Firefox's developer tools are really becoming great. "Copy as CURL" for the network tab was something in chrome that I really missed. Box model view which was super handy is now editable. Console showing stack traces along errors was long due. New Canvas debugger, eyedropper are very handy.
Font viewer, responsive design tools, the 3d view (very handy in debugging silly box model issues) and highlight painted area mode (handy in optimizing visualizations) are all things I love that I haven't been able to find good analogues for in other browsers.
Yeah it's pretty good, but it still has issues. My game uses a manifest.cache and the console/networking tab does not give good feedback with it. In chrome, the console will tell me, "This file could not be found". In Firefox, it gives me no feedback that was a problem... things just don't work as expected.
Also, I'm pretty convinced the dev tools have some sort of memory leak in them. Every 3-5 days I have to restart Firefox or my game runs so slow it's unplayable. I don't think it has to do with my code because I've got no persistent store and refreshing the page doesn't fix it. Restarting the browser always does.
Sometimes the debugger gets confused and puts breakpoints on things I no longer have breakpoints on.
The profiler does not have a very useful view. I can't figure out how to delete reports I've made without closing the dev tools. More importantly, I can't find a convenient way to answer the question, "What are the slowest parts of this code?" I have to do a lot of manual labor expanding deeply nested trees.
As I said, it's pretty good. These are just things I'd like to see improved, not things that drive me crazy. I prefer the dev tools to firebug because firebug takes a relatively long time to start up.
Re: this update, I'm embarrassed to say I didn't know console.error even existed (I don't consider myself a JS expert). I've been using `throw new Error("message")`. Will that show stack traces now, too? I always have to put breakpoints on the line with the throw to figure out the call stack. This can add a lot of time to my debugging if I'm not sure how to reproduce the error.
That is not true. I am sure add-ons can access the same stuff that built-in tools can. Most of the stuff are exposed as JS objects and there are APIs for a lot of stuff. APIs for devtools may be a bit hard to find the documentation sometimes. I guess it is because it is still being developed.
For me Firebug seems ~2x as slow (eg, for a 1s operation in native tools, firebug takes 2s). Opening Firebug on my workstation causes a noticable pause in all activity on page for several seconds, while opening the native tools only takes about half a second. All of Firefox's dev tools are still a ways from being as fast and bug-free as the Webkit tools (both Chrome and Safari). I can't cite specific examples since I dropped Firebug in Firefox once the native tools reached a certain level of quality. I just suggest trying to use 'em and seeing if you like them more than Firebug. I did.
I have a little bit of a hard time believing this: Firefox Addons can use almost all the APIs in the browser that built-in code can, and I'm at least led to believe the new debugging APIs the built-in tools are using are no exception to that rule. https://developer.mozilla.org/en-US/docs/Tools/Debugger-API seems to agree on those points. Or do you mean the tools themselves aren't extensible?
It's been a few years since I've written any Firefox code or touched XPCOM, so I could be looking on this without up-to-date information: I'd be interested if the Firebug authors or others have written any thoughts on struggles they have with this.
I'm also a bit worried about the privacy implications of this. Especially given that the feature is hidden, and the only two ways to turn it off involve going into about:config, which most people don't even know exists. I wouldn't even know it was doing this if I hadn't read the changelog.
Reading the changelog, it also mentions having local and remote blacklists, but how FF chooses which one to use wasn't clear to me. Local blacklists are not as scary to me for obvious reasons. Being able to use this without the remote blacklist would be nice.
I would really love some additional info on this feature from Mozilla, as well as a more user-friendly way of disabling it.
> and the only two ways to turn it off involve going into about:config, which most people don't even know exists.
There's a third way according to the feature's development documentation [0]: uncheck "Options"/"Preferences" -> "Security" -> "Block reported attack sites".
It seems like the last few updates of Firefox are each hiding yet one more important setting... I think that is what will finally drive me away from Firefox (I've used it since the beginning even if chrome was much faster for a while) ... time to browser shop.
There's no "seems" about it. Useful functionality has been obscured or removed from Firefox time and time and time and time again since Firefox 4 and their rush to imitate Chrome. It isn't a new problem at this point; it has been happening for years now, and people have been angered by it for years now, too.
It's a shame that your comment has been voted down, as well. The more that the Firefox community goes out of its way to deny that users are unhappy, or even to censor them like in this case, the harder it will be for Mozilla to remain relevant.
Firefox is pretty much Mozilla's only semi-successful offering at this point, and even its market share has been steadily dropping (it's probably well under 20% by now). If this slide continues, nobody will have any reason to listen to Mozilla. Their influence over the web, already waning, will unfortunately become non-existent.
They might be "hiding" the setting by not including it in the main settings page, but there is no other (mainstream) browser with anything close to the equivalent of the flexibility of about:config.
Though judging by the logic espoused in the tab-close-button bug step one is to move it to about:config and then step two is removing it all together for not being discoverable enough.
yes, but they also change about:config "Preference Names" possibly to confuse users so that they stop messing with settings? for example the preference name for the number of lines scrolled when using the mousewheel used to be "mousewheel.withnokey.numlines", I believe it is now "mousewheel.acceleration.factor" ... I didn't keep track of all the changes, but I am sure that there is many others.
A switch on/off in the preferences is needed (because privacy choices must be easy for everyone) and I'd choose "off", but I think this bit from Google may ease the mind of someone here.
> It’s important to note that any time Safe Browsing sends data back to Google, such as information about a suspected phishing page or malicious file, the information is only used to flag malicious activity and is never used anywhere else at Google.
As a Firefox user, I really think it's poor form on the part of Mozilla to not provide any additional information or re-assurances on their website about the data captured and recorded by Google. Mozilla's slogan on their homepage is "Commited to your privacy and an open web". Being committed to privacy means being open and explicit about the data captured through their browser services, no matter how innocuous the data might seem. It also means presenting that information clearly and making it easy to find. People can then make informed choices about whether they want to use such services.
This is the sort of thing I would have expected them to have said:
"When you download a file from a web page, Firefox checks that the file does not contain a virus or malware before you save it. To check the file is safe to download, Firefox contacts Google to use a service they provide called Safe Browsing. Google checks if the file is harmful or safe, and sends this information back to Firefox (this normally happens in a few seconds). If the file is safe, Firefox will start the download. If the file is harmful, Firefox will block the download and display a warning message.
When Firefox uses the Safe Browsing feature, it needs to send Google information about your download. Google records the following information from Firefox: your IP address, the name of the file you are downloading, the address of the website, and [insert any other data recorded here]. [Also insert a re-assurance that Google does not keep a record of all your downloads against your Google account or against your IP address - assuming this is the case. Also explain how Google uses that info. and how long it's kept for etc.]
[Then finally explain how to switch off this feature if you don't want to use Google's Safe Browsing feature.]"
They used to use a bloom filter with a regular download list, which would allow local checks without revealing every URL, but it looks like that changed at least for Chrome: https://code.google.com/p/chromium/issues/detail?id=71832
Google's malware scanning service also generates false positives and provides no contact info for submitting an appeal. A little too automated for my liking.
I don't think that there is a good place to do that, unfortunately.
You could file a bug, but it'll be pretty much lost among the thousands upon thousands of other ones that already exist and haven't been dealt with.
You could comment here or in some other discussion forum, but those in the Firefox community will probably just vote you down, and continue to pretend that there isn't a problem.
You could write an article, but again, that probably won't help much, unfortunately.
The best thing to do may be to find an alternative browser, or use older browsers that aren't broken in this manner.
I just don't think that those working on Firefox these days truly care to listen to what the remaining Firefox users have to say. Time and time again lots of users have loudly expressed extreme displeasure with monumental mistakes like Australis and the removal of useful functionality, only to be totally ignored by the Firefox developers. I don't see why this situation should be any different.
It's also checked against a remote one unless you either disable the feature entirely or set "browser.safebrowsing.appRepURL" in about:config to an empty string.
They removed browser.tabs.closeButtons. It's now impossible to hide the close buttons without css fiddling. I don't get it. What benefit is there to removing that feature?
Same with clearing the download list. They removed that feature with the last major UI overhaul.
The number of tiny addons I have to install just to get features back that have been removed for no reason at all is starting to get too large...
I already have to use some user CSS because they removed the ability to set the tab min width via `browser.tabs.tabMinWidth`, because that property wasn't "worth it" according to one developer.
Edit: Turns out Classic Theme Restorer has an option for that.
That is a nightmare to set now if you want tabs to go below about 40 pixels, thanks to the Australis swoops. It used to be one line of CSS to set min width to 0-ish, now I have a giant blob of custom CSS on top of Classic Theme Restorer.
The only thing I want from Firefox is to stop freezing completely when a single tab goes crazy. It just makes the whole experience feel clunky compared to Chrome.
Usually it happens with Gmail so maybe there is a conspiracy there to drive adoption for Chrome :p
Multi-process Firefox (aka Electrolysis or just e10s) is till under development. If you want to test it, I recommend using Firefox Nightly (34) because it has recent e10s bug fixes and its File menu has a "Open e10s Window". You can test e10s and non-e10s windows without flipping an about:config pref or restarting Firefox. You can flip the optional pref browser.tabs.remote.autostart to true to enable e10s by default, which also enables extra "shims" for add-on compatibility so more add-ons will work than with "New e10s Window".
Chrome uses a separate render process for each tab[1], so when one crashes it doesn't KO the whole browser. However this also means Chrome has a larger memory footprint, since each tab is taking up the resources of a little browser. Firefox doesn't do this, and so it keeps a smaller memory footprint than Chrome (oh how times change) but the downside is--as you mentioned--the whole browser freezes if one tab fails.
Try changing the values of dom.max_script_run_time and dom.max_chrome_script_run_time in about:config to lower values so that unresponsive scripts can be killed faster.
tbh this happens to me more often in chrome than firefox these days..
i do get slowdowns in firefox when some tabs are loading tho. but the complete freeze in chrome is when the ui or something goes wrong, its rarely a single tab (which doesnt really happen in firefox)
In Firefox 29 I continued to see weird font rendering issues on Windows 8.1. So much so, that I can't use FF on Windows right now. I don't see the same issues with Mavericks. Has anyone else experienced this and does it continue into this version?
I have the same issue in Windows 7, but have not found what exactly causes it. I work around the issue by highlighting the incorrectly rendered text with the mouse.
"Firefox is already running" just means it failed to shut down. Any number of things (including malware and bad plugins) can cause it. I think you're referring to a specific bug that caused more regular users to see the warning, but that doesn't mean the warning is gone.
Just that Firefox itself shouldn't cause it any more.
"What happens when you download malware? Firefox checks URLs associated with the download against a local Safe Browsing blocklist. If the binary is signed, Firefox checks the verified signature against a local allowlist of known good publishers. If no match is found, Firefox 32 and later queries the Safe Browsing service with download metadata (NB: this happens only on Windows, because signature verification APIs to suppress remote lookups are only available on Windows). In case malware is detected, the Download Manager will block access to the downloaded file and remove it from disk, displaying an error in the Downloads Panel below.
How can I turn this feature off? This feature respects the existing Safe Browsing preference for malware detection, so if you’ve already turned that off, there’s nothing further to do. Below is a screenshot of the new, beautiful in-content preferences (Preferences > Security) with all Safe Browsing integration turned off. I strongly recommend against turning off malware detection, but if you decide to do so, keep in mind that phishing detection also relies on Safe Browsing."
The MathML torture test shoes some very nice progress[1] but makes me ask myself once again, why do TeX fonts always look like crap? Isn't that the one thing that TeX should get right? Anyway, at least they look great as rendered by Firefox 31.
I'm not sure "crap" is the word I'd use for that, but I think I understand what you're getting at. The problem is that the TeX fonts are designed to work at a much higher resolution (as is typical for being printed). They look much better in that case.
Yeah, i don't earn any points for eloquence, but just looking at the side by side examples on the MathML page the contrast is pretty stark. Have often wondered if it was due to low screen resolution so thanks for confirming that. I'll try to be less grumpy the next time I get a TeX generated PDF and maybe even print it out before trying to read it ;o)
I don't see the problem. At that link, the browser column is higher contrast. Switching to the Stix fonts gives characters that look more like the TeX column, but seems to be missing a few things. I did not try all the font options.
Really? Maybe it renders differently for you, because here the difference is stark. People keep using TeX so maybe most people don't see a difference. For me, almost every TeX document seems to use a font that looks like something very old and unprofessional. But perhaps as suggested elsewhere it is optimized for higher resolution output devices.
I had the opposite experience, with the same machine (Mid-2011 MBA). Firefox, even with plugins, was superior to Chrome for some time now. Anecdote and data and all that jazz.
4.8ghz quad core, 8gb ram, gtx 580. Firefox is consistently slower on my machine sadly (I've also got 3 script blockers/whitelists in chrome with tons of extensions and still outperforms.)
I recently (this past week) got a brand new MBPr, 16G RAM, etc. FF provides a better experience Chrome, still. So, once again, anecdote and data and all that jazz.
> Maybe you and others in the Firefox community don't realize this...
We do.
> This is about users having bad experiences when using Firefox.
Every company creates bad experiences for some users. Hell, I've had bad experiences with Apple technical support, the apparent gold-standard of customer support.
> Your positive experience in no way negates these bad experiences.
And vice-versa. So what? What's the point? If I can't share my positive experience, they can't share their negative ones?
Maybe you think comments like "My experience was bad" is helpful and worthwhile. Maybe. But only if the inverse is true.
So, don't complain when people share anecdotes you don't like. Either accept them, or don't.
This seams to be a problem with your machine. I also use a macbook air (End 2013) and dont have this problem (or feel any Lag on Tab-Actions). Even on my old Ubuntu-rig it doesnt have this problem. Either in Firefox or Chrome.
Your response is the kind that has given the Firefox community a rather bad reputation when it comes to addressing problems reported by users.
Instead of accepting that Firefox may indeed have performance problem, you immediately discount this possibility, instead blaming it on mr_november's computer.
It's irrelevant that it isn't happening on your laptop that's over 2 years newer than his is. It's irrelevant that it isn't happening on your Ubuntu system. None of that matters.
I don't doubt for a second that he is in fact running into performance problems with Firefox. He isn't alone. Many people report Firefox having worse performance than Chrome does on the same system. I've experienced this, too.
Yet instead of addressing and fixing these very real performance problems that have been brought up time and time again by many users, the Firefox community and developers seem content to deny that they exist, or refuse to consider that it may be a problem with Firefox (like you've done), or point to useless and totally unrealistic benchmarks to suggest it isn't a problem.
But worst of all is how mr_november's comment has been voted down. It's one thing to deny that the problem exists, but it's much worse to try to actively censor those who have merely pointed out a very legitimate and troubling issue.
Firefox has been losing market share for some time now, and this trend will only continue as long as Firefox's performance problems go unaddressed, and the Firefox community mistreats anyone who dares mention that such problems still exist.
But worst of all is how mr_november's comment has been voted down. It's one thing to deny that the problem exists, but it's much worse to try to actively censor those who have merely pointed out a very legitimate and troubling issue.
'Actively censor ... a troubling issue'? You make it sound like a police state.
Firefox has been losing market share for some time now, and this trend will only continue as long as Firefox's performance problems go unaddressed
Firefox's problem is shedding it's past reputation. It really isn't that slow anymore - it depends on what you're doing in the chrome vs firefox wars. I work with a bunch of chromeheads, and they all spurn ff because of that past reputation.
In actuality, they have as many problems with chrome as they do with firefox - I'm constantly saying "[bug] not evident on firefox" and they express puzzlement that The Awesomeness That Is Chrome actually has a problem other browsers don't. Not to mention that a lot of the bugs I do run into on FF are because of their 'designed-on-chrome-for-chrome' default mindset. The bugs do get sorted out, but the perception that FF is so much worse than chrome isn't reflective of the truth of the matter. They're pretty similar these days, swings and roundabouts.
Yes, voting down a legitimate comment here so that it's greyed out, thus making it more difficult to read, is a form of censorship.
As for Firefox's reputation, I think it still has a reputation for poor performance because, contrary to what you and others may claim, a lot of people still find recent releases to be slower than Chrome and other browsers.
It will never be able to shed its reputation for poor performance as long as it still suffers from those problems. And these problems will persist as long as the Firefox community continues to deny that they exist, or go out of their way to suppress discussion of these very real performance problems.
I work on the desktop performance team and my job is to address performance issues. It is mistaken to suggest that we don't care about performance and responsiveness.
Can you please address why we haven't seen much progress, even after so many years?
You and others involved with Firefox may choose to deny it, but a lot of people still find recent versions of Firefox to be slow and/or to suffer from stability problems.
For example, just look at the discussion at Slashdot today about the Firefox 31 release:
Whenever Firefox is discussed, those kinds of comments seem to be quite common. While some may choose to brush them off, to me it indicates that Firefox has some real problems, and these problems just aren't being fixed. We wouldn't see people continually bringing up these problems if they truly had been fixed.
I have never denied anything and I do not appreciate you making such accusations toward me. Having said that, I have a few remarks about this post:
1) People making vague complaints on discussion forums or social media is not going to get the right information to the right people. We can't fix problems by spending all our time scavenging (HN|reddit|slashdot|whatever) for complaints. Even if we could, they wouldn't contain enough information to act on them. It's super important for the community to help us out: If you are having performance problems, you absolutely need to be filing them at https://bugzilla.mozilla.org/enter_bug.cgi?format=guided
2) Just like any other type of bug, we cannot do anything about people's problems unless we know what is wrong. Ideally we would have information about the user's hardware and OS, which extensions they have installed (and better yet, is it reproducible with no extensions at all), steps to reproduce and diagnostic information to help us.
about:memory and the Gecko profiler (https://developer.mozilla.org/en-US/docs/Mozilla/Performance...) are essential tools that we provide for diagnosing this stuff. If you're savvy enough to be reading HN, you're savvy enough to use those, and you should attach output from those tools to the memory/perf bugs that you file.
3) For people who are not tech savvy, we try our best to analyze telemetry information on their behalf, but for privacy reasons we only do so on the release channel if users opt in. The more users that opt in to telemetry, the better the data that we receive.
4) Finally, IMHO many people need to come to grips with the fact that modern web browsers are much more complex and capable than their pioneers were. I see lots of complaining from people who want their browser to have the same memory footprint as Netscape 2.0 and that just isn't reasonable.
Have you tried going to the special "about:support" URL and clicking the "Reset Firefox" button at the top right? This creates a new profile and imports all your existing bookmarks, history, saved passwords, etc. but resets preferences back to defaults and recreates all the databases from scratch. If there's any settings from old version of Firefox that are confusing the new version, or poorly-laid-out database files that can't be efficiently updated, this should clean them up.
I run it on a macbook white plastic (old). Firefox runs well. It does get tired sometimes (around 3pm). I usually go to about:memory and clean up.
That or restart it.
Very excited to see hash-source being implemented and enabled. This means you can load files from a CDN and have their integrity checked by the client so the CDN cannot trick the user.
You probably don't have flash installed, and Firefox doesn't have Media Source Extension support for h264 yet, which YouTube has required for 480 and 1080 video since last October (IIRC). According to this bug, support for MSE h264 will be landing in Firefox 33.
https://bugzilla.mozilla.org/show_bug.cgi?id=1027875
The 1080p option isn't shown because YouTube requires Media Source Extentions enabled (HTML5 DRM) for this. You -can- enable this under about:config (media.mediasource.enabled) but it's still experimental for now. When I tried it, it did show the 1080p option, but then I had very mixed results from the blank video, no audio, or the higher resolution video not actually loading.
This should also only apply if you have the HTML5 player enabled on YouTube, the Flash player should still show the 1080p option.
...and FF finally caught up with IE and Chrome habit of reporting every single URL you visit/download to check for malware.
Since there is no free-lunch, i'm pretty sure the people running those services are mining that data for advertising or something. I know google is with chrome. Malware protection and translator services phoning home on every page, and pre-fetch adding the links you haven't clicked yet to that list...
Font viewer, responsive design tools, the 3d view (very handy in debugging silly box model issues) and highlight painted area mode (handy in optimizing visualizations) are all things I love that I haven't been able to find good analogues for in other browsers.