Developers › A Journal from Behind the Trenches
Is Firefox doomed?
Randall C. Kennedy at InfoWorld seems to think that Firefox is doomed.
Why is it doomed?
It isn't entirely clear to me, really. His argument seems to be that the delay of Firefox 3.6 shows that Mozilla is in chaos, and that their talk about restructing their development process means that they are in real trouble. That they can no longer maintain their product. Shorter release cycles apparently means that Firefox won't be tested much anymore, as they are bypassing the lengthy testing process and releasing numerous small changes without testing them (sufficiently) first, like Microsoft. And even Microsoft can't pull it off, he says.
I find his argument to be rather weak.
Why is it doomed?
It isn't entirely clear to me, really. His argument seems to be that the delay of Firefox 3.6 shows that Mozilla is in chaos, and that their talk about restructing their development process means that they are in real trouble. That they can no longer maintain their product. Shorter release cycles apparently means that Firefox won't be tested much anymore, as they are bypassing the lengthy testing process and releasing numerous small changes without testing them (sufficiently) first, like Microsoft. And even Microsoft can't pull it off, he says.
I find his argument to be rather weak.
First of all, this is certainly not the first version of Firefox which has been delayed. Indeed, delays are very common in the world of software. If delays were a recipe for disaster, I don't think there would be any software companies left! And how did Firefox grow to become the dominant "alternative browser" if delays automatically mean disaster?
I also don't see the logic in the claim that they will be in trouble after restructuring the development process. If Mozilla identified ways they could improve their development process in a way which leads to higher quality, how does that mean that they are in trouble now, as opposed to before the restructuring? Can he predict the future? Does changing something to address a problem automatically mean that you are in even more trouble after addressing any problems you identified earlier? I would have thought that being able to adapt to new situations and changing things when necessary is a positive sign.
And how does shorter release cycles with fewer changes mean that they won't be able to test Firefox anymore? Did they announce that there would be no more nightly Firefox builds? Are they going to prevent the community from testing Firefox while it's still work in progress? That seems to be the assumption since he mentions Microsoft's security patches.
Of course Mozilla isn't going to stop releasing test builds to the public. And even if there is less time to test things, won't fewer changes mean that there are fewer things to test, which means that testing will take a shorter amount of time anyway?
Furthermore, if shorter release cycles are such a disaster, then why isn't he also claiming that Chrome is in trouble? It seems to me that Mozilla's new development process is similar to Chrome's in many ways.
After all this speculation about Firefox's new release process, Mr. Kennedy claims that there will be no room for Firefox in the market in the future. Google and Microsoft will squeeze Firefox from above and below, he says.
Yes, I'm sure the competition will heat up, but so what? Why wouldn't there be room for several browsers in the market? There are hundreds of millions of people online, and even if only a small portion of those chose Firefox, Mozilla would still be able to make a comfortable living for themselves.
He also seems to forget about Mozilla being a non-profit organization. That means that a lot of people are likely to donate simply because they like what Mozilla is doing. In fact, they have this whole movement thing going for them.
In conclusion, I don't think Firefox is doomed at all. It might see some of its market share eroded to a certain degree by other browsers (or it might continue to steal users from IE), but I think it will definitely stick around.
In fact, I really hope Firefox will stick around and continue to be a force to be reckoned with on the Web. Recent developments in HTML5 VIDEO shows that we need someone else to help push for a free and open Web, rather than a Web based on closed, proprietary and patent-encumbered solutions. More browsers in the market also helps prevent a Web monoculture. We need more browsers out there, not fewer of them.
If Firefox is indeed doomed, then that's a big loss for the open Web. But I don't think that's the case at all, and Mr. Kennedy certainly wasn't able to convince me that this is the case.
Readers' comments 23
Left under the original article, as archived.
About this article
| Published in | A Journal from Behind the Trenches |
|---|---|
| Date | January 22, 2010 |
| Author | Haavard |
| Tags | infoworld, mozilla, firefox, development, open standards, browsers |
| Original | Internet Archive |
StaretzFriday, January 22, 2010 4:18:17 PM
Good job, Haavard.
Opera will flourish because of folks like you.
karolpFriday, January 22, 2010 5:26:23 PM
endless loveFriday, January 22, 2010 5:34:07 PM
tomassplatchFriday, January 22, 2010 7:47:16 PM
http://www.osnews.com/story/22764/Why_Firefox_s_Future_Lies_in_Google_s_Hands
not those nonsense reasons...
TeoFriday, January 22, 2010 7:58:50 PM
PurdiFriday, January 22, 2010 8:25:38 PM
Originally posted by tomassplatch:
And you think other search engines wouldn't be more than willing to pay Firefox to be the default search? Maybe they would make less money, but they would still be making lots of it!
TeoFriday, January 22, 2010 9:37:42 PM
Originally posted by Purdi:
There are handful of search engines that matter: Google, Microsoft, Yahoo-soon-to-be-microsoft, Rindex.ru, Baidu.
Given that many of the most vocal Firefox users have political agenda (free, open source, open speech), making deals with MS or Baidu may be suicidal to Firefox.
Today with great surprise I found that an entire department of the company I work for has switched to Chrome. The company does not have a policy what browser to use - so they did it just because they like it. They switched mostly from IE7/8 but some of them are former Firefox users.
Cutting SpoonFriday, January 22, 2010 9:56:25 PM
Hurr Hurr, looks like someone was miffed by all the speculation. I don't really buy Mozilla's claims in the release notice, any more than I buy this guy's claims that Firefox will implode and eat the sun on its way out.
Much more likely: In the near-term, Mozilla Foundation's repositories need to be examined and cleaned. They need better project management and source management, better cooperation between salaried employees and volunteers, and a diversified revenue stream.
Mozilla Corporation is a for-profit company, as opposed to Mozilla Foundation. MoCo derives too much of its current income from one Google deal concerning Firefox. If the contract is not renewed when next it expires, MoCo will be crippled and may even be purchased.
So, yes I may go so far as declaring that I despise Firefox and hope it will fail; but there will be hope if they can arrest this freefalling open-source fragmentation of the code base.
As I've stated before, Open Source is not Open Method. And it requires the most powerful and responsible leadership to avoid cancer and collapse.
Charles SchlossSaturday, January 23, 2010 2:21:10 AM
+1
walterSaturday, January 23, 2010 3:40:34 PM
thats it
cousin333Saturday, January 23, 2010 5:37:37 PM
Although I do think, that Firefox's golden days are over. But - just like Opera which seems to be back in competition with 10.5 again - they can also give themselves a boost.
On the contrary maybe sticking to their current development cycles would doom them.
Cutting SpoonSunday, January 24, 2010 2:29:35 AM
Or try wandering through Bugzilla - a major reason why I prefer that Opera's developer's maintain their privacy. The development of new features begins with:
"So-and-so did 'X', can we do this?"
"I dunno, give me a weekend and if it looks good we'll try it."
And that's why the new features in Firefox 3.5 exposed so many existing performance issues, which they've been frantically patching ever since. Opera's recent leaps in automated testing are great, and they have most likely aided in tuning Carakan's JIT and Vega's internal render structures. I've heard that 'Table' and certain other plain HTML elements are currently slower in Presto 2.5, yet granted the release comments this is due to re-optimization of older fixes.
And that's what Firefox needs: possibly a near-total development freeze, and a major consolidation of individual components. The Download Window? Still buggy, still separate, still doesn't warn when a download fails. Mozilla needs to know what it has before further improvements will be possible.
Aaron HayesThursday, January 28, 2010 12:37:54 AM
power does not come from small Northern European Companies, American Companies, Search Engine Giants, or Steve Jobs fat new oversized iPhone that doesnt make calls... it comes from what the marketers call the "enduser". They are the ones using an item and creating uses for it. The reason for Software patents that stifle innovation and create oppressive monopolies is to keep the average user from being anything other than an obedient consumer, and realizing that THEY call the shots. If patent trolls like the MPEG-LA operating under the guise of legitimate business are insisting on holding the world hostage for the sake of their pocket books, then they are opening the door wide for opportunity. Dont believe me? Ask the RIAA. We cant let greed destroy freedom.
Cutting SpoonThursday, January 28, 2010 2:11:58 AM
IlgazThursday, January 28, 2010 11:51:30 AM
Cutting SpoonFriday, January 29, 2010 7:32:40 AM
I had the same question about Opera UI a while back. Menu text and skins use common file formats. Most of the rest is compiled into machine code.
Comments in plain, readable, editable text like "NOTE: This function is vulnerable and still needs to be fixed." should not go into the public release. Also, being able to replace or modify the contents of browser.jar is a huge risk.
PurdiFriday, January 29, 2010 11:14:57 AM
Originally posted by hellspork:
Why? A release is a compiled binary. Why would they waste time removing comments in the source?
What question? The application is compiled. The settings are plain text.
Cutting SpoonMonday, February 1, 2010 4:34:36 AM
2nd foolishness 1st: I once HAD such a question, the REST of what I wrote is the ANSWER I received. Okay? YES, most of OPERA is compiled to machine code. Settings and localization are text. We are saying the same thing.
Addressing the first matter: I believe that Opera's compiled binaries do not contain source comments, they are automatically removed in the build process. A few settings files in Opera (such as browser.js) contain HELPful comments; for that matter browser.js is SIGNED and cannot be MODIFIED casually. This is a good balance. However I was complaining about Firefox. Much of that browser's logic is written in a hybrid scripting syntax, stored as plain text inside a .jar file (named browser.JAR, not browser.JS). Comments could be stripped from the RELEASE copy with a single command-line argument, reducing file size and security risks. The sources and dev releases should retain all comments and disabled code. I am unimpressed by the etiquette and security model of Firefox development, is this clear now?
PurdiMonday, February 1, 2010 10:10:24 AM
Originally posted by hellspork:
Uh, yes, that's what a compiler does. it's designed to ignore comments.
Pointless. It won't be noticeably smaller, and there's no security risk.
Cutting SpoonTuesday, February 2, 2010 6:59:19 PM
Originally posted by Purdi:
It's done routinely on third-party issues of Firefox, reduces the baseline size of the program by about half. Since Mozilla could save roughly 1/3 the bandwidth of each download, it'd be a significant cost savings. On security, visit BugZilla yourself or check the recent profile of web attacks.
Saurabh KumarMonday, February 22, 2010 7:44:46 PM
http://www.infoworld.com/t/business/unfortunate-ending-357
Edit: ZDNet has done a investigation into the issue.
http://blogs.zdnet.com/BTL/?p=31024
HaavardMonday, February 22, 2010 8:11:03 PM
Well.
beritakuWednesday, April 14, 2010 2:48:57 AM