I'd like to comment on goals for engineers in general:
If your goal is X million users or some such end-user metric, you have created a goal for your salesmen, not your engineers.
An engineer cannot sign up a user, except indirectly through quality work. "X million users" is only indirectly under the control of the engineer: intermediates like sales or getting-the-word out are out of the engineer's control and can be screwed up by others. If the engineer does not get to pick his own features to work on, then the work he does do is limited by the wisdom of the person making assignments.
In other word, a goal like this is discouraging for an engineer because he lacks serious control over whether or not the goal is accomplished. It is an anti-motivator.
Instead, goals for engineers should be things directly under their control. Are you assigning specific features to the engineer? Getting a list of them done with high-quality solutions can be a goal since that is entirely under the engineer's control. Want to get your low-cost solution adopted? Set a goal for the engineer to reduce memory space by Z percent or improve performance by Y percent.
Always make sure goal achievement is controlled as much as possible by the person receiving the goal.
And speaking as an engineer working on this project, I was very happy to see X million users / end users metrics.
I care whether the software I develop is used and enjoyed, its my primary motivation, the quality of the software and the technical health of the project have a huge bearing on whether people will use these it
The goals are discussed and created by community members and employees, engineers and otherwise.
It's already been addressed that individual teams at Mozilla have more specific goals with more actionable metrics, but in Mozilla's case I disagree that engineers cannot sign up users. Firefox didn't have a sales team to get its current marketshare.
We do it in person for friends and family, we do it through discussion and diagnosing problems on hackernews and twitter etc.
Source: I work in Mozilla engineering and I help people out online and offline with Firefox all the time, clear up misconceptions people have etc.
Firefox succeed because it was so obvious to people that IE sucked, period. IE was a shitty product, MSFT did not care.
Firefox with it's tabs,it's plugins was a natural choice for internet users.It was faster,more reliable and gave the best experience.No need for a sales team to promote that.
By the way i think that without Mozilla there would have been no Google success story,but that's a discussion for another time...;)
Certainly those sorts of goals shouldn't be how engineers are judged (or worry that they'll be judged), and they should have their own KPIs. However, it's very beneficial for them to be included in the overall goals, as it can increase both their sense of purposes (why are their individual KPIs important?) and their sense of being a team (we're all working together to grow the userbase, not just the sales guys).
Mozilla is just the best company. They took a stand when it mattered and where the billion dollar money makers Google, Apple, Facebook and the like failed spectacularly.
They are no bs, fight for internet freedom, celebrate transparency, are the centroid of internet activism and the extension of Aaron Schwartz' ideas. Mozilla is what makes me want to stay in tech and every company should try to learn from Mozilla what it means to be a tech company.
Apple was the first to bring implementation of things that now a big part of what we call HTML5: canvas, CSS transforms and animations, etc. Mozilla deserves credit of taking over from IE6 and moving web forward in the first years of XXI century (when there was no Facebook at all and Apple was hardly a billion dollar money maker), but around 2007 Gecko stalled and Apple and then Google driven WebKit took over. Maybe Mozilla was a tech company but now I see them mostly as politics company.
I must've imagined all the things like emscripten, all the new WebAPI's that make Firefox OS possible, WebRTC, asm.js, etc...but then again I'm not a random internet troll which is posting vague and baseless comments all over HN.
emscripten depends on LLVM, guess who is the biggest backer of that project?
Would you mind to expand a bit about "vague and baseless"?
Or maybe you could spend some time and learn the history of canvas element, or CSS transforms?
Note that at this point Firefox had DataChannel, and it would take Chrome months to catch up there. Firefox also introduced the used audio codec.
That said it's a bad example, because WebRTC standardization is currently a strong effort between both companies and it'd be great for the web if more standards went this way rather than ending up with having to try to standardize half-assed implementations from one rendering engine. (cough WebSQL cough)
I really miss focusing on the details of their desktop product here: About a week ago I tried to switch away from Chrome and back to FF and its been a rough week. Speed wise FF is actually great, absolutely no complaint here, same goes for memory. What I didn't expect and what kind of shocked me is how broken some web pages are on FF. For instance in GMail I can't use Cmd+arrow keys to jump to the beginning or end of a line and on the new Google maps (the one that has been around as a preview for more than a year) zooming by scrolling simply does not work. It's not just Google sites though, I had stutering scrolling on some more complex sites (where Chrome has no issue) and some startup pages have oddly shaped buttons and similar layout issues. It seems like many sites aren't testing on FF anymore.
I really expected FF to be on par with Chrome in terms of rendering quality but that really turned out not to be the case in many small, yet annoying corner cases. They got big by incorporating the quirks of IE and then improving on that, maybe it is time they start thinking about implementing some of the WebKit rendering quirks and go from there.
I really want to keep on using FF and to love it but if it means I have to switch browsers for some of the sites I use daily that will be difficult.
> For instance in GMail I can't use Cmd+arrow keys to jump to the beginning or end of a line
This is a bug in gmail, not a bug in Firefox. In standard browser text widgets, Firefox reads your standard OS X key bindings and treats them properly.
However, Google is unsatisfied with standard form fields, and so they reimplemented their own glitchy, poor-performance, half-broken text widget, which happens to semi-work in Webkit, but not handle shortcuts like cmd-arrows in Firefox. It’s in no way fair to blame Mozilla for that.
This doesn’t address your main point, but Mozilla’s side of the Gmail+cmd bug has been fixed in Firefox 29. Google now needs to remove their mitigation on their side: https://bugzilla.mozilla.org/show_bug.cgi?id=341886#c67
tbh I think some of this has less to do with FF and more to do with OSX. (used one for almost three years.)
As far as I'm aware of there is no universal standard that every developer follows on OS X. Some choose CMD + arrow keys, others the classical shell version (ctrl a/ctrl e). Even Apples own apps are (was?) not consistent in this matter.
Nearly all OS X apps use Cocoa and standard Cocoa text widgets. Ones which don’t, such as Firefox, usually try to read the standard Cocoa keybindings and respect them. Firefox actually does a relatively good job with this now (within the past 2 years or so; before that, they didn’t).
Both cmd+arrows and ctrl+a/e should work in almost every text box on OS X.
[Unfortunately, applications like browsers override cmd+arrows to switch tabs or navigate through history or something, depending on context... which is IMO a really stupid and obnoxious choice. Browsers typically use cmd+arrows for back/forward except in a text box, where the text bindings take over. I hate it, because it leads to accidentally going back when the user intends the text box movement command.]
The problem in the GP’s particular case lies with gmail’s custom form field though, which is not a regular text widget, not with Firefox.
I might be wrong but from what I remember from two years ago the constant annoyance of switching keybindings between each app was one of the things that prevented me from becoming a fan of OSX and I really tried. (I think the resident fanboy at that site had an explanation about Cocoa and Carbon(?) keybindings but to dumb me it just didn't cut it.)
The zooming issue may be an OSX problem, but it works fine here on linux. I've also never noticed any rendering problems, websites look like they should. Do you have any examples?
The stuttering during scrolling is probably because firefox is one process, so if lots of javascript is running on the page, it interrupts the process. In general, scrolling is smooth for me though.
This would get a lot better if FF had development tools built in that felt less unfamiliar to people accustomed to the Safari and Chrome dev suites. (Firebug is less bad in this regard, but being built in counts.)
In one sentence: they don't want to have the only income from having the search engines (especially Google) in the search boxes and links, so they want to have their own cloud and services and in-browser ads.
I thought more privacy & less leakage to third-parties would be nice. Especially with all the government access to databases. Ie adding more control of cookies, JS, requests, HTTPS etc while providing an interface that my grandma could understand.
Both the team and some of the underlying technology are now at work on Firefox Accounts, which is a simpler (non-federated) system that will power the new Firefox Sync service, the Firefox Marketplace (for installable web apps on Firefox OS and other platforms), and other Mozilla services.
The only change is that Mozilla isn't funding new feature development in Persona this year since Firefox OS has slightly more time-sensitive / urgent needs. Persona is still actively supported by Mozilla, and is a critical part of our own infrastructure.
Persona ain't going away so long as we're still relying on it.
I'm not sure what the long-term plan will end up being. I personally hope it can find a new direction to grow eventually. (Note: I'm a Mozilla developer but I don't work on Persona or Firefox Accounts.) We use Persona extensively for our own sites and tools, so I'm sure there would be some sort of migration process if Persona were ever shut down in favor of some replacement.
In hindsight we did indeed spend too much time experimenting and tweaking, and not enough time saying "it's good enough" and driving it to completion.
But it's open source :) we're still working on getting it there. The next release will dramatically scale back the API complexity, lock in the APIs and data formats, and pave the way for native browser extensions, self-hosted polyfills, etc.
I hope they step it up in the browser area again. I know they were top dog, but as we all know, Chrome is now the reigning champ in speed. Firefox has come a long way. I still try it out for a couple weeks every few months, but I continue to come back to Chrome.
I couldn't imagine switching away from Firefox + Vimperator. It's an incredible combo, and I don't think the speed difference between FF & Chrome is noticeable nowadays.
Chrome does no such skipping (on the same machine). RequestPolicy and ABP seem to help, but not eliminate the problem.
Okay, so that test is totally artificial, but surf the internet for a couple days without restarting Firefox (27.0.1) on Ubuntu (13.10) and Firefox still grinds to a half, taking more than half a second to change tabs or register other UI input.
but surf the internet for a couple days without restarting Firefox (27.0.1) on Ubuntu (13.10) and Firefox still grinds to a half, taking more than half a second to change tabs or register other UI input.
Shrug, I do this all the time as I never restart the browser. And your timer test is WORKSFORME too :-/
Speed wise, I really think Fx is pretty much on par with Chrome. There are areas where Fx is faster and Chrome is faster. In general, both are great.
If you install a lot of addons/extensions, I will expect some performance hit. But that's an not excuse, yet I personally not a big fan of addons (well, addons are common in web development) anyway so that doesn't really concern me.
Memory hike usually happens with plugins (flash, java applet) but normal browsing is fine. Plus I have 16GB to spare...
Wow. That's going to take like... every single piece of my firefox UI except maybe the location bar, and either reset it to the current defaults I can't stand or simply drop it all on the floor. Given that it's looking like I may have to rewrite a good chunk of the UI in an extension to fix this, I'm wondering if I'll even be able to find the time to upgrade.
Mozilla has been making some horrific decisions lately, and it seems they are going to be yet another business set to learn the hard way that surprising and pissing off users that you depend on for survival is a BAD idea, when they can leave for your competitor at any time.
Just about the worst thing you can do when users don't like your Cool New Design is to say "you'll get over it" and repeat how the features they cared about were removed.
I don't have much opinion about Australis. I was referring to performance regression, addon/extension compatibility. Mozilla has a lot of work to put into after enabling it in the release; I believe it's 29?
Sure, but projects like Rust have been very successful in terms of getting external contributors interested. A goal regarding getting more people involved and made aware of Mozilla Research efforts wouldn't seem out of place.
I hate to see a monetization/sustainability plan based on ads. AdBlock will kill Tiles, and if it doesn't, that'll kill adoption amongst occasional users.
A better idea IMO: get on the user-is-the-customer-not-the-product train and build the open web at the same time, by blessing a distributed social networking technology and selling installs of that technology, Automattic-style.
Sell a well built MozBook with 4GB+ and a full HD screen and some nice extras like a backlit keyboard for double the price of a typical Chromebook and I am in. Perhaps. I probably would not have to queue for it.
Mozilla are facing some real problems I am afraid and they aren't technical ones. They aren't the default browser on any significant platform so for people to switch they need to have something wrong with their existing browser. That is decreasingly the case with IE which has improved considerably and then they have to be the best alternative browser available which for many people is questionable.
Going the platform route is interesting but they are competing with Android with a massive installed space at the low end.
The elephant in the room is the Chromebook which I am writing this, and most everything else, on and the trouble with the damn thing is it is good enough. Not stunningly, mind blowingly amazing. But good enough and selling by the millions. And chrooting it and installing Firefox is not a typical scenario.
So, let's not try to build the perfect plan. Let's build a good enough plan, and move fast and change things as we go.
Please don't. Moving fast and breaking things is the #1 thing that's been turning me off Firefox ever since the move to six-weekly updates and dropping more than token support for older versions.
Notably absent from the list is the one thing I, and I suspect many others reading this, actually care about: making Firefox a better browser, i.e., fixing bugs, improving performance, and adding support for useful new features. Do that, and any concerns about how much/little it's being used will take care of themselves.
I really don't care about yet another UI rearrangement (I'd rather you didn't), I really don't care about integrating the kitchen sink into the browser (I'd rather you didn't), and I'm still waiting to discover anyone who does care about these things and doesn't work for Mozilla.
Sadly the maintenance of this wiki page has fallen behind, so goals for current and previous quarters aren't posted yet. :( Anyway, I can tell you they are largely focused on performance (startup speed, smoother scrolling, and responsiveness), developer tools (the new memory profiler; CSS source maps) and some experiments that may lead to new features in the future (like built-in translation of web pages between different languages). I'll ask around to see who has the official list and can update the wiki page.
There are also engineering goals for the teams that build the Gecko platform (shared by Firefox, Seamonkey, Thunderbird, and Firefox OS), which are at:
> Notably absent from the list is the one thing I, and I suspect many others reading this, actually care about: making Firefox a better browser, i.e., fixing bugs, improving performance, and adding support for useful new features. Do that, and any concerns about how much/little it's being used will take care of themselves.
Stop the FUD already will you? (Check my comment history, I'm mostly nice)
Mozilla has been iterating tirelessly on speed and lower memory consumption etc etc.
I can't remember when the last major ui redesign was.
So for everyone who hasn't tried FF in a while: give it a try!
I said it many times before and I will repeat this again there: this goal was dropped as soon as Blake Ross left the project. And I am starting to suspect that even if Mozilla wanted to make Firefox a better browser they no longer know how.
Wikipedia fails to say anything about monetization and googling is not readily enlightening. Does anyone have a link to a quick run down of how they monetize this?
The general idea is that the contract from Google and other competitors are not enough to grow Mozilla.=, especially hiring more staff. Mozilla Inc exists to pay taxes and accept business contract (search engine option ranking).
We have been spending around ~$115K per month for Amazon AWS infra for the past 4 months....
Even non-profits have to pay the bills somehow (I have done lots of volunteer work and read up on such things). Plus, there is a taxable Mozilla Corporation. So, somewhere, there is money being made, somehow.
Mozilla Corporation is not for profit, which basically means that they're there to get money and pay taxes (non-profits don't pay taxes). All they need to do is pay the bills (and employees).
Mozilla Foundation completely owns Mozilla Corporation and there is no shares available.
Firefox has too many bugs and when you file them they don‘t get fixed very quickly at all. That’s why it isn’t my primary browser. Maybe something about that should be in this document.
> Objective: Become a force in the cloud while delighting our users by adding services to our products.
> Measurable goal in 2014: End the year with 20M people using a Firefox account.
That's what I want to see from Mozilla. I want to have a johndoe@firefox.com account and use it as my login account to the firefox cloud, to all the services mozilla will provide.
They need engagement, I was an early user of phoenix more than ten years ago. I dumped firefox for safari and never touched a single mozilla property again. No need to.
Now they have to seduce me again, they need to bring me day after day to their properties. They need email, chat, cloud storage, documents, notes, blogs, everything I can do, everything I want to store, send, exchange, communicate, enjoy, play with.
Everything.
Then they can insert ads and compete with google. Billions over billions.
Then they can offer premium accounts with premium features. Billions over billions.
Its sad to see Mozilla go.
I really liked them as a software community, but I think I will soon have to be looking for a different browser.
Or maybe someone at Mozilla will make a stand, and tell out that Mozilla accounts/device sale figures is just wrong, and that it does not make the web more open, but ties them to another "ecosystem"
Your post makes no sense to me. Why do you think web applications built on standardized HTML are "another ecosystem"? Mozilla has been pushing this path since...forever.
If your goal is X million users or some such end-user metric, you have created a goal for your salesmen, not your engineers.
An engineer cannot sign up a user, except indirectly through quality work. "X million users" is only indirectly under the control of the engineer: intermediates like sales or getting-the-word out are out of the engineer's control and can be screwed up by others. If the engineer does not get to pick his own features to work on, then the work he does do is limited by the wisdom of the person making assignments.
In other word, a goal like this is discouraging for an engineer because he lacks serious control over whether or not the goal is accomplished. It is an anti-motivator.
Instead, goals for engineers should be things directly under their control. Are you assigning specific features to the engineer? Getting a list of them done with high-quality solutions can be a goal since that is entirely under the engineer's control. Want to get your low-cost solution adopted? Set a goal for the engineer to reduce memory space by Z percent or improve performance by Y percent.
Always make sure goal achievement is controlled as much as possible by the person receiving the goal.