Firefox 29 beta arrives with easy Sync tool.... | Hacker News Reader

Firefox 29 beta arrives with easy Sync tool....

freshtechapps.com

39 pointskumarrahul71 comments

Firefox 29 beta arrives with easy Sync tool, customization mode and Chrome-like user interface Australis

You can always use Aurora which is Firefox two versions ahead of release. So you would have had those features about a month ago.

I was for a long time hesitant thinking that it would be unstable, or cause compatibility issues with my favourite extensions... Nothing like that. At least in my experience. New sync worked completely flawlessly. I now use it on both Ubuntu[1] and Android[2] almost exclusively.

On Android it's faster than Chrome beta, perhaps partly due to AdBlock, and it has a nicer interface, especially closing tabs is much easier. It's not available from Google Play but it updates itself so you only need to download the .apk manually once (make sure you do it over ssl).

On desktop some changes brought by Australis are, let's say, not to my taste. Of course, there's already an addon to fix it[3]. You can keep whatever you like from the new UI, and revert whatever you don't. For example, I like the drawer but didn't like the inability to move back and forward buttons to the right of the url bar. They're back on the right[4]. Same goes for rounded corners of tabs, or any other control issue.

[1] https://launchpad.net/~ubuntu-mozilla-daily/+archive/firefox...

[2] https://www.mozilla.org/en-US/mobile/aurora/

[3] https://addons.mozilla.org/en-US/firefox/addon/classicthemer...

[4] http://i.imgur.com/LE8UZHY.png

In this new version, Mozilla presents a major user interface overhaul Australis that makes it easy to customize your browser.

The amount of doublespeak Mozilla has been putting out recently is astounding. Read the comments here for example (I don't know if they've changed anything since then, but I doubt it): https://blog.mozilla.org/ux/2013/11/australis-is-landing-in-...

All they appear to be doing is chasing Chrome's design decisions.

I worry that desktop Firefox will go the way of the Android version or the neutered Firefox OS browser, where you cannot install the vital security add-on NoScript due to the incompatible add-on model that the mobile versions use. That used to work on Android, but no longer (and it seems the noscript dev is not a fan of the mobile platform). There are many more add-ons that only work on the desktop version. They should all be cross-platform and not desktop specific :(

While I agree, Australis only breaks visual add-ons like Vertical Tabs, and the like. Not sure where NoScript comment is coming from?

For now Australis only breaks visual add-ons, but I can see a drive to align, i.e. remove features from, the desktop version so it is more like the mobile versions. The thinking being - Australis is visually more like the Android version of Firefox, why not make its add-on model more like the Android add-on model? i.e. Neutered.

The NoScript comment is because that is the "killer app" for Firefox for me, and on mobile Firefox it is not available. There's the amusingly titled "NSA" addition - http://noscript.net/nsa/ - but as you can see from that page it is pretty much dead. A forum post from November 2013 (1) indicates that the author is aware of the problem, but the diverging code bases I guess make it difficult to maintain.

1: http://forums.informaction.com/viewtopic.php?f=25&t=17687&si...

We have indeed seen a severe dumbing-down of all of the major desktop browsers over the past few years, as mobile devices have become more widely used. So I think that what you're saying is happening, and will continue to happen.

Desktop browser users should not have to suffer because mobile users choose to use platforms that are crippled in so many ways. Nor should desktop browser users have to install numerous extensions just to bring modern desktop browsers back to the level of usability we had a few years earlier. Yet with each new release of the major desktop browsers, we find ourselves in the position of reduced functionality, and the need for more and more extensions to restore the now-missing functionality.

Instead of dumbing-down desktop browsers to match mobile ones, I think there should be more of a push to getting mobile browsers to desktop usability levels.

It might take a few more levels of UI indirection since the screens are smaller, but I don't see any fundamental reason why a mobile browser couldn't have all the features of a desktop one.

They are doing a major UI overhaul - the design itself is unstable. So there are valid reasons for current visual add-ons not working - the majority of bugs reported by users would be caused by plugins designed for a different user interface, making it useless as feedback. That would defeat the purpose of having an open beta. Meanwhile the plugins should not yet update because the interface can change any update. So for now it's a chicken/egg problem but I'm sure that once the interface stabilises things will be as they used to.

So while it is good to stay on your toes, I don't think you realistically have to worry about this going to be a permanent issue.

NoScript is unavailable because current NoScript only works with the classic 'everything in one process' Gecko model & associated plugin model.

Firefox for Android isolates content into multiple processes (this is literally a requirement on Android due to their horrible memory management), which means plugins written in that fashion simply can't work. Unfortunately, at present this excludes NoScript. It's possible it could be fixed, though; I'm not sure.

The problem is that most existing extensions are based on XUL, which is terrible and impractical for good mobile UIs.

There is a more abstract addon API (used by restartless addons) and your favourite ones just need to be updated.

Also, NoScript is not a "vital security add-on".

> Also, NoScript is not a "vital security add-on".

The vast majority of exploits depend on JS being available, since (at the very least) it'll be used to unobfuscate/execute some attack code. Remember the Tor browser vulnerability that was exploited? That wouldn't have happened without JS.

(Coincidentally, I find it amusing that not too long after, that Firefox made it even harder to disable JS, by removing the checkbox for global enable/disable.)

That's almost like saying the vast majority of exploits depend on you using a Turing machine.

Breaking most websites for the sake of slightly reducing the attack surface doesn't seem worth it.

You have never been in the Russian-speaking part of the Internet, have you?

The definition of "most websites" varies between people; the majority of the Internet browsing I do (mostly reading content in the form of text or images) does not require JS. If I come across a site in a search that refuses to show content without JS (and Google's cache, text-only, doesn't give me what I want to see either), I'll just move on to the next search result.

It's far better for websites to break, than my machine. The tradeoff will be different for someone who regularly browses a different demographic of sites.

Firefox has no XSS mitigation built in unlike Chrome and even Internet Explorer. NoScript provides that for Firefox, even if you allow global scripts for practicality purposes you're getting XSS protection.

And I realise that there are design trade offs, but it seems sad that mobile has to drag the desktop into the realm of relative uselessness - not all add-ons can be updated to use the new abstract API (since the API will have gaps that the old API used to support).

Well, yes and no. They seem to think it's the winning design, or their user studies indicate it (I wouldn't doubt there is benefits to tabs on top).

Also their not Chrome design, is the fact they are also pursuing tablet users with huge icons.

Actually, Australis has been at work for years. I seem to remember that the first mockups of Australis are actually older than Chrome 1.0. It just took us lots (and lots, and lots) of iterations to build something that users actually enjoy.

I've been patiently waiting for Firefox to fix their ugly UI, I look forward to finally ditching Chrome and returning to a browser made by a company whose primary concern isn't accumulating data on its users.

If it's only the UI holding you back you could have found an extension/theme long ago.

If it's only the UI holding you back then you having been following the NSA spying / Google cooperation stories lately.

FXChrome solved the problem for me (chrome-like theme for firefox, its perfect). I am glad someone finally told them their last design iteration was horrendous. Honestly, who makes these decisions over there?

Lots of people tell Mozilla that their recent UI changes are bad. Just read the comments of any blog post Mozilla makes about such changes. Mozilla never listens. Their current approach is always deemed the "correct" one, even if it's seen as totally wrong a mere 3 to 4 months later, and subsequently reworked in some significant way.

These recent moves to imitate Chrome to an even greater extent are not welcomed, even if Mozilla forces them on its remaining Firefox users. Many people use Firefox these days precisely because its UI still retains a few of the vestiges of the more more usable UIs of its pre-4.0 releases. Getting rid of these makes Firefox even less appealing.

Perhaps the only thing keeping people on Firefox is that the UIs of the other browsers have gotten far worse (like in the case of the new versions of Opera and IE), or were never very good to being with (like in the case of Chrome).

I love that the same thread has people complaining that Firefox is simultaneously not enough like Chrome and too much like Chrome.

Mozilla literally cannot win here; no matter what they do people will just get angrier. They should just strip out the whole Firefox UI and make you run it from a terminal.

"Ugh, the new newsfeed is terrible. I'm quitting facebook. For real this time, not like the last 4 times."

I mostly like Australis. And I can see why Mozilla doesn't listen to most of the noise about it. As with everything else, nobody gets really loud about how much better it is.

People do "get loud" about stuff that's inherently an improvement. C++11 is a good example of this. It generally does improve C++ in significant ways, and there are a lot of people who are willing to express this.

The fact that we overwhelmingly hear silence or anger toward Mozilla's UI changes is indicative of them not being very good at all.

Don't get me wrong, I don't consider chrome's design to be the end-all of browser design. The difference is that chrome is pretty. Firefox (pre Australis, have yet to try that) was just down right hideous. Whether its usable or not is irrelevant if a user can't stand to look at it. The "chrome" on your application should not be off-putting, however functional you think it may be. It is a lot like furniture; functional is important, but none of that matters if its hideous to begin with.

Why is it surprising or notable that there are people who have different needs?

Sure, there are some people who make very minimal usage of a web browser. They can get by just fine with a dumbed down Chrome-style UI. They might even like that experience.

Not everybody falls into that camp, however. There are a lot of users who push web browsers much harder, and expect a different experience. They need a UI that exposes the functionality they need to constantly use.

Firefox used to be good at giving power users what they needed, while still being relatively lean, easy-to-use and useful for users who pushed it less intensively.

We've seen this change over the past few years, however, as they've stripped out useful UI functionality in their quest to imitate Chrome. Firefox is now a weaker, less-practical browser for many people. These people are burdened with trying to install numerous addons, just to restore Firefox to something reasonably close to its previous state of usability.

Firefox isn't alone, of course. Just look at the whole debacle that Opera has experienced with the new versions of their desktop browser. It's perhaps even worse in that case, given how they're building upon Blink and Chromium directly. The new versions of Opera are horrible compared to Opera 12 and earlier. It's just really unfortunate to see this happening to software products that were once pretty good.

UX people and graphic designers, based on extensive user testing. Have you run any user tests lately? ;)

I've participated in them, and one thing you learn very quickly is that users don't know what they want. But users can tell if something is ugly--you should listen to them there.

The majority of the users being tested are not power users. In fact the power users are probably the ones who don't like the idea of their browser phoning home to report their usage habits, and so disable those sorts of things.

Australis is such an improvement that I'm tempted to come back from Chrome. Everything feels very fluid and fast. I'd be curious to see the latest benchmarks on startup time, Javascript performance and such.

I don't know how long you have been a Chrome user, but Firefox has felt snappy and responsive for a few months now. Australis re-design shows that often speed is more a matter of impression than benchmark results.

I switched from Firefox to Chrome the week it came out. So maybe 5 years? I've kept Firefox installed for testing web development and such though.

You're absolutely right that perception of speed is largely about experience. The animations are smooth, it doesn't jump around. But I can tell that the boot times are much faster now as well.

Both teams have done an excellent job. Seeing this release has encouraged me to finally donate to Mozilla. They do a lot of good for the web.

IIRC, modern Australis is actually tuned for better rendering performance; older themes used more gradients and alpha transparency in UI elements, which made them more complicated to render. Australis makes lighter use of these elements and this has a measurable impact on how much CPU time gets spent repainting tabs/address bars/etc.

http://arewefastyet.com/

"In addition, Firefox Sync is also upgraded so that all current users will need to create new account"

Seriously, NO! How to alienate your existing customers.

I'm not sure that part of the article is correct. I've been using my old sync account in Nightly builds (now up to version 31) and it's still working fine.

Edit: I just checked with a clean install, and there's a button that says "Using an older version of sync?" It brings you to this page https://support.mozilla.org/en-US/kb/how-to-update-to-the-ne... which basically says "unlink all your devices and re-link them." So I guess adding new devices to your old account doesn't work.

So now I need to give them my email address plus remember yet another password for sync? Do not want. I always thought the old model with the sync keys was pretty sweet. Is the synced data end-to-end encrypted? It better be!

In times of free and secure password managers "remembering passwords" should really be a non-issue.

Let's just plug KeyPass here since its open source and brilliant: http://keepass.info/ Also for the linux and co (incl osx & win too actually) users: https://www.keepassx.org/news/

You can run your own Firefox sync server, but it's a pain to implement.

http://docs.services.mozilla.com/howtos/run-sync.html

Once it's up and running, though, it's seamless.

Edit: Hmm. Looks like they're happy to break this:

" Some unknown proportion of the user base uses their own Sync servers ... [Asa] I don't think we should worry too much about this group. If we can identify them and message them that'd be sufficient for my concerns."

https://services.etherpad.mozilla.org/sync-migration

Well, thanks.

Here's a code-dump of the new Firefox Account server. Not sure yet if I can implement this:

https://github.com/mozilla/fxa-auth-server

That's not a very reassuring attitude. Hopefully the new Firefox Accounts authentication server will be easy enough to deploy and configure as an alternative to private Sync servers.

> Is the synced data end-to-end encrypted?

Yes.

Unfortunately, according to our research, an enormous majority of people (read 95%+) hated it. So we had to switch to something less surprising, but that most users [believe that they] understand. Sad but necessary.

So.. can I disable australis somehow?

Have you at least tried it or are you simply change averse?

I have done the move a few weeks back, and I don't regret my decision. The toolbar and my add-ons located there still work very well, also the new menu gives a quick access to useful features.

I, for one, have tried it and it makes things I do frequently, more difficult: there's no longer available a bookmark button to open the bookmarks sidebar, there's no longer a menu item to show all bookmaks, when I bookmark something I click the bookmarks star twice in order to enter tags but the second click is only sometimes registered. For me, the previous theme worked much better than australis. I struggle to see the improvements in australis.

Opening the bookmarks toolbar takes indeed two clicks now (but you can use the keyboard shortcut CMD + B to open it instantly). Regarding your second point, the double click works great, and the bookmark details overlay appears.

I'm on Ubuntu, if that makes a difference about the double click on the bookmark star so perhaps it's a platform issue. But I was referring to the bookmarks sidebar not toolbar. How do you open the sidebar?

On Windows Ctrl+B opens the bookmark sidebar, and ctrl+shift+b opens the "library" popup with bookmarks selected.

My point is that there used to be a button that you could place in a toolbar and clicking the button would open the bookmarks sidebar. That button is now gone. I'm sure that in time I'll get used to the changes in australis but I think it would have been better if they would have been introduced gradually. As it was introduced, lots of things stopped working at once the way I was used to. For me, this is friction and it's not good.

I have tried Chrome and I didn't like it.

As nightly-build user I've worked with it a few weeks. I got used to everything - except the tab's on top (more mouse-movement and they just don't make sense, I don't get why they removed that option). But now working with classic theme restorer, so as long as they don't break that plugin everything's fine.

I got the Australis updates yesterday, tried it for half an hour. Never use the menu, want my tab real estate back, so I got the addon to fix it.

The shiny gloss over everything is still annoying, and they made the RSS icon monochrome rather than the normal RSS icon when you can sub to feeds, but otherwise I can just get rid of the menu I'd never use and have my old UX back.

If by tab real estate, you mean the amount of tabs that can be shown at once, you only lose 1/8th of a tab http://i.imgur.com/uT0oHl3.png

There's an extension called Classic Theme Restorer that aims to do it, as well as customize australis.

The fact that such an extension exists, and that it may get broken again in later versions, says a lot about what Mozilla really think of customisation.

> says a lot about what Mozilla really think of customisation.

Oh? And what is that? "Customization is great and you should have the option to do so"? Try having a "classic theme restorer" with Chromium.

> it may get broken again in later versions

Future changes may cause incompatibilities in any project. That is why such changes are released in major new versions so that add-on developers have plenty of time to make adjustments and fixes.

It's different when breaking changes are infrequent and optional. This is no longer the case with firefox: breaking changes can come at any time and you generally don't have the option to upgrade (although I do recall there being an option to turn off auto updating, I'm not sure this is a good solution though).

Mozilla periodically releases 'ESRs', or extended service releases, that are exactly what you complain they don't offer: the ability to lock yourself to a given release without being forced to upgrade (the most recent ESR gets security updates, IIRC.)

What used to be possible with the stock browser, without any extensions, now needs one or more. They're not quite at the level of Google, but it looks like they're heading in that general direction.

They didn't make the addon. What, you want them to go back on 15 years of history and make the UI non-customizable? What are they, Google?

I think he just wants them to not screw up the UI so badly in the first place that numerous addons and customization is needed to bring it back to a minimal level of usability.

I'm not sure what were the datapoints that suggested to them that the UI was in need of an overhaul but I would have been much happier if they lowered the cpu usage of Firefox. To be fair I have a massive number of bookmarks (12000+) and tabs :-/ I was happy with the UI as it were. I'm using Ubuntu, btw.

I've fallen in love with rekonq recently, it always has really low memory footprints, loads instantly, and uses webkit so 99.9% of sites that work in Chromium work in it too. (I say that because it hasn't updated to blink yet).

It is supposed to get Chrome extension support this year too, and if I ever had the time I'd try to expand their sync options (currently Opera + Chrome) to sync tabs and such.

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

In the linked markup, the difference between the bookmark star when the page is bookmarked (vs not bookmarked) is that the star is filled in grey rather than an outline.

I think this is a subtle design problem that doesn't easily show "bookmarked" vs "not bookmarked" state. I think users may have come to expect the typical yellow star when bookmarking a page. Usually if I want to bookmark a page I glance at the star to see if it's already bookmarked, and the outline vs. grey states don't make that clear.

There's still some stuff before I can consider switching from Chrome, most are OS X specific-ish:

- No swipe animation for going back/forward in history.

- Doesn't have that native feel for "over-scrolling". You know when you try to scroll beyond the page and see that grey linen stuff.

- No Keychain integration (yes I know there's an addon, but I would rather see this integrated).

- Even though there's been plenty of improvements, the dev tools in Chrome feels more polished and feature complete.

The real strength of Firefox vs Chrome is in FF's real (as in not android-"open-source") open source philosophy which makes forked projects like Palemoon possible (I personally prefer Palemoon at the moment due to more conservative release cycle and UI plus strong security focus)

I agree that the freedom to actually maintain viable forks like Palemoon is excellent.

On the other hand, I wish Palemoon didn't exist: It's a compatibility nightmare for anyone trying to build HTML5 apps or games using recent APIs. All his changes tend to cause mysterious memory usage/cpu usage issues or simply break things, and 'use a different browser' is not a great answer to have to give a customer.

I don't know how you could consider Palemoon security focused unless everyone using it is building it from source and they've carefully examined his changelists. At least with trunk Firefox you know the commits are exhaustively reviewed and tested (perhaps too exhaustively; some FF contributors find the extensive review process stifling).

EDIT: Removed complaint about Pale Moon builds not being reproducible; he used to have a big angry rant about how he wouldn't share his build settings with anyone, but he's apparently removed it. Nice move!

I've been using it for quite a while (as "Aurora"), and it is quite a significant improvement over regular stable Firefox builds.

Reply on news.ycombinator.com