File:  [mozdev] / books / www / articles / bloom.html
Revision 1.10: download - view: text, annotated - select for diffs - revision graph
Thu Aug 29 21:05:20 2002 UTC (17 years, 6 months ago) by david
Branches: MAIN
CVS tags: HEAD
updating draft

<h1>Let a Hundred Browsers Bloom</h1>


<p>The recently released <a href="http://wp.netscape.com/eng/mozilla/ns7/relnotes/7.html">Netscape 7</a> may be the most well known browser built with Mozilla, but it certainly is not the only one.  Mozilla is being used as a framework to create many different types of applications including OEone's <a href="http://oeone.com/products/desktop.html">HomeBase DESKTOP</a>, ActiveState's <a href="http://activestate.com/Products/Komodo/">Komodo IDE</a>, and <a href="http://www.mozdev.org/projects.html">all of the projects</a> hosted on mozdev.org.  People are also using Mozilla to create their own custom browsers [link to brian's article].


<h2>Why Do We Need More Than One?</h2>

<p>One of the benefits of Open Source development is that it prevents someone from having to reinvent the wheel whenever they are working on developing something that has been done before.  Since the Mozilla community is already working on a browser, wouldn't it be better if everyone just focused on making that browser as good as it can be?

<p>Instead of being a bad thing, the several different browser development projects that are currently underway are one of the Mozilla community's greatest assets for the simple reason that one browser can not be all things to all people.  Each new browser that gets built is filling a need that is not being met by any other existing option.  Each new browser that is built also has the potential to appeal to a whole new audience that will help expand Mozilla's adoption.

<p>Another positive benefit of having multiple browsers is that it helps avoid compromises that don't make anyone happy.  AOL is interesting in Mozilla because they want a browser that appeals to novice Internet users.  The Mozilla developers who contribute their time to the project want to create a powerful browser with a collection of advanced features.  If the community is locked into working on only one browser, then the end result of this development process will be a browser that has a bewildering array of features and that doesn't appeal to either intended audience.

<p>If one browser can't possibly to beginning users and power users at the same time, why not create two different browsers?  For that matter, why not create as many different browsers as there are different types of users?  Since all of these browsers are built using Mozilla, web developers can create sites that work with Mozilla and users can browse with whatever tool suits them best.  Everyone wins.


<h2>Gecko Based Browsers</h2>

<p>There are two main types of browsers that are built using Mozilla.  Some developers choose to create their application using XUL, Mozilla's XML-based User Interface language.  Other developers prefer to use just Gecko, Mozilla's rendering, and then create the GUI of their browser using one of the toolkits native to a specific platform.  There are Gecko based browsers for each of the major operating systems in use today, including Windows, Linux and Mac OS X.

<p>*** need screenshot *** The goal of the <a href="http://chimera.mozdev.org">Chimera</a> project is to create a best-of-breed browser for the Mac OS X platform with an user-interface that is as simple and as clean as possible.  The most recent stable release, <a href="ftp://ftp.mozilla.org/pub/chimera/releases/chimera-0.4.dmg.gz">Chimera 0.4</a>, is available for download along with <a href="ftp://ftp.mozilla.org/pub/chimera/nightly/latest-trunk/">nightly development builds</a>.

<p>Chimera uses <a href="http://www.mozilla.org/ports/fizzilla/Cocoazilla.html">Cocoazilla</a>, a variant of <a href="http://www.mozilla.org/ports/fizzilla/">Fizzilla</a> that consists of a UNIX back end connected to a Cocoa front end.  Since Chimera uses a native toolkit to create it's GUI it can't run on any platform other than OS X, but since it doesn't use XUL it is faster than the default Mozilla browser on the same computer (applications written with XUL will always be slightly slower than applications written with native toolkits because *** need explanation here, ask brian ***).  Reference to screenshot...

<p><a href="http://galeon.sourceforge.net">Galeon</a> and <a href="http://kmeleon.sourceforge.net">K-Meleon</a> are projects that also have the goal of creating a simple standards-compliant browser using Mozilla's rendering engine.  Galeon uses Gecko to create a browser for the <a href="http://www.gnome.org/">GNOME</a> desktop and K-Meleon uses Gecko to create a Windows only browser.  The latest stable version of Galeon can be <a href="http://galeon.sourceforge.net/download/">downloaded</a> for a variety of Linux distributions.  There are also alpha versions available for <a href="http://galeon.sourceforge.net/galeon2/">Galeon2</a>, which is a new major version of the browser that takes advantage of the huge changes in architecture in the new GNOME 2 desktop.  K-Meleon download is at...

<pre>
SkipStone               http://www.muhri.net/skipstone
Q.Bati                  http://qbati2.sourceforge.net
</pre>


<h2>XUL Based Browsers</h2>

<p>A few of these projects are working on improving the basic Mozilla browser interface. One of the first Mozilla applications, Aphrodite
(http://aphrodite.mozdev.org), has an interface designed and created by members of the Mozilla community. Another project called m/b (short for
mozilla/browser) addresses shortcomings that some people see with the default browser interface. In many ways, these projects are similar to
themes, but they take this idea one step further and use XUL, CSS, and JavaScript to change the layout of the browser interface and not just the
look of the browser.

<pre>
Aphrodite               http://aphrodite.mozdev.org
	include aphrodite themes
skyline                 http://bugzilla.mozilla.org/show_bug.cgi?id=154414
m/b                     http://www.blakeross.com/images/mb (screenshot)   
        phoenix                 http://bugzilla.mozilla.org/show_bug.cgi?id=161041
        project piglet          http://mb.mozdev.org
</pre>


<h2>Conclusion</h2>

<pre>
	- links to other browsers
	- minotaur reference???
	- let 100 browsers bloom
</pre>

<p>If you think of a type of browser that is needed but that is not being developed right now, use Mozilla to create your own browser.


-------------------------------------------------


<pre>
notes

others

Beonex Communicator	http://beonex.com/communicator/
BrowserG!		http://browserg.mozdev.org
Dino			http://dino.mozdev.org
more			http://www.mozdev.org/categories/browsers.html




aol and compuserve clients	gecko based or mozilla based?


minotaur  - alternative mail client
	phoenix reference on the page
	http://www.mozilla.org/mailnews/minotaur/



http://www.deftone.com/blogzilla/archives/some_major_mozilla_ui_changes.html#comments

Lot's of changes are under way for the Mozilla browser


- The "Skyline" project, which is a version of Mozilla is being worked on internally at Netscape, "is a short-term project to build a working 
prototype of a new web client that provides for the daily needs of novice-intermediate users, is fun to use, and really shines in a broadband 
environment"

- Project "Phoenix", "...a redesign of the Mozilla browser component, similar to Galeon, K-Meleon and Chimera, but written using the XUL user 
interface language and designed to be cross-platform" (via mpt's site)

- The mozilla/browser project, mentioned a while ago on this site has resurfaced on mozdev. There are some screenshots, but the installable 
XPI is not working, yet.


http://www.mozillazine.org/weblogs/asa/2002_08_01_asadot_archive.html#80090450
http://www.blakeross.com/archives/2002_08_11_index.html#80310485
http://www.deftone.com/blogzilla/archives/some_major_mozilla_ui_changes.html
http://www.mozilla.org/mailnews/minotaur/index.html (stand-alone mail client)



-------------------------------

reasons for creating alternate browsers

from http://www.mozillazine.org/weblogs/hyatt/2002_04_07_mozillian_archive.html#75307435

In some cases features that are of more obvious benefit to one app end up affecting the design of all the apps. For example, profiles are of far 
more use to mail than they are to a browser. Are profiles even really necessary if you were designing just a browser? In this era of Windows XP and 
Mac OS X, both of which have fast friendly user switching, I'd claim that they aren't.

Another example of horrid user interface is the Mozilla Preferences dialog. Again, all prefs are shoved down into the hierarchy by an extra level, 
because the top level must necessarily be given over to individual application branches. User interface that logically belonged in preferences has 
been torn out because it would have ended up too buried to be useful, e.g., the Mail/News account settings.

from http://www.mozillazine.org/weblogs/hyatt/2002_04_07_mozillian_archive.html#75279564



The end result of these colliding forces is Mozilla 1.0, a lumbering beast of an application suite that boasts a bewildering array of features
supplied by contributors whose respective agendas are inevitably at odds.  What's even worse is that none of the warring factions is happy with the
end result. Mozilla ends up with geek features like the Links Toolbar that make Netscape unhappy, and Mozilla ends up with a lousy toolbar design
in order to keep Netscape happy. [...] You'll never see eye-to-eye, because you're not trying to produce the same browser.

The only real solution I see to this problem is for Mozilla to escape the one-browser mold. After Mozilla 1.0, Mozilla.org should relinquish
control of its flagship application to Netscape and strike out on its own with new browser projects that are truly innovative. Let one group of
people try to design the power user's dream browser, while another works on a simpler elegant user experience.


</pre>

FreeBSD-CVSweb <freebsd-cvsweb@FreeBSD.org>