Followers Or Leaders?

This is a sort of guest post, since it’s the English translation of an article originally published in Italian by Giacomo Magnini (prometeo), a long time contributor to the Italian Mozilla community. Even if I don’t agree with him on some points, I share the overall feeling that comes from his considerations and I thought it could be worth sharing this article with other Mozillians.

I would like to share some not really positive gut feelings about the Mozilla project as a whole. I’m very puzzled about two ongoing trends within the project that frankly I don’t like at all:

  1. Cloning Chrome
  2. Focus on marketing

For the first point, I’d say it started with separate processes for plugins (and the ultimate goal is separating each window or tab of Firefox in its own process, just like Chrome) and is proceeding in a similar manner with graphical appearance. And this happens while many people are complaining about the removal of support for Mac OS X 10.4, the incentive to use Jetpack (once finalized, because for now it is on the high seas) over real extensions and the implementation of not very useful features (and I would say even questionable) like Personas.

It really seems that we are rapidly losing sight of the technological superiority to chase ghosts and competitors, instead of presenting new features. And when you also have to recover architectural holes that you’ve been putting off for years (separate processes, limitations in Gecko, browser’s startup time, old fashioned JavaScript interpreter, etc.), this situation becomes much more difficult.

On one hand it is true that Firefox has achieved very good market share, albeit somewhat uneven overall, but on the other hand the arrival of new very aggressive competitors has found MoCo/Mofo rather unprepared (to say the least). And the heat of competition is pushing the project to throw everything possible in the field: all the available technology, even if not ready, as well as technologies possibly taken/cloned from others, complete rewrites of large parts of the software to eliminate structural weaknesses, etc.

In addition to this, and passing onto the second point, there is this new push in the marketing field, hoping to reach new promised lands where the verb has not yet arrived (without even glancing at the pastures of enterprise installations, as usual), where new recruits should increase market share or at least compensate for the more “geek” ones who have already left or are moving to other shores. So we get the hive of activity of social marketing, viral marketing, guerrilla marketing and so many “cool” or “modern” terms. The problem is that Mozilla launches itself into such operation against three competitors who can rely on unlimited resources and who’ve been in that field for decades: Google, Apple and Microsoft (not to be underestimated, never). And the other problem is that without a better and different (from the others) product, marketing is what it is, i.e. very little.

Mozilla as a platform/embedding software has died, rightly ousted by WebKit: after losing Gnome, few products are still standing, perhaps not for long. The promise of stable long-term releases supported for years has been made to disappear. Instead of having to manage twenty different development branches, and their own releases, Mozilla has introduced a ploy to bring out new and invasive features as a “minor update”, reducing support for older versions to only 6 months. We’ll see if even this commitment holds since there’s a run-up to Chrome in action.

Mobile Firefox has some space only through MeeGo, though it is already light years better than the situation until the last week with only Maemo… The market share of WinMobile is disappearing like snow in Rome: the version 7 will have to sweat a lot and be damn good. Android, iPhone, RIM and webOS are closed. The version for Symbian exists only in its epitaph.

I deliberately did not comment on Thunderbird: let’s say I gave it the benefit of the doubt, given the huge technological gap that it had to fill in a short time. I’ll just note that Postbox has some very positive press and it seems to be moving beyond TB abilities. And do not tell me that email clients have had their day: there is no gmail-like thing, not even in picture, which holds its own compared to a well done specific client.

I hear the frame creaking, and I do not like it.


Posted

in

,

by

Tags:

Comments

54 responses to “Followers Or Leaders?”

  1. […] questo articolo di un collega mozilliano. Se poi volete farvi del male, potete anche leggervi la relativa traduzione in (quasi)inglese. Tag Technorati: ballot screen, commissione europea, firefox AKPC_IDS += […]

  2. David Adam Avatar
    David Adam

    I’m only tangentially engaged with the Mozilla community, but like Francesco I agree with the broad brushstrokes of this argument.

  3. Adam Avatar
    Adam

    Yeah. I agree with the setement there.

    Mozilla is a lot about marketing instead of getting a product that is the best that it can be. Why does mozilla need 50+ websites? Every new project does not need a URL… Also new projects don’t need a big announcement and a youtube video, before a week of code has even been started.

    Also, with personas, how will that be useful when the new skin comes out that crunches out most of the layout space?

    Also.. From what I have seen on blogs on planet mozilla.. there are only 2 main developers for bugzilla? Shouldn’t that be sorta high on the priority list of having people work on?
    ___
    On the other hand, I will give mozilla credit for awareness of web issues like privacy and layout, and standards… not that the browser follows it itself.

    On the other hand.. mozilla’s marketing of the browser is more fluff then anything.

    Firefox does not have Acid 3 support, or the fastest rendering of anything people currently pay attention to.

    I have been thinking of moving to chrome, not so much because I don’t like firefox, but because the mozilla brand has become more hipocracy then I can palate.

  4. FP Avatar
    FP

    As someone who has been very peripherally involved (as a nightly tester and extension developer) with Mozilla for the last 5 or 6 years, I’m also starting to lose hope. Mozilla has lost the support of the important power user and web developer camps and seems uninterested in reclaiming it. In case anyone cares here’s what I’d do to try and win it back.

    1. Pass ACID3 now. I understand the arguments that there are more important web standards to implement and that WebKit arguably cut corners in order to pass the test. However as Adam’s post above demonstrates the vast majority of web developers and power users do not understand this, they see Mozilla as not focusing on web standards anymore, this is clearly wrong but nonetheless it appears to be a widespread belief. ACID3 is an incredibly important marketing tool and not passing it has hurt Mozilla badly.

    2. Sort out the memory issue once and for all. Again, I’ve seen all the benchmarks, I know that Firefox almost always wins memory benchmarks. However that doesn’t stop memory usage being one of the biggest complaints I see on power user sites. It may be that people who experience it are using poorly written extensions, or perhaps certain sites cause issues, who knows, but Mozilla needs to reach out to the people having memory issues and figure this out once and for all.

    3. Sort out the bloat perception. A phrase I often see used to describe Firefox is “it has gotten slow and bloated”. Yet again, I don’t believe this for a second but it appears to be a common perception. Mozilla Marketing could help by gathering a wide range of benchmarks and putting up a site showing the results for 1.0 vs 1.5 vs 2.0 vs 3.0 vs 3.5 vs 3.6, the site could provide easy links to the browser downloads and the benchmarks in order that the results could be easily replicated. It’s also possible that some people are genuinely seeing Firefox get slower with each version in which case again Mozilla should reach out directly to these people and figure out why.

    4. Politeness. Stop being so damned rude to people in Bugzilla, anyone who has taken the time to report a bug should be thanked and given gentle guidance on what extra information is needed, not brusquely told that they didn’t file it correctly. Every time that happens you lose a potential contributor.

    End of moan. Yeah I realize this stuff is much harder than it sounds and that as an outsider I don’t really have the right to tell you what to do but I had to get that off my chest.

  5. anonymous Avatar
    anonymous

    I would rather say in order to keep up with the competition, Mozilla must innovate to form better products. I think you just don’t know about the competitive strategies. Mozilla, per se, is not cloning Chrome. As I’ve said earlier, Mozilla is just innovating to keep up with the competition. Jetpack and the new UI is just an example of such innovation(and not a copy of Chrome)and both projects are still under development not subjected to a final Firefox release.
    By the way, you’ve said that “It really seems that we are rapidly losing sight of the technological superiority to chase ghosts and competitors, instead of presenting new features”. How about Jetpack and the multi-process plugins(OOPP) would you not consider them as new features? You’ve also said earlier that “…browser’s startup time, old fashioned JavaScript interpreter, etc.” FYI, Mozilla is now currently working with Jaegermonkey a new addition to the Javascript engine, Spidermonkey, to further improve Firefox speed and to keep up with Chrome. Moreover, Mozilla also has this Electrolysis project wherein it is specialized to improve Firefox’s launch time, responsiveness and stability.

    A tip for you: Try reading the Mozilla wikis and be updated with all of what’s happening within Mozilla.

  6. flod Avatar

    @FP
    It sounds like you just read my mind 😉

    @anonymous
    Even if I’m not the one who wrote the original article (and he knows Mozilla a lot better than me), I’m following Mozilla’s development – with all my technical limits, since I’m not a software developer – and still I share his sensations.

  7. Toni Hermoso Pulido Avatar

    @anonymous and @flod , certainly there are many improvements going on, and imitating, not blindly, when things are good must not be a bad thing per se. One historical example is tabbed browsing, which firstly appeared in IBrowse and Opera.
    However, the present apparent abandonment of, IMHO, innovative projects such as Ubiquity in Mozilla Labs makes me sad… More resources and attention should be devoted to people striving in this kind of research.

    Regarding Gecko, when Epiphany decided to switch to Webkit, almost 2 years ago (http://blogs.gnome.org/epiphany/2008/04/01/the-future-of-epiphany/) this was a clear warning that things maybe had not been going the right direction. Related to this, I don’t know if a rumour is Flock switching to Webkit http://techcrunch.com/2009/03/02/flock-ditching-firefox-moving-to-google-chrome/

    My 2 cents.

  8. Michael Lefevre Avatar

    I don’t know if the previous comments think they are agreeing, but they seem to be going in different directions. The original post is complaining that Mozilla is following Chrome, instead of doing new and different things. Then the comments are complaining that Mozilla isn’t focusing on doing the things that Chrome has done (stability with multiple processes, etc).

    On the technical side, I don’t think it is mostly the case that Mozilla is copying Chrome. Several of the things that Chrome has done have been worked on already in Mozilla, but Chrome has started later and finished earlier.

    As for the marketing, I agree that all the different marketing activities and websites seem a little crazy, but that is not taking resources away from the coding – a community marketing person cannot be “reallocated” to working on the code.

    So I kind of agree with the sentiment a little, but I think the main issue is that people are expecting Mozilla to keep up with Chrome by working the same way, which is rather unrealistic – Chrome has a much much bigger budget (for coding and for marketing), and doesn’t rely on volunteers for doing a lot of its work.

  9. flod Avatar

    certainly there are many improvements going on, and imitating, not blindly, when things are good must not be a bad thing per se

    That’s certainly a good point. The problem is that, in the last period, we seem to improve only by catching up with others instead of improving through innovation as we used to do.

  10. […] @flod! You touched interesting topics in your latest post, and when I started crafting my comment it got so lengthy, I decided to use my platform to deliver […]

  11. Peter Lairo Avatar

    I very much agree with the sentiment and details of the article.

    – Copying Chrome’s tabs-on-top is a mistake (you carry the whole UI in each tab, and you lose the page’s title in the title bar, tabs disjointed from content).

    – The Mozilla platform (Gecko?) should already be much more widespread.

    – Corporate use should be more emphasized (e.g., MSI packages, bug 436259)

    – Thunderbird is directionless (e.g., lots of work done on messing up the header pane instead of fixing important bugs (bug 250539, bug 391057, bug 446444)

    I love Mozilla, and hope they will find a better focus.

  12. Michael Lefevre Avatar

    “That’s certainly a good point. The problem is that, in the last period, we seem to improve only by catching up with others instead of improving through innovation as we used to do.”

    But, in the previous period, there was more innovating, but then everyone (like FP above) was complaining about the lack of progress (and even moving backwards) on performance, memory use, crashes, and fixing bugs in ACID3.

    You can’t do everything at once, and, as I already commented, you especially can’t expect Mozilla development to move faster when it has less resources and more history behind it.

  13. Michael Lefevre Avatar

    Oops, missed a tag there somewhere…

  14. Pascal Avatar
    Pascal

    I find your point about Personas intriguing – you write: “[…] the implementation of not very useful features (and I would say even questionable) like Personas.”

    This for me a typical engineering driven argument – Personas is used by 20% of Firefox 3.6 users, has more than 50,000 created Personas and thus is clearly one of the most widely and successfully adopted features in 3.6.

    You might not like it and even find it “questionable” – but arguing in this way you only show that you don’t understand the typical user of Firefox.

  15. flod Avatar

    @Michael
    Fixed the closing tag 😉

    But, in the previous period, there was more innovating, but then everyone (like FP above) was complaining about the lack of progress (and even moving backwards) on performance, memory use, crashes, and fixing bugs in ACID3.

    If you change, a lot of people will complain, no matter what direction you take. It happened for the Awesome Bar in 3.0 or the behavior of the last opened tab in 3.5, it’s happening for the way new tabs open in 3.6.

    IMO it’s already difficult to innovate, almost impossible if you want to remain “mainstream” and have a six months release cycle with limited resources.

  16. Pavel Cvrček (JasnaPaka) Avatar

    I’m Mozilla contributor for 7 years. This is not first post where I see something like “clone Chrome”. But I don’t think that Firefox is cloning Chrome. When was Chrome released for the first time many users said “hey, this dialog looks like from Firefox”, ” hey, I know this from Opera”… etc. Was Chrome clone of Firefox or Opera? No, they have a look at browsers and they inspired. Do you remember how some years ago many users was saying “hey, Firefox just copy Internet Explorer”. I see analogy.

  17. Nuss Avatar
    Nuss

    First of all, this is how the market works. A newcomer, makes a heck of a good product. It wins market share at the cost of the product that owns the market. Repeat.

    Granted, Mozilla has a set of problems. The biggest problem is that development is slow. Too slow. This is something that has gotten better, and seems to become even better in the future. Transforming Mozilla Suit to Firefox took ages. DeCOMptamination takes time, losing old technology takes time. Fixing startup time takes time. Everything takes time. You have to see the whole evolutionary picture of Firefox to understand why development is slow.

    What Apple and Google have done is remarkable. Of course Firefox should have OOP, Aero Glass support and all the new technologies that the competitors have. I can’t understand why anybody would be against that. Of course new features that doesn’t break compatibility should be introduced in minor updates (read OOP). Jetpack and Personas are innovation, Taskfox is innovation. Firefox Mobile is innovative. Weave is innovative.

    For me it seems like Mozilla has gained momentum and I sincerely thank Google and Apple for the competition.

    As for embedding: it’s sad to see developers abandoning Gecko, but Gecko is improving and while WebKit is better for them today (as Gecko was better then). The tables may turn once again in the future.

  18. flod Avatar

    You might not like it and even find it “questionable” – but arguing in this way you only show that you don’t understand the typical user of Firefox.

    You’re absolutely right when you say that I’m not the average Firefox user. Enough said, I’ve always used the default theme from the times of Phoenix and have less than 10 extensions installed.

    Maybe I’m also a little biased about Personas, because from a l10n point of view it was quite a mess.

    Personas is used by 20% of Firefox 3.6 users

    Interesting. My perception, based also on the number of questions that I see in our support forum, was that Personas isn’t that popular. Are these statistics available somewhere?

  19. Pascal Avatar
    Pascal

    Stats on GetPersonas.com here:
    http://mozillalabs.com/personas/2010/02/12/57000-and-going-strong/

    Not sure if the 20% number was written up yet…

  20. Të Avatar

    One important thing to remember here, Mozilla is so much more than just Firefox.
    http://www.mozilla.org/projects/

    As pointed out by Michael Lefevre, a lot of things that are “copied” have actually been worked on or at least been on the todo list since long before Google Chrome.
    For example separate processes for plug-ins: https://bugzilla.mozilla.org/show_bug.cgi?id=156493

    I’m not surprised if the open-source nature of Google Chrome helps to push Firefox in certain directions.

    I can’t see what’s wrong with the marketing. It’s usually fun or interesting stuff posted on the site appropriate for the target audience.
    There’s not really that many sites either that do marketing, of the top of my head I can think of The Mozilla Blog, Mozilla Labs, Spread Firefox and about:mozilla (well, sort of).

    Although, I would like to see some long forgotten pages be updated, like the various guides ( http://guides.mozilla.org/Mozilla_Community ) and the Mozilla Memory Bank ( http://mozillamemory.org/ ).

    Mobile Firefox isn’t even a year old yet, is it? Calling it dead already surely sounds pessimistic.

    @Toni Hermoso Pulido

    InternetWorks and NetCaptor were the first browsers with tabs.

    Ubiquity is being developed, they have just ran into a roadblock.
    http://jonoscript.wordpress.com/2010/02/18/quick-update-on-ubiquity-and-firefox-3-6/

    Hopefully Taskfox will appear some time soon though.

    @FP

    1.
    ACID3 isn’t a standard of any sort. It would be better to teach the web designers¹ that the parts in ACID3 that Firefox doesn’t support correctly are minor stuff.
    https://bugzilla.mozilla.org/show_bug.cgi?id=410460
    http://spreadsheets.google.com/pub?key=pNgBCwWdyRTT2JeiZn4B2Yw
    ¹ I would never call someone who doesn’t understand this a “web developer”.

    I would really not like to see them do hacks like this just for marketing.
    http://trac.webkit.org/changeset/31322

    2.
    Once and for all? Sure as long as you solve the feeble problems of world peace and starvation. 😛
    Detecting bad extensions aren’t really feasible given the nature of extensions. The future multi-process tabs should help with the bad behaviour of websites though.
    There are tools available for manual testing though: https://wiki.mozilla.org/Performance:Leak_Tools

    3.
    Like this?
    http://www.mozilla.com/firefox/features/#super-speed

    Mozilla can’t be omnipresent, it’s the users who’ll have to reach out. Which they can easily do by simply pressing F1 in the browser.
    Although, they are trying with the SUMO project.
    http://blog.mozilla.com/sumo/

    4.
    How long ago do you remember seeing this? I haven’t looked at every bug out there, but I can’t remember seeing any of this behaviour.

    By the by, this blog post is number nine on Google, for me, when searching for: politeness on bugzilla mozilla.

  21. pascalc Avatar
    pascalc

    just for the record and because i am probably the pascal that Francesco is usually in contact with, Flod I didn’t write the above comments signed Pascal, that’s another Pascal 😉

  22. flod Avatar

    Flod I didn’t write the above comments signed Pascal, that’s another Pascal

    Yep, I knew from the e-mail address 😉

    Too many William and Pascal in Mozilla 😛

  23. Boris Avatar
    Boris

    Just one thing… “The promise of stable long-term releases supported for years” is not something Webkit has. So that’s not an issue in the platform/embedding space.

    What _is_ an issue is that most people want to embed an HTML renderer, not a web browser. There’s a huge difference between the two. Webkit is the former. Gecko is the latter.

  24. prometeo Avatar
    prometeo

    @anonymous:
    I’ve been following bugzilla since December 1999, long before a wiki existed, so I guess I know a bit about Mozilla’s story… I do follow some wikis and many newsgroups. As an example, since he wrote a lenghty response, I’ve seen Gandalf coming, going to Flock and coming back.
    Did I ever say multiprocess is bad? No, actually Mozilla started to think about it after Milestone 13 (or was it 15?), when Mailnews came out and crashed on launch: they wanted separate processes in order for mail not to knock down the browser.
    Did I say multiprocess is not original? Yes, since chrome (and even IE in some ways) got there first! Mozilla spent 10 years doing nothing about it while others delivered: that’s simply the whole truth. Now they are catching up using Google code (which is not a bad thing IMHO) with some problems they have not fully understood since the code is not theirs (memory usage at compile time for the virtual machine being one of those: did you read this on a wiki? I guess no).
    About Jetpack: I’ll believe it when I’ll see it working in a real release (minor, probably) and it is proven that old style extensions are still possible (e.g. Chromebug). See the discussion on the newsgroups for details, at least the only viewable groups, since the main group for Jetpack is not easily readable.
    About JavaScript: Mozilla’s JS engine has always been the most standard compliant, thanks to Brendan Eich being in the project since the beginning. But many other engines came and run away, based on new reasearches and papers, and on new techniques which were hardly adaptable to the current Mozilla engine: so we got SpiderMonkey, TraceMonkey and tomorrow JaegerMonkey. But WebKit, Google and Opera engines are already here and are delivering now.
    Did I say Mozilla engineers are uncapable? No, actually I have spoken (and eaten and drinked) with bright examples of smart and passionate people.
    Were they put at good use by developing Personas? (BTW, just an example, this is not a war on Personas!) YMMV, but I guess no, when you have much more pressing technical nodes to solve. And which someone else has already put behind their backs.

  25. prometeo Avatar
    prometeo

    @Boris:
    true, but integrating webkit is a one liner, and that means ANY version of webkit, for years probably. Integrating Gecko & friends has been a nightmare, always: 10 years ago it was acceptable, since no competition was around, nowadays you have to play catch up with others running away…
    But if so many projects are switching to webkit, probably integrating the whole browser is not so useful. And losing other eyeballs looking at your code is not good, IMHO. Yeah, you probably earn more users by getting “pretty”, but they don’t contribute a line of code, ever.

  26. prometeo Avatar
    prometeo

    @Nuss:
    I see and saw the whole evolutionary story, and I understand that working today on the same things Mozilla was working 10 years ago is not good. And now there is competition.
    Did I say they did everything wrong? No. When Mozilla decided that memory usage and crashes were real limiting problems, they delivered and in a short time, IMHO.
    Fennec/Mobile is innovative, no question. Still, there is little room left there, for a number of reason, one of them being late to the market. But why not be innovative in the same way on the desktop? Why copying chrome UI instead of providing something new? We all use super wide screens and you are still putting tabs on the top? Just an example.
    What I’m saying is that trying to innovate while you are trying to close the gap you let open for long time won’t be easy. And trying to use marketing to stay up there until you can make a come back technically is scary and dangerous. And if Personas is an example of technical supremacy, than I see no future.
    For embedding, see previous post.

  27. gerv Avatar

    This post doesn’t really explain what the poster’s suggested alternatives are.

    E.g. is having each window/tab in a separate process a good idea or not? If it’s a good idea, we should do it whether Chrome got their first or not. If it’s not a good idea, explain why.

    I think it’s a great idea, as is out-of-process plugins (based on the same work) because when Flash crashes or hangs, it won’t take down my browser. That’s a feature worth having.

    And is having a stable API (Jetpack) for 90% of extensions so they don’t have to be updated with each release a good idea or not? If it’s not a good idea, and you’d prefer to stick to the world where we have to hassle thousands of addon authors to do work every time we release a new version, and where users can’t update because their addons don’t work yet, explain why that’s actually better. Because it doesn’t sound better to me.

    And what’s your point about marketing? Should we be doing no marketing at all? Or if we are doing marketing, what sort should we do? You criticise our “social marketing, viral marketing and guerrilla marketing” – but what are the alternatives? Where do we get the money to do poster advertising across Europe like Chrome are doing? Their marketing budget for a year is about the same size as Mozilla’s entire budget for everything. Would doing no marketing at all suddenly make the product better? Do you want all the marketing people to start coding?

    And you say it’s terrible that we are no longer embedded by anyone – but you acknowledge that this requires commitments to long term stable branch maintenance that you agree we can’t make. So what’s your plan to get us back to being embedded by lots of people? And how is that a bigger win for keeping the web open than focussing on growing the market share of Firefox and Fennec?

    This all sounds a bit like “Mozilla isn’t doing what _I_ want”, without explaining why what you want is the best thing for the web and for Firefox’s 350 million existing users.

    Gerv

  28. prometeo Avatar
    prometeo

    Dear gerv,
    I cannot help you if you don’t understand what I was trying to explain: I think the post is clear and I had some more precise comments up here, please read them.
    The alternative is going back to the drawing board, just ike Mozilla did with Fennec, and come out with something new, showing the way ahead to the competitors, instead of simply catching up with them! Something is in my last post, so don’t try to shut me up by saying that I’m only complaining, that is not really fair.
    I’m just a contributor, a user, a localizer, a supportive voice, I don’t decide on Mozilla’s behalf, but I’m not a fanboy and I keep my eyes (and brain) wide open.
    If you cannot bear some open criticism, then the situation is much worse than I thought.

  29. prometeo Avatar
    prometeo

    @Toni:

    More resources and attention should be devoted to people striving in this kind of research.

    So there are people out there who did understand the meaning of my post! Thank you, I guessed it wasn’t clear enough.
    About multiprocess, please read the other comments: I have never said it is a bad thing. I’m saying that the problem was known and clear to Mozilla for the last 10 years, and it was solved by Google: not something to be proud of in my vocabulary.
    And another sign of lost technological supremacy.
    If people still thinks that this can be compensated by implementing Personas…

  30. gerv Avatar

    But “new” is not a good thing in itself. Why is process-per-tab so bad that we shouldn’t implement it? Why is having a stable extensions API a bad idea? Why is it a good thing to go in a totally different direction to e.g. Chrome, when they have smart people thinking about the same things we are thinking about? If two groups of smart people consider a set of problems, they often come up with similar solutions. This is confirmation that they might be good solutions, not evidence that they are bad ones.

    And who said I can’t bear criticism? I wasn’t objecting to your criticism, I was suggesting that you need to back it up with reasons. “Process per tab is bad” is not an argument, it’s an assertion. Why is it bad?

    Gerv

  31. Boris Avatar
    Boris

    prometeo, “integrating webkit” is nothing like a one-liner, unless you’re using someone else’s solution for integrating it. And if you are, then you could just as easily have used someone else’s solution for Gecko, had someone written one.

    So it sounds like you real complaint there is that someone (not the webkit project last I checked) has written a good wrapper for webkit for your environment while no one has stepped up to write a Gecko wrapper for your environment?

  32. iacchi Avatar

    @gerv: I think you still haven’t caught the point of this post. I’m not prometeo, but I’ve discussed with him a bit and read the original italian post (maybe the translation can be bad in some point) so I think I can reply for him (he told me he’s out for work).

    First of all, let’s get one point straight: are out-of-process plugins and window/tab in a separate process a bad idea? Obviously not. It’s a great thing to have, and fortunately it’s being developed.

    Is jetpack a bad thing? I think prometeo’s answer is: time will tell. I don’t know so much about jetpack, so I don’t want to comment any further on this point.

    So, what’s the “bad” prometeo’s writing about? Why he says marketing is bad? Using his example, Mozilla thought about the OOPP and separate processes about 10 years ago, but instead of innovating and setting a new technological level, they just put the whole thing in stand by and started considering it againg only when Chrome and the other competitors already implemented it because it couldn’t be delayed anymore; this (I think) is what we want to underline with this post.

    About the marketing, from what I can see (and, I admit it, it is very little actually) in these years Mozilla has hired a lot of new employees, but mostly in the marketing department. So: marketing (unfortunately) is important for a product to grow, but wouldn’t have it better to focus a little less on the marketing and a bit more on the code?

    I hope it’s a bit more clear now.

  33. Robert O'Callahan Avatar
    Robert O’Callahan

    About Acid3:
    If you turn on the HTML5 parser, we get 97/100 on trunk. The three remaining test failures are SVG fonts, which absolutely no-one uses or cares about. I bet most people who complain about the Acid score don’t even know what SVG fonts are, and wouldn’t use them if they did.

    About Epiphany “switching” to Webkit:
    How’s that working out? That’s been in the works for two years, and as of the last Gnome release they still don’t have basics like an HTTP cache.

    As for innovation, how about this stuff, for example?
    http://hacks.mozilla.org/2009/12/file-drag-and-drop-in-firefox-3-6/
    http://hacks.mozilla.org/2010/01/industry-support-for-woff-and-firefox-3-6/
    http://hacks.mozilla.org/2009/10/font-control-for-designers/
    http://hacks.mozilla.org/2009/10/orientation-for-firefox/

  34. Robert O'Callahan Avatar
    Robert O’Callahan

    If you look back at the history of Mozilla, it’s hard to identify a time when we could have taken a couple of years off to reengineer everything for separate processes. Taking a break from the battle with IE would have been suicide for us and for the open Web.

    We’re only able to work on it now (while also doing releases, improving performance, adding features, etc) because we have a lot more developers than we ever had before. We have hired a lot of developers over the last few years and we have far, far more developers than marketing people. I don’t know why iacchi thinks otherwise.

  35. flod Avatar

    We have hired a lot of developers over the last few years and we have far, far more developers than marketing people. I don’t know why iacchi thinks otherwise.

    Maybe I’m wrong, but I think it’s a common sensation watching Mozilla from the outside (and with outside I mean both “outside US” and “outside Mozilla”). Or maybe marketing people are just more visible 😉

  36. Robert O'Callahan Avatar
    Robert O’Callahan

    Of course marketing people are more visible, it’s their job!!!

  37. Robert O'Callahan Avatar
    Robert O’Callahan

    About innovation in general, hacks.mozilla.org has a good collection of articles about where we’re innovating for Web developers — for example, WOFF fonts, File API and drag-and-drop, orientation sensing, and better CSS gradients. (I’d post links but this blog seems to reject comments with links as spam :-).)

  38. flod Avatar

    @robert

    Sorry, I’m going to check the spam queue, approve your comments and remove duplicated comments where necessary 😉

  39. flod Avatar

    Just a note: recovered also Të’s comment here 😉

  40. iacchi Avatar

    @robert: as I said before my POV window is quite narrow. I have a big bunch of thing to do during the day, and I spend my “Mozilla time” doing things more that getting informed. For example, I didn’t know hacks.mozilla.org, I’ve just added the feed on my reader.

    Anyway, the OOPP was only an example as I said, and I’m sure prometeo (not me) has many more of them when he’ll be able to come back to the discussion.
    Just an other thought on this subject: couldn’t have we (you, they, whatever) hired more developers to work on this, say, two or three years ago instead of now? Surely Mozilla had the money to hire them two/three years ago, so why not?

    About marketing, as flod said, from the outside this is what you get. Maybe is wrong; still, this is what you get.

  41. gerv Avatar

    iacchi: so it seems the argument about process separation is not “we should be doing something different now”, but “we should have done something different 10 years ago or 5 years ago”. Whether that’s true or not (and I think roc has explained well why it’s not), why is it a useful thing to say?

    Why didn’t we hire more people 2/3 years ago? Because organizations can only grow at a certain speed. Netscape grew faster than that speed, and it caused all sorts of problems. Doubling your organization in size every 6 months is not a recipe for sustainable growth. You need the right people, too.

    And on the marketing point, looking at the org chart, I think Mozilla employs roughly 25 people who could be said to work in marketing. For an organization with 350 million customers, that’s not very many. And it’s far fewer than the number of developers, which is a couple of hundred, I think.

    Gerv

  42. FP Avatar
    FP

    Some people seem to have misinterpreted me. I wasn’t claiming that the missing ACID3 points are important in their own right, I was claiming that not implementing them has seriously hurt Mozilla’s image amongst web developers and power users. See Adam’s post for an example, as far as I can tell from reading tech sites that sentiment is widely held.

    p.s. @Të I also wasn’t claiming that ACID3 was a standard in its own right, I meant “… there are more important web standards than the web standards that would need to be implemented in order to pass ACID3…”, perhaps I should have spelled that out more clearly but at the time I didn’t think it was necessary.

  43. Të Avatar

    @Flod
    So that’s why I never got any subscription emails! 🙂

    @FP
    Ah, ok, sorry about that. It was a bit late (or early actually :P) so it were probably just me.

    I still do think that these web designers should get their act together and actually learn the basics of their field instead of complaining about things that doesn’t affect them.

  44. prometeo Avatar
    prometeo

    Dear Gerv,
    in my answer to anonymous I wrote:

    Did I ever say multiprocess is bad? No…
    Did I say multiprocess is not original? Yes…

    and you are still putting words in my mouth I’ve never used.
    Well, the answer I wanted was given by RoC:

    If you look back at the history of Mozilla, it’s hard to identify a time when we could have taken a couple of years off to reengineer everything for separate processes.

    I do remember Benjamin writing off OOP just a little before chrome came out, because applying it to the UI and the extension system was simply “very hard”: do I remember rightly here, or am I a visionary?
    Then, just a few days after Google released the code, Benjamin or another engineer (don’t remember exactly here) had plug-ins running in a separate process: am I right?
    So it could have been done before, but was decided otherwise: that’s a strategic decision I won’t and can’t question. So when the willingness to do OOP was swapped with the need of chasing the competitors, devs have found the needed cycles, and delivered.
    I’ll repeat myself once again: multiprocess is not bad is late!

    Going back once again to my original post, I was simply expressing my fears that marketing is being used to cover up technological gaps FF has collected through the years while adding the same features offered by new competitors: I wasn’t implying that marketing is bad, per se.
    I wanted some constructive talks, with glimpses of great news coming, not a flame war. But no sign of the first (except for RoC and MaK on his Italian blog) and lots of signs of the latter: seems like I hit a raw nerve.

    BTW, (and this is not a flame, just a funny note), 3 different Mozilla employees have used the example of the 2 groups of smart people coming to the same conclusions… Company’s mantra? 😛

  45. prometeo Avatar
    prometeo

    Dear RoC,
    for how long Gnome shipped with Gecko? How much work did they put into refining the integration, fixing Gecko bugs and more? Still, they decided to jump ships: probably a “marketing” decision as Boris suggests, and it wouldn’t be the first one (mono anyone?). But in the end, less eyeballs looking at your code in the open source ecosystem is not good: do we agree on this?
    XUL Runner had no official release for what? 2-3 years? Static FF/TB (even SM I suppose) will make it even less useful or widespread. Am I right here?
    Some of the features you mentioned are great indeed. Some others you are working on (2D acceleration, WebGL, etc.) are even more attractive. But is Mozilla leveraging on those inside its own products or are they just adding items to the list of features the others don’t have? How are they making an impact for FF daily users?

  46. gerv Avatar

    “I do remember Benjamin writing off OOP just a little before chrome came out, because applying it to the UI and the extension system was simply “very hard””

    Yes, I think he said something like that. But it turned out that we have very smart people who can do hard things. And I’m sure looking at Chrome’s design helped.

    And bear in mind we are a distance from having it working completely and shippable, even now. We’ve been focussed on the more limited goal of out-of-process plugins. Getting that total separation is a lot more work.

    “I was simply expressing my fears that marketing is being used to cover up technological gaps FF has collected through the years”

    You would prefer that our marketing department concentrated on telling people about our weaknesses?

    “3 different Mozilla employees have used the example of the 2 groups of smart people coming to the same conclusions…”

    I have not consulted anyone else in writing my responses to your blog post; there is no “company line” and no collusion.

    Responding to your second post: in your view, why is doing XULRunner an important goal for the Mozilla project?

    “But is Mozilla leveraging on those inside its own products or are they just adding items to the list of features the others don’t have?”

    Hang on a minute… earlier you were complaining about the lack of innovation and differentiation. Now you are saying “what? are you just _innovating_? That’s not good enough…” It’s starting to look like you are working hard to find something to complain about.

    Gerv

  47. prometeo Avatar
    prometeo

    Did I say Mozilla doesn’t have smart people? Never, I said quite the opposite.
    I would prefer having engineers filling holes instead of marketing people trying to hide them, that’s all. Reminds me of other big companies’ strategies, which I don’t like but neither I care. But I care about Mozilla, and I don’t like it going the same route (or at least that’s what it seems to me and others).
    If XUL Runner isn’t an important goal (IMHO spreading the platform is always good, since more people use your code and find bugs or improve it), just say so; that’s fine. Mozilla already did this with MAS, and the project lived on anyway: why not doing the same here?

    On the last part of your post, I’ll try to explain me differently with an example.
    Long ago, someone came out with the bright idea of using some sort of db engine to manage bookmarks, history, etc. So sqlite was added, Places was born and the Awesome Bar was developed. This is what I mean with leveraging on your own technologies: adding features that allow real innovation, impacting all FF users. All was done with a goal and a direction in mind, planning most of it, right?

    I’ll give a counter-example of a technology added to Mozilla which was of no use: P3P, and we all know how it ended. Adding “n” features for the sake of it is just feature creep, which leads to bloat, not to innovation.

    I have to repeat myself once more (sorry, but this will be the last time):

    this is not a flame, just a funny note

    There is none so deaf as he who will not hear…

  48. gerv Avatar

    “I would prefer having engineers filling holes instead of marketing people trying to hide them, that’s all.”

    But it’s not an either/or. It’s not like we could grow the engineering team any faster if we didn’t hire marketing people.

    Success at Mozilla’s mission of keeping the web open requires Firefox market share. Without that, we are irrelevant. And a way to grow market share is to make sure people know about Firefox. We’ve got an amazing community who do a great deal, and (as I’ve said) we have far fewer marketers than most companies with 350 million customers.

    Are you saying we shouldn’t do marketing at all? Or that all our marketers should be volunteers?

    “If XUL Runner isn’t an important goal (IMHO spreading the platform is always good, since more people use your code and find bugs or improve it), just say so; that’s fine.”

    My understanding is that it’s been said several times over the past few years. But don’t take XULRunner policy from me; get it from someone like bsmedberg.

    You seem to be saying we should only add technologies if we know they are going to be successful. In other words, we shouldn’t make mistakes. Can I borrow your crystal ball, then? It would be very useful… 🙂 But seriously, that’s why we now have Labs – so we can do experiments and see what works and what people like. Personas was a successful experiment. Snowl was interesting but didn’t go very far. Ubiquity, we haven’t quite cracked yet – the jury’s still out.

    Gerv

  49. Të Avatar

    About XULRunner
    I’m pretty sure there’s a new stable out for every new release of Firefox. No?
    http://releases.mozilla.org/pub/mozilla.org/xulrunner/releases/

    About ACID3
    A semi-related comment by Alex Faaborg on the default placement of the tab-bar, a minor issue, in my opinion, that is perceived as important by the general user-base much like the last points on ACID3.

    There is a general perception that placing tabs in a more conceptually pure
    configuration is “modern and sleek” while the traditional placement is “old and
    clunky.” This is purely an emotional design and brand level consideration,
    primarily caused by Chrome being the “new” product in the marketplace, and
    Firefox presently being the “old” product in the marketplace. Had we made the
    change with Firefox 3 on Vista, we would not currently be facing this
    situation. However we are now at a state where deploying the more conceptually
    pure design is “copying” or “catching up,” and failing to deploy it may be
    interpreted as “stagnating” or “falling behind.” The product perception issues
    related to tab placement appear to be a lose/lose at this point.

  50. prometeo Avatar
    prometeo

    Gerv,
    if I really had such a crystal ball, I probably wouldn’t have started this discussion at all…
    If you really can’t make mistakes, I’d bet on you anytime. Obviously that’s impossible.
    Adding all possible technologies doesn’t help much, right? But adding a few as well planned and delivered as Places/Awesome Bar is well in the reach of the talents you have there / you are.
    Labs are a wise and smart investment: but reading that Ubiquity was put on hold to finish up other projects is a sign of under staffing in my book. They also came well before marketing was expanded so much, or did I get it wrong?
    Better technology is what made FF successful, and I really hope Mozilla won’t forget about this (as well as doing all the other wonderful things they are doing, ok).

Leave a Reply

Your email address will not be published. Required fields are marked *