Mozilla competes with 3 massive multinational companies (Google, Apple and Microsoft) who have between 30 and 200 times more employees than we do, allied with development and marketing budgets which dwarf our entire income. There is a clear and obvious revenue stream (advertising via search) tied directly to success in our market. They have awesomely powerful distribution channels that we just don’t have. How can we possibly win? What are our competitive advantages?
I suggest we have 3:
Community
Hard numbers are difficult to come by, but the often-quoted figure is that 40% of the Mozilla codebase is contributed by people we don’t employ. However, that doesn’t count the massive efforts people put in doing QA, documentation, support, local activism, and encouraging their friends to download, install and use Firefox. IE doesn’t have fans so keen they do this. Or this. When the Mozilla Foundation was 6 people in a borrowed office somewhere in Silicon Valley wondering what they’d do when AOL’s $2 million ran out, the community was what kept us going. And they’ve been vital ever since.
One thing which prevents the community growing is that we keep hiring people out of it. Which is fine – but if we want to keep employing people who believe in what we do, rather than people who are just here for a few years as the next step on their career ladder and because Mozilla looks good on the CV, it’s important to keep that source of passionate talent healthy. Which means giving responsibility to people we don’t pay, levelling the playing fields inside the project, and enabling meaningful participation.
We need to make sure we don’t take our community for granted.
Ideals
If you promote your product as “the new shiny”, then people will leave it pursuit of the newer and shinier. I think that a while back, someone realised this, and that led to the Mozilla marketing being refocussed to emphasize more that we are a non-profit, and are in this for the good of the web, not the good of the shareholders.
I’m sure a large proportion of Chrome users are ex-Firefox users. In the time before Chrome, when we were the new shiny, we missed an opportunity to educate them about why Mozilla is different, why the open web is important, and why having the coolest, fastest, slickest browser around is a great thing but it’s not the most important thing. So when something they perceived as cooler, faster and slicker came long, they left us for precisely the same reason they arrived. We didn’t tell them why they should stay.
We need to make sure we don’t keep missing that opportunity.
Goodwill
Goodwill is Community x Ideals. People who believe in our mission and what we are doing have gone to great lengths to persuade their family and friends to use Firefox. As users regained control of the web with popup blockers and the like, they realised that here was an organization which cared about more than the contents of their wallet. And they installed it for all their friends. But now, all those friends are ringing them up complaining that Firefox updates too often, the UI keeps changing and their addons no longer work. Heck, some of their own addons no longer work. They don’t understand why all this is happening, but they do know it’s a pain in the ass.
People who believe in us have fought inside their companies against recalcitrant IT departments, indifferent managers and our lack of formal support for enterprises to get Firefox to be an accepted or even the default option on their corporate desktops. These people understood that we didn’t have the resources to make their lives easier, but up until recently, what they also understood is that we empathized. We couldn’t give them technical help like MSIs or Group Policy support, but we could say “thanks very much”, and “we appreciate your efforts”. And that made it all feel so much better. Now they think we don’t give a monkey’s about them, and are wondering why they bothered.
We need to not give our friends the finger.
Very well put, and correct on all points from where I sit.
On the “open” front, how do you compete with chrome? Are you more open?
@Jon
For starters, Mozilla doesn’t include closed-source binary components in their browser (e.g. Chrome’s internal PDF reader) and doesn’t back problematic codecs because it’s convenient (Chrome still ships a H.264 decoder despite Google’s announcement to remove it).
Excellent post with some excellent points presented in an excellent manner. Excellent. Sums up a lot of my recent concerns far more succinctly than I can manage.
“I’m sure a large proportion of Chrome users are ex-Firefox users. ”
Like me, perhaps they found Firefox became a bloated monster. With stupid ‘features’ like modifying view source before displaying it and as a result ‘fixing’ tags etc, this is not developer friendly. Mozilla have simply wandered off into Internet Explorer territory.
Your comment on ‘the new shiny’ doesn’t line up with your latest crazy idea of iterating through major version numbers for minor updates reenforcing the ‘new shiny’ Firefox. At the rate you’re going we’ll be having “Now that’s what I call Firefox 86” by the time I complete this sentence.
Your take on goodwill is spot-on, particularly in the context of the “enterprise kerfuffle” a few weeks ago. You remember, that’s when a Mozilla product director stuck his thumb in our eye, flipped us off, and frankly revealed that he hasn’t a small clue what “IT support” actually does for its customers. Correcting that horrible communication error through a comparatively small investment would be one of Mozilla’s best opportunities to lead — by enablement. Organizations produce much of the web. Make it easier for them to do that “the right way,” end to end, and the Open Web will be well served indeed.
That’s Mozilla’s job.
Having some experience with this from the FLOSS/Linux world, I would urge you not to rely on idealistic reasons only. It makes a lot of sense for many people (me included), but if you push it too much you get to the weird philosophical corner of RMS.
I think for every idealistic company (including my employer) it is very healthy to keep very close look on the purely pragmatic strengths you have. Trying to emulate your biggest competitor and liquidating your own strenghs you are not doing favor to yourself. Yes, I am tolding about Chromelust and retreating from Firefox as platform.
I completely agree with this. At the end of the day, would it really, really be that much effort to devote some development time to some of these issues? It might not be Mozilla’s primary focus, but it seems like something that can just be sorted out once and would just stop being an obstacle to corporate adoption.
At the end of the day, pushing open standards needs to happen on all fronts, not just the ones that are convenient or trendy. Corporate IT isn’t going to go away, and it _is_ going to stay on Internet Explorer unless concessions are made by Firefox or Chrome developers.
I think as soon as you have to educate people into using Mozilla, then you’ve lost the battle. If you can’t keep (the majority of) people using the browser when they’re already on it, then there’s something wrong. I used to love Firefox, but it went in a different direction than I wanted, and now I barely use it. Chrome is my first choice, and IE (even though it’s slow) comes in second because of website compatibility. Personally I think Chrome’s gain is because it caters for developers. I’ve worked on projects where Chrome wasn’t officially supported, but it always worked because as developers we used it so much that we took the time to fix up any bugs that crept in. That used to be Firefox’s role…
As for goodwill, does that stretch to writing a version of Firefox that runs on iOS? ;-)
I did laugh when I read Scaredycat’s “your latest crazy idea of iterating through major version numbers” given your recent post about renumbering :-)
Gerv, this is really well-done. Thank you!
Gerv, I completely agree with your arguments, but I think one other large advantage we have, even if that’s more on the product level, it comes out of our open and web-centric view: customization possibilities. A large amount of built-in customization opportunities paired with unprecedented power in what add-ons can do is another extremely large advantage of Mozilla and Firefox.
And, don’t forget Firefox crop circle:
http://maps.google.com/?ie=UTF8&om=1&t=h&ll=45.123747,-123.113729&spn=0.001243,0.002173&z=19
Spot on. (I think I may have posted a comment to this effect recently.)
@Ken: Mozilla is an institution, with Firefox its mascot. Google is a company, with Chrome its product (with an emphasis on the possessive ‘its’). Ain’t nothing wrong with that — Chrome is a very good product. But you can never really put any faith in Google having your interests at heart. Would Chrome be as good as it is if Google had held a (near) monopoly the way MS did? Is Google going to put your privacy ahead of its revenue interests? Is Google going to give fair time to everyone who wants to have a say, including competing businesses?
Yes?
Are you sure?
Keep in mind, it’s not as if badly-behaving BIG companies don’t exist (see current hysteria over Murdoch press for one example).
@voracity: you’re right. I can’t trust that Google will have my interests at heart, but I don’t believe I can with Mozilla (not because they’re evil, but because our interests aren’t necessarily the same). Granted there’s more chance that Mozilla will behave in a way that I’m happy with long term, but in the shorter term, I’m much happier with the Google lighter weight approach. Sufficiently happier that I’m willing to take the risk that Google might be a little evil some day.
As an aside, I freely admit that I see Google through rose-tinted spectacles, which was part of the reason that I used Chrome as soon as it was released, even though it didn’t work for lots of sites. Now I do genuinely believe it’s better *for me*. I still think Firefox is marginally better for most non-developers.
Personally I’d be happy with three big players constantly fighting it out in the marketplace and all having around a third of the share. Looks like that might actually happen some day soon, albeit briefly…
Gerv, thank you for this blog post. Many contributors in Europe are still waiting for a more community-friendly signal from the MoCo…
Agreed, but unfortunately the customization/addon possibilities aren’t as good now as they used to be when Firefox (and subsequently, XULRunner) was supported for ~12 months.
” And, don’t forget Firefox crop circle:
http://maps.google.com/?ie=UTF8&om=1&t=h&ll=45.123747,-123.113729&spn=0.001243,0.002173&z=19 ”
This is a good example Atul! And remember the New York Times campaign? Or Firefox Flicks?
I think that Mozilla has always been at its core a grass-roots movement. I know that nowadays, technology change too fast and there should be always someone to drive the community innovation and someone who should stay on the top layers and make sure that all is developed in an open way (and thus, go faster w/ decision making process).
But what I can’t understand is the leak of communication and conversations. The complete indiference for “them”, the community (is that fear or simply indiference?).
Decisions are taken at All hands and Team offsites, where the ousiders (community volunteers) don’t have access. With a few exceptions, I never seen summaries from those meetings, or presentations, thoughts (though at Mozilla’s core is openess, transparecy, participation).
“We are taking away the dino head, we are going for rapid release cycles that indirectly affects add-ons ecosystem” etc.
Those decisions could be understandable for me…and I suppose that for others as well. But what I missed in all those processes was the conversation in open: why we should do this, which are the implications and risks and how we could prevent it etc.
I’m not saying that hiring people from outside community is a bad thing (that’s needed, if you want new ideas, visions) but those people should be aware of where they come to work – in a community-driven org., a different kind of “company”.
Not taking the community for granted, as you say.
I don’t have nothing against “rapid cycle release” (actually I think that this is critical now) but I’m not feeling comfortable about how the situation was handled (too much noise, indiference, misscomunication, frustrations etc.). Indeed, not a good way of work for an open community/company.
Thanks Gerv for your great blogpost!
Gerv, this is the finest Mozilla post that I’ve read in quite a long time.
Seriously, it’s awesome and it’s spot on,
Thanks. It’s very refreshing.
For me personally, I learned about Mozilla through Spread Firefox. I was one of those excited about the new shiny thing and it was there that members enlightened and educated me on Mozilla, what it’s all about, and why it’s important. I say it as it’s a single entity. Collectively, I guess that it is. I also learned about the open web and open source there. Oh, and from you too. Remember my massive page full of Mozilla trademark violations when you and Frank Hecker handled all of that? :)
No such place or system (like SFx) currently exists to get the word across that there’s more to Mozilla than just Firefox. No place to really energize and educate those just discovering Firefox who are really excited about it and are actually wanting to do more and be involved. No, clear, defined, and ya, polished place or system.
I know there’s been some good progress with the get involved efforts (aside from the Join Mozilla one), but they’re really just links to Mozilla sites when people need solid guidance and direction.
A few of us tried to get something going in anticipation of SFx being shut down because we thought that a community driven Mozilla site was important, but we needed more help and support but we couldn’t obtain it so it crumbled before it got going.
Retaining people is of course essential, but so is bringing in and guiding new ones.
A mentorship or buddy project, whatever would be nice.
I tolerate (but get pissed at times) some of the issues that I have, and come across with Firefox, Thunderbird, and some Mozilla sites and services because I have an understanding of the ecosystem, the processes, and the available resources. The majority of people don’t. They jump ship and move to the next shiny new thing. People don’t realize that their complaints and issues are being addressed and worked on.
They probably don’t because they’re just used to the way that most software makers, organizations, and companies operate. Primarily, behind closed doors.
Now reading Alina’s comments about just that is interesting, but at some point, someone has to pull the trigger because it’s their responsibility and job to do so, and at that point, I suppose we just live with it, or move on.
I just hope that their decisions were based upon, and out of a consensus from the community.
Anyway, thanks again for this post.
I sincerely hope that something good and productive comes out of it.
Here are things I don’t agree: Mozilla must be on the bleeding edge to gain new users. There is no other way. Users never cared whether something is built by profit or non-profit organization. It might help them to switch to newer product later, but that’s it, nothing more.
I must say that I am not fully confident in 6-week release cycle, but I am fully confident that 2 years release cycle has become insane recently, and this is surely good thing to happen.
The things I agree: Mozilla’s employes gained too much power somewhere at the working on Firefox 3 (but not whole cycle), and it made community less interested in.
Gerv, thanks for this and the previous posts in the last couple days. They all have been spot-on and are very much appreciated. I wish more of “decision makers” would join the discussion and make their arguments heard that resulted in the situation we have right now.
Alina, I completely agree with you. I have felt the same for some time now. Like in Asa’s comment on http://blog.gerv.net/2011/07/firefox_version_numbers_cognitive_disson.html#comments: “We posted and had public phonecalls discussing this for weeks.” Phone calls are fine, but make it very hard for outsiders to take part (time zones, work, etc.). So is there a summary of those calls somewhere? Some written documentation how the version number conclusion was reached, what the ups and downs of the individual options were? Not that I necessarily disagree with the conclusion, but the process how it was reached could be improved I think.
Once again, Gerv keep up the great work!
Well, I agree wholeheartedly that Mozilla is more open than most corporations, because your decision-making processes are almost all public. This has historically created 2 small problems within and outside the community:
1. Managing expectations: Decisions are made well in advance of any code being written. When an exciting new thing is announced, people get frustrated when they have to wait for the thing to happen. Google, Apple, and Microsoft can just “surprise” their users with the new shiny and can SEEM more innovative, without ever having to be.
2. Navigating the torrent: The rapid-release process was communicated through about a billion different channels. We knew this was coming, although the swiftness of it was shocking, even for a long-time Mozilla fanboy like me. So much info was communicated about this, it was hard to determine where and to whom meaningful feedback should be sent.
@voracity, I think it’s a mistake to paint the “Google is a corporation that doesn’t care about you, Mozilla does” dichotomy too strongly. Remember that Chrome was started by a team of ex-Firefox contributors (so we share a lot of the same ideals as the current Mozillians) and has a goal of making “the best possible browser for users”; and that regardless of intent, neither group has the ability to personally listen to every user’s feedback and validate every user’s desires.
I think Gerv’s original post got closer to the mark. While Firefox and Chrome both have user and contributor communities and are both open-source, Firefox obviously has the larger, longer-lived, and in some ways more passionate such community. And while both teams are filled with idealistic people, Chrome’s stated ideals focus more on the product, while Mozilla’s stated ideals focus more on the web as a whole. (Practically speaking, I think the two teams are actually closer in this regard, because Mozilla wants to advance the open web by making a great product, and Google wants to use their great product to advance the open web.)
Right now, as Gerv noted, there’s a minor crisis of goodwill among some Firefox users (and a few developers), for various reasons. I think (and hope) this will eventually pass. Firefox is why I’m a browser developer today, it’s a valuable choice in the market, and it would be a real loss to the web if it faded away.
While Chrome devs, or at least many of them share Firefox’s ideals – they’re the developers, not the company.
Google has a clear focus and clear goal for Chrome – control the market, make sure data isnt encrypted or stored away far from their reach, control the advertising market.
Mozilla does not suffer from that.
@Erunno You don’t ship the binary components, but then 90% of your users just install it manually, right? Would you agree that you’re doing a lot of your users a disservice by not helping them do it correctly? (Keeping it up to date, verifying compatibility, etc.).
@zeebo Most binary components still work w/o recompiling now. Only binary components using XPCOM must be recompiled each time.
And with the creation of a Javascript proxy, they could get rid about that dependance.
Of course, that some work to be done, but that’s the recommanded way to do it since 2007.
Also note that a lot of extensions will need to be rewritten for electrolysis. All those accessing the content from the chrome will need to use the Message API or broke. This API is already there now, so they should be porting their extension to it.
Mozilla has a communication problem with add-ons writers. It must tell them far ahead how to make their extensions evolve w/ the platform. For e10s & Message API communication should happen *now*, not in December when it will be almost ready (and too late for add-ons authors).
Note that XULRunner-based applications have a deeper problem, and a LTS may be of use for them. But this would be maybe a community job, helped by Mozilla. (Like Seamonkey).
Gerv’s remarks are spot-on re. involvement & listening to the community, but I also wonder if there is not simply a basic lack of software project management competence among the current Mozilla Co. leaders:
I’m just a Firefox end-user (and Netscape/Mozilla suite since 1995, before Firefox), and the major Firefox attractions for me are a) up-to-date support of WWW markup currently in use, and b) extensions that increase the utility of the browser or that integrate it better with the operating system & other programs.
Of the two, (b) is actually more important, since other browsers (e.g. Chrome, Opera) do (a) very well. (b) is really a defining feature of Mozilla suite (er, I mean Seamonkey) & Firefox, a major reason to continue to use it.
But the Firefox management team’s inability to provide a stable Gecko & add-ons API that includes explicit, formal guarantees about the lifespan of usable interfaces is breaking the experience for me and other end-users.
You’d think that extensible software that has had a user-base in the millions for many years, and a whole ecosystem of 3rd party developers/enthusiasts working with it, would have worked out an architecture by now that allowed stable APIs to be retained and deprecated in some kind of organized “no surprises” fashion (explicitly versioned interfaces, or whatever mechanism is necessary) while impeding as little as possible the evolution of the core nuts-and-bolts of the browser itself.
Other large, long-lived, multi-developer software projects manage this process (e.g. FreeBSD, Java, etc.). Why can’t Firefox? After all, the project is far more centralized and driven faster by fewer people than in the days of Mozilla suite–what’s stopping them from making Firefox an even easier & stable platform for developers and users?
The latest casualty of the rush to bright shinyness for me is the Lenovo ThinkPad Password Manager extension, which uses binary components to communicate with Lenovo’s authentication software. It’s a small but significant utility extension: instead of typing passwords and other confidential text, swipe your finger and have the text field(s) completed for you (and you can easily set it up so that only your fingerprint authentication will release the text). And the confidential text thus collected & stored can be used in any other application that integrates with the Password Manager.
I’m keeping Firefox at 4.0.1 until Lenovo releases an update for version 5. But what if 6 breaks the extension again? And 7? Lenovo will be entirely justified in dropping Firefox integration–the effort is unreasonable.
I accept and expect that major versions of Firefox are going to change internal architecture in significant ways that will certainly affect binary components. We all want a significantly evolving browser. But not at the expense of an open, extensible browser, and the excellent extensibility and utility that are Firefox’s real market arguments.
I missed teoli’s comment above. I’m not a developer, so probably I’m missing something, but even if recompilation is necessary for binary components, why is there no agreed set of stable interfaces that binary components can invoke?
E.g. the Lenovo Password Manager extension fails to initialize on 5.0 with an error message box “TypeError: Components.classes[TvtPwmComponent_CID] is undefined”. Surely one of the purposes of XPCOM (among other approaches) is provision of known “frozen” interfaces that can be invoked with some degree of version independence? But, as I say, I don’t know much about this stuff, just that other projects seem to manage it better…
Matěj: I am not suggesting we rely on idealistic support only. There are several great browsers, but idealism is one of our competitive advantages – the things we have that the others don’t.
Thanks, everyone, for your thoughts and comments.
@Zeebo
I wasn’t commenting whether I agree with Google’s strategy or not. I was only answering to Jon’s question where Mozilla is more open than Google. I’m sure there are more examples though (both technically as well as organizational).
The binary blobs are not a problem in itself but they are closed source or come with questionable licenses attached. Mozilla already presented that there are alternatives. For instance, the recent pdf.js project aims to solve the same problem as Chrome’s binary PDF viewer in a much more open (and “webby”) way. Google could have also developed or licensed a PDF reader which does not exclude non-Chrome Chromium users.
@Insert Real Name
About stable extension APIs: I strongly recommend you to read this articles why extensions break:
http://www.oxymoronical.com/blog/2011/06/Why-do-Firefox-updates-break-add-ons
The gist of it is that extensions have full access to all the browsers internals. This allows for some pretty powerful and creative extensions but comes at the price that the whole browser becomes the “API”. The alternative is Chrome’s approach by exposing limited extensibility via API’s which improves compatibility immensely but limits the developer drastically at the same time compared to “old-school” extensions. For this Mozilla has introduced the Addon SDK (aka JetPack) which works similar to Chrome’s extension mechanism as an alternative way to write extensions (and unfortunately probably the only way in the long term).
I’m afraid that this is one of these “can’t have your cake and eat it” cases, i.e. maximal compatibility while retaining the current level of extensibility.
@Insert Real Name
Lenovo Password extension also sounds like it could be an ideal candidate for js-ctypes, which would likely in the end be simpler to maintain, and would fix most of those compatibility problems.
This transition phase while things move over is likely just going to be a bit painful, but someone who has the notebook and wants the feature just needs to write it.
DigDug2k: Forgive me, but that’s a bit of a dismissive response.
I just pretended to be an extension author with a binary extension, and spent 10 minutes reading about js-ctypes on MDN. I now have a fair idea how they work. but no idea how to convert an extension to use them. And also, there’s a big warning which says this doesn’t work with C++ libraries. Mozilla’s libraries are C++ aren’t they? Also, what about backwards-compatibility? I still have users on 3.6 and 4.0. Does it work there too?
If js-ctypes is our answer for binary extensions, we need much more extensive documentation. And perhaps a little more appreciation that asking people to rewrite large chunks of their extension is not trivial for them, and something we should perhaps have given them more warning about.
Gerv, js-ctypes are for accessing 3rd-party libraries from JS (without a dedicated binary component). I don’t believe there is a need to access Mozilla libraries that way (I would be much interested to hear one).
And js-ctype are there since Firefox 3. They were refined since their appearance in 2007, but that shouldn’t be a major problem for compatibility.
I do agree with you that we need more documentation, especially migration tutorials. And to communicate that better to add-ons writers.
teoli: Then I don’t understand. I thought that the problem with binary extensions is that they were using unfrozen internal Mozilla interfaces (which is all of them, now) and therefore needed recompiling for each new version of Firefox. And js-ctypes is proposed as the solution for this problem. If they can’t access Mozilla libraries, how can they solve it? What have I missed?
There are two reasons to use binary addons. One is to use internal Gecko c++ APIs, as you mention. That is not helped by js-ctypes. The other is to include non-Gecko native code in the add-on – for example, code that calls operating system APIs directly, or links to an existing third-party library. This second use case is easily handled by js-ctypes which avoids the need to link to Gecko libraries or recompile for Gecko version changes.
@Ken: My comments may have misled. I certainly wasn’t criticising you for choosing to use Chrome. You don’t have to justify your choice to random interlocutors like me. I was taking issue with this comment: “I think as soon as you have to educate people into using Mozilla, then you’ve lost the battle.”
I don’t think that’s true because Mozilla isn’t a typical product-maker — and there is a measure of open-mindedness and learning required whenever people encounter something atypical. Part of that process is to educate people as to *why* Mozilla is *necessarily* atypical: what are its relative advantages and what are the costs of those relative advantages. Gerv covered some of these relative dis/advantages from the Mozilla side. I covered some from the traditional company side.
These aren’t just issues that may pop up in the future. These are issues that affect Google(/Microsoft/Apple) and its users today that don’t affect (and don’t ever need to affect) Mozilla as a non-profit.
@Peter Kasting: Great comments, but I didn’t say “Google doesn’t care about you” because that would be a mostly incoherent anthropomorphism. Organisations have behaviour, but it would be a stretch to ascribe emotions, whether positive or negative.
Unfortunately, you can’t just will away the dichotomy between a company (any company, not just Google) and a non-profit. For example, one behaviour common to all public companies (but not non-profits) is to always strive for growth (excepting pathological cases). When a company can grow freely and legally, employees and customers benefit.
This is what’s happening with Google at the moment. Thus, we have:
– Web Search
– GMail
– Documents
– Spreadsheets
– Presentations
– Drawings
– Calendar
– Maps
– Earth (and Moon)
– Groups
– Scholar
– Reader
– News
– Books
– Code
– Desktop Search
– Sites
– AdWords
– AdSense/Ad network
– Analytics
– Fonts
– Google+
– YouTube
– WebM
– Sketchup
– Picasa
– Blogger
– Android Mobile/Tablet OS
– Chrome OS for netbooks
– Chrome
Forgive me if I’ve forgotten some. (You might also want to compare and contrast this list to point 2 here: http://www.google.com/about/corporate/company/tenthings.html) I should note I use quite a lot of the applications on this list, in preference to the alternatives.
But a company’s positive growth can’t continue indefinitely. My question for you and for Google is, What happens to Google then? How does a company, that stores more information about the people of Earth than any company in the history of humanity, keep growing? To chain two cliches together, if knowledge is power and power corrupts…
So I hope you’ll understand and maybe forgive me if I draw a very sharp divide between Mozilla and non-profits on the one side and Google and for-profits on the other. Common origins notwithstanding.
None of which should be perceived as personal or particular criticism, because I think the Chrome/Chromium team are doing a great job and (for the most part) are doing it in the right spirit.
Extensions that consists solely of a binary components must be re-arched. They need to have a js stub that will access to the XPCOM internal API. This js stub will use js-ctypes to access the remaining binary libraries.
This will remove the need to recompile the binary component each time.
But this isn’t something neither trivial nor easy to do: it is an important rearchitecturing of the former binary-only extension. This may not be possible in all cases (though in most).
When the API change, no need to recompile, but, like with any other extensions, only the js part need to be changed.
And the automatic tool from AMO used to check compatibility can be used. That way compatibility can happen at the beginning of the aurora phase rather than late in the beta phase. Testing is also easier as the binary lib won’t need to be retested each time as it is not recompiled each time.
“We here at ConceivablyTech still believe that Mozilla can only succeed if it will use all of the assets that has made it strong in the very beginning: It will need a considerable push of innovation as well as its community and ideals.”
Suggested reading
http://www.conceivablytech.com/8450/business/mozilla-we-need-to-not-give-our-friends-the-finger
Gerv, your posts keep popping to the top of Planet Mozilla when new comments are added (just FYI in case it’s not intentional).
Erunno, DigDug2K, Gerv: thanks for the comments earlier. I should make clear I’m not a developer, just an end-user trying to make sense of what’s going on with Firefox these days.
Firefox just like Linux,
Chrome just like Windwos.
I still can’t find Pentadactyl and foxyproxy equevalent plugin in Chrome. Gecko is slow but it hosts variaty. Chrome, like many app of Google, relies heavily on mouse and it seems that webkit is hard to customize on those points.
Firefox should be more……. faster! The memory consuming problem also need to fixed. Because Chrome can do these, they should not be technical problems. More firefox users will run to Chrome till firefox becomes faster and less memory consuming. In my opinion.
gialloporpora’s status on Monday, 18-Jul-11 09:58:54 UTC
Mozilla’s Competitive Advantages: http://bit.ly/oDdmzD. Mozilla competes with 3 massive multinational companies (Google, Apple and Microsoft), how can we possibly win?
teoli: What documentation do we have to assist addon authors in doing this non-trivial rearchitecting?
can you tell me how to hack pls i always wanted to hack someone
trulyinter: read this.
I am one of the many who switched from Firefox to Chrome; I did so about two months ago. Though Chrome does strike me as cooler, faster, and slicker than Firefox — a matter of fuzzy perception and probably marketing as well — that’s not why I switched. I switched because Chrome has stronger security, by virtue of its multi-process architecture and sandboxing. The immediate motivation for me to switch was that the Gmail account of one of my brothers was compromised, and maybe his whole computer, too. He used Firefox, and I suspected that his system had been compromised through an out-of-date installation of Adobe Flash, since Firefox makes no attempt to sandbox plug-ins AFAIK. So because, as far as I know, Chrome has stronger security than Firefox by design, I switched default browsers for the first time in several years. In my opinion, Firefox needs to catch up in this area as soon as possible, or the talk about community, ideals, and goodwill doesn’t count for much.
When I first got to know and test the Mozilla Firefox browser, I fell in love with it. and yes, I still do, far above the so called Google Chrome browser.
Great deal to learning this concept,it is awesome to learning of this blog.nice job ,
please send to mozilla software
I personally think that the goodwill doesn’t convert in any short-time results. But it is the best option for now. I think if the Mozilla development would be more open, transparent, so the ‘world’ could see who actually builds this great software, people would switch, or continue using Firefox. Google uses this strategy all the time.
Making Firefox ad fast as Chrome would help too.;)