I highly encourage all of you to go read Mitchell's post, linked to in my previous post, and when you're done with that, head over to Red Hat and read the interview with Chris Blizzard. Chris is one of our long-time unix hackers, an employee of Red Hat, one of drivers@mozilla.org (with me and a dozen others) and on the Mozilla Foundation board of directors.
Here's an excerpt (but you really want to go read the whole thing):
RHM: So Firefox is largely from the same code base?Posted by asa at January 21, 2005 07:01 PMCB: Parts of Firefox are largely the same, yes. The core rendering engine for webpages is the same, as well as the underlying technologies that deal with networking and most web features. Where there is divergence is how certain features of the web (like popups or webpages being able to resize windows) are handled. Also, the UI has been rebuilt from the ground up with a focus on ease of use.
RHM: Is that why the browser project was renamed to Firefox?
CB: I think there was a lot of pressure to make sure that people understood that there was a pretty clear difference between what Mozilla was and what Firefox had become. It's not just a fresh name on an old face, but a new proposition to our users that states what we believe in and how good we feel the web can be with the right technology.
RHM: What was Mozilla then, in your eyes, and why was this "new proposition" necessary?
CB: In reality, Mozilla was a reincarnation of the old Netscape 4.x suite. It hadn't changed much in a few years, and many people considered it a mess from a user interface perspective. The Firefox project (originally named the Phoenix project) started as an effort to get rid of all of the things in the user interface that were probably not needed by 99% of end users.
It was also seen as ground to experiment with technology and try to innovate in the browser space, something that hasn't really happened in a long time. In the end, I think that the Firefox project was needed to prove that we could still make positive changes on the web.
RHM: Give us an example of where Firefox has served as that "ground to experiment."
CB: Firefox's popup blocking was derived from previous code found in the Mozilla project, but only in Firefox was it possible to make the changes required to turn it on by default and refine it to the point where it works as well as it does today.
It's also important to note that Firefox as a browser is also a platform for a huge list of "extensions." These extensions have two purposes. One, power users can still add all the functionality they want without requiring everyone to have the same feature set. Two, it means that developer's experiments can be tried out. From time to time an extension will be pulled back into the project and made part of the product.
The extensions system in Firefox is largely based on the one that was available in Mozilla, but it's easier to use and has tools to manipulate those extensions. What made it possible to install or not install the Mail component of the old Mozilla browser is largely the same technology that is used to install extensions in Firefox today. It's just much better refined.
RHM: What are Red Hat's plans for integration of the Firefox browser?
CB: We know that Firefox is one of the most successful open source projects to date, and we also know that the Firefox people take the user experience seriously. We will continue to bundle Firefox for our users because we understand that a great user experience is crucial to our own success as well.
I was sort of dissapointed from the article. Mostly because I saw the questions as poorly worded, and responses small due to them being about stuff he didn't deal with.
I would love to see an indepth story about:
What makes Firefox different besides the UI
Why hope in Mozilla Suite was later put into Firefox
At what point the Mozilla development moved twards making an end user product.
I know there are simple quick answers to this, I just haven't had the motivation to corner people and get the information.
Posted by: larfnarf on January 22, 2005 12:30 AMNo, I think they were good questions, and Chris didn't seem to understand the last one. The irony of a product made to 'get rid of all of the things in the user interface that were probably not needed by 99% of end users' and which is promptly stuffed with tens of extensions after installation is bought out nicely too.
W.
Posted by: Wally on January 22, 2005 08:00 AMAny examples of extensions that were "pulled back into the project"?
Posted by: aldiboronti on January 22, 2005 10:43 AMlarfnarf, those sound like good questions for Ask Asa. I suspect that I have sufficiently accurate information for most of what you're wondering.
aldiboronti, depending on how far back you want to go, tabbed browsing was first implemented as an extension. More recently, the extension manager was first implemented as an extension. I believe that sidebar bookmarks were first implemented as an extension as well. Per-site styling started out as an extension and was recently incorporated into Gecko.
Lots of features or sub-features started life as extensions or other hacks to the browser and that functionality was eventually built into our apps. With Firefox, we've been a lot more conservative about what we incorporate into the app because we finally have a much better extension management mechanism and an improved versioning and update system.
I can imagine other great extensions like the Flash Click to View feature or Adblock features eventually finding their way into Firefox in some form.
--Asa
Posted by: Asa Dotzler on January 22, 2005 12:02 PMThanks Asa.
I posted that comment in a fit of mini rage. I appreciate your positive response to my post, and also praise you for still having a blog that allows for comments.
Thanks.
Posted by: larfnarf on January 22, 2005 01:20 PM