can sandbox e.g. social buttons by putting them in a sandboxed iframe. You can also populate the iframe's without extra HTTP-requests by the iframe, by using "srcdoc". This is a perfect way to sandbox user generated comments on your blog.
I recommend reading [1] as it covers all of this.
[1]: http://www.html5rocks.com/en/tutorials/security/sandboxed-if...
The version number is now basically just a step counter.
You mean Aurora, and I'm not sure why you use "(sic)" here. It means "that's how it was written upstream, I didn't invent this!"
How exactly they manage to pull this off I don't know, but mad props seem apropos... :]
NEW Web Audio support (https://hacks.mozilla.org/2013/10/songs-of-diridum-pushing-the-web-audio-api-to-its-limits/)
NEW The find bar is no longer shared between tabs
CHANGED If away from Firefox for months, you now will be offered the option to reset it to its default state while preserving your essential information
CHANGED Resetting Firefox no longer clears your browsing session
DEVELOPER CSS3 background-attachment:local support to control background scrolling
DEVELOPER Many new ES6 functions implemented
HTML5 iframe document content can now be specified inline
FIXED Blank or missing page thumbnails when opening a new tab
FIXED Security fixes (https://www.mozilla.org/security/known-vulnerabilities/firefox.html)
new ES6 math functions
Map#forEach and Set#forEach
Array.of
Number.parseInt and Number.parseFloatAt least with the addons I have installed moving the close button of the addons bar is not possible.
I prefer the old position of the search bar buttons. The new one forces me to move the mouse to the other edge of the screen and also look at the opposite sides of the screen.
Unlike other elements it isn't possible to change the positions of the search bar buttons.
Not sure why they changed it. Could it be a bug?
https://bugzilla.mozilla.org/show_bug.cgi?id=686747
Edit: I would fix this myself, but don't even know where to look...
Wait, is that a new feature? Well let's file a bug.