Reasons not to worry

The higher-ups at Mozilla like to say that we should focus on building the best browser we can, and not to pay too much attention to our competitors.  This is probably wise when you consider that we are competing against Microsoft, Apple and Google.

But I can’t help reading comment threads relating to browsers on sites like Slashdot, Reddit and Metafilter.  Judging from these threads, few people love Firefox;  a few say things like “I can’t switch because I couldn’t live without Adblock Plus/NoScript/etc.”  Lots of people complain about Firefox being slow/bloated/a memory hog.  Lots of people praise Google Chrome, mostly for its speed.

I realize these communities aren’t reflective of web users overall, but they do represent something of a leading edge.  As do Linux users, which is why I am troubled by Ubuntu’s plan to use Chromium as their default browser.  Also, IE9 looks like it will be a high-quality modern browser;  the sleeping Microsoft giant has finally awoken.

It’s all rather depressing.  So, gentle reader, please tell me why my perception is wrong.  What am I overlooking?

60 Responses to Reasons not to worry

  1. My perspective is that it doesn’t matter that e.g. Chromium is a little faster than Firefox at rendering a page. We’re talking about milliseconds here, and you have to be one sharp-eyed eagle to notice the difference.

    But there are other performance-related issues that *do* bother me. For example, when the UI blocks when there is a lot of hard drive activity. Or the UI blocks when Firefox is loading a website in a different tab. In my opinion, the basic rule is that Firefox should almost never block the user interface.

    The positive side is that there are people working on solving these problems. Provided that they receive the appropriate attention prior to the new release, I will remain a much happier camper than I would be when using Chrome.

  2. Mozilla is unlikely to have more money to hire developers and designers than Google, Apple, and Microsoft. Consequently, Mozilla could decide that it wants its competitive edge to result from engaging developers in collaborative development. Unfortunately the Mozilla team confuses ‘open’ with ‘collaborative’; it would take a major shift in thinking to succeed. Of course they will label this kind of comment as not representing the true community, because they are already ‘open’, which is the highest possible state of being.

  3. I have to reiterate that I think the love for Chrome seen on the web is a little… biased at this point. I have friends and colleagues, some of whom even work at google, who don’t really like Chrome’s UI. I know a few others who do. Its choice. Users finally have it (at least on the desktop). Its an exciting time for the web because of it.

  4. I’ve been a Firefox supporter from day 1 (my name was in the NY-times ad). I’m also a Linux users, as I’m sure a surprising percentage of people in that NY-times ad are. As the recent Mozilla survey showed, increasingly web-developers use Linux as their desktop OS (Ubuntu in particular). Linux users are also *very* likely to be running 64-bit.

    Yet Mozilla brushes-off complaints about the *embarrassingly* slow 64-bit Firefox performance, and poor Firefox performance under Linux in general. The excuses seem to be:

    * No one uses 64-bit
    * No one uses Linux
    * Linux distros are stupid for shipping 64-bit builds to their 64-bit users, should run mixed 64/32 bit environments instead

    Those excuses are sad coming from the usually forward-thinking Mozilla we all know at love.

    Firefox 3.5 was released well over a year ago with TraceMonkey, yet a huge percentage of your important early adopters and influential web-developers still don’t get to enjoy any JavaScript performance improvement. From what I gather, it seems the situation still wont be fixed in 4.0.

    Mozilla is taking a big risk alienating Linux users this way, a mistake Google isn’t making.

    • Nicholas Nethercote

      Jason, I can’t guarantee anything, but from what I’ve heard Firefox 4.0 will support 64-bit.

  5. Nicholas, I hope so. I’d love me some fast JaegerMonkey under 64-bit Linux!

    And for the record, I do *hugely* appreciate all that Mozilla has done. I just don’t like feeling Desktop Linux isn’t taken seriously at Mozilla. It breaks my heart, really.

    Just because Desktop Linux has a small marketshare now, doesn’t mean it will forever. Firefox too had a tiny marketshare not long ago, and now look.

    Cheers!

  6. I don’t know about other OSes, but I’ll never use FF on the Mac because the UI is so horribly non-native. Chrome manages to sit just on my side of the imaginary line in the sand.

    FF seems slow, too. Perhaps it’s slow startup time (since it’s not my default browser, that matters), perhaps its unresponsive UI. Perhaps it’s all in my head. Doesn’t make much difference.

    Should FF be worried? Probably. Standards only take you so far; if market-share drops to the point that people only test sites against IE and WebKit then FF will end up fighting the impossible battle to render more and more like one or the other.

    The advice to build the best product you can is sound. On the other hand, the easiest way to confirm you’re fastest, leanest, stablest, etc. is to perform direct comparisons against the competition…

    If FF does succeed in outstripping the competition, the same people who are switching to Chrome now will switch back quite happily, and are tuned-in enough to discover the reason exists when it does.

    Regardless of what happens, FireFox has played a major role in bringing us to a significant point in web technology where standards actually matter. That will always have been worthwhile. The only question is, should you be investing your time now in JavaScriptCore rather than TraceMonkey?

  7. My 2 cents:

    Stop adding features for a moment or two and fix the multi-year-old bugs that still affect Seamonkey 2, Firefox 3.6 and 4. Seriously.

    Add SVG fonts support, even just the bare minimum necessary to pass the ACID3 test. It’s not the best test, all right, but it is one of the reference tests. Just pass it and be done with it.

    Add small useful features your competitors added long ago but that Mozilla developers refuse to add even when many, many people have been begging them for years. Some examples: Paste-And-Go/Search, ability to clear upload form fields, ability to dismiss infinite Javascript trap dialogs…

    Also, you need to do something about profiles. Keep them clean or have an easy way to clean them. Many people complaining about crashy/slow/memory-hogging Firefox are using these dirty profiles. Having an easy way to clean them/import preferences into a new profile would greatly improve things.

    Extensions are great, but some of them are not very well written and tend to crash Firefox or leak like crazy. Identify them, work with the developers, let the users know that some extensions may make their Firefox experience worse.

    Lags/pauses. Firefox experiences lots of them. When a tab is loading, the whole UI freezes, for example.

    Better information at Mozilla’s home page. Have easy-peasy tutorials on how to use Firefox and take advantage of its many built-in features.

  8. To me, Mozilla’s biggest problem is that a lot of the best ideas are left on the table for too many years before somebody finally wakes up and says “Hey, maybe we should be working on this.”

    #1 on that list is mobile. A little more attention back in the days of Minimo and Mozilla could been leading, instead of trying to play catchup like they are today.

    Dirty profiles is another one. Even back in the days before Firefox 1.0 was released, Mozillazine was loaded with posts saying “Clean your profile or create a new one to fix a lot of slowness/crashes.” Why, WHY are you just now getting around to looking at this?

  9. I’ve been thinking about why I switched from Firefox onto Chrome for a while.

    I think it’s mostly due to (fairly) trivial UI things that just repeatedly annoyed me. Updating Firefox (on MacOS) is overly slow, and after the upgrade I have to click through updating plugins. Running Chrome, I just don’t have to know – everything updates itself in the background, and I can get on with browsing.

    On Linux there was a bug (fixed in FF4) that caused the proxy authentication popup to appear multiple times on startup. If you had done an upgrade, this box appeared once-per-plugin as a result of the plugin compatibility dialog. Unfortunately Fedora ship a lot of plugins by default (language packs), so you end up with hundreds of windows appearing. A fairly simple bug, that languished in the bug tracker forever, so I ended up using other browsers.

    .. so I guess I’d say I’ve switched over to Chrome because there aren’t these little frustrating annoyances.