ExtremeTech Preview of Phoenix
Monday December 30th, 2002
ExtremeTech has a glowing preview of Phoenix. The article, which appears to be based on the 0.5 release, highlights Phoenix's small size and uncluttered interface. Tabbed browsing, multiple home pages and the search bar are discussed as well. They also included a screenshot of MozillaZine, which makes us like this review even more.
Thanks to the_village_treky for the news.
Printer friendly version: <http://www.extremetech.co…cle/0,3998,a=35184,00.asp>
Also, too bad they didn't grab a nightly build to show off the new theme.
I'm using Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.3a) Gecko/20021224 Phoenix/0.5 which is great except for the tooltip bug: <http://bugzilla.mozilla.org/show_bug.cgi?id=185965>
> Also, too bad they didn't grab a nightly build to show off the new theme.
this new theme sucks, at least in Linux. I really liked the previous theme and now it is gone... the orbit theme on the theme sites is really different, I want the other theme. Was it too difficult to have / maintain two themes in a release?? just this new, ugly, theme as the default (since so many people like it) and the previous theme as classic or something.
"Was it too difficult to have / maintain two themes in a release??"
Probably not, but it takes up extra space and the whole point about phoenix is that it isn't bloated.
#15 Re: I partly agree ...
Tuesday December 31st, 2002 1:10 PM
We plan on offering the old Orbit theme shortly.
Interesting, another one mixing up features with, I cite: "junk". I am sick and tired really of people doing this. If they do not like more than very basic features, fine. Phoenix will give them that and everybody is happy. No need though to bash Mozilla for what makes it the preferred browser for many other users: its advanced features and configurability. Many people prefer Mozilla because it has the other apps (Mail, IRC, calender) included or addable, because it has *many* more useful features and because empowers them to do the things they want more efficiently. I have never seen a Mozilla user bash Phoenix for its lack of features and flexibilty, why do Phoenix aficionados insist on bashing Mozilla and calling those features they might not need or understand "junk" or "bloat"?
#3 Toolbar grippies are gone...
Tuesday December 31st, 2002 3:22 AM
I think the problem with Mozilla isn't the features, but rather some design decisions that made the extra features expensive in terms of performance and user interface clutter. This is a criticism of Moz, but not of the developers. It's almost impossible to get everything right at the beginning of a project. The fact that projects like Phoenix and the calendar can emerge based on Mozilla is a testament to all the right decisions that the Mozilla architects did make.
The value in Phoenix is that it points the way to what can be done with the mainstream Mozilla. Divide it up into separate processes. Simplify the user interface. Keep the features, but not the bloat.
Hmm. Sorry about the title of the last message. Must remember to edit it by hand before I post.
#8 Mozilla as THE development platform
Tuesday December 31st, 2002 9:30 AM
Mozilla guys have always said that Mozilla is the develoment platform from which technologies are developed and others grab to produce their own product. But Mozilla is so good that it can be used by end users. Let Mozilla keep developing the technologies and incorporate them into their suite. As many have reported, I see very little difference in performance between Mozilla and Phoenix (subejctive feeling). As computers become more powerful and with more standard memory, the peromance issue becomes a moot point (you get diminshing returns as you go up in the scale).
I am sure Mozilla will incorporate fixes and features from offshoots like Phoenix and others. Toolbar configurability is an example of one technology that can be ported back (codewise or ideawise) to Mozilla the same way that tabbed browsing was developed after Multizilla and Opera showed how neat the feature was.
Lots of people criticise so called bloat but as the Office suite wars of the early to mid nineties showed, "more" gets more attention than "less" because my 10% share of features I use is not the same 10% of the features others use, yet most users want to have something in common. What is "bloat" to some is "feature rich" to others,
My 2 cents
#9 Mozilla-the-platform vs. Mozilla-the-browser
Tuesday December 31st, 2002 9:44 AM
> Mozilla guys have always said that Mozilla is the develoment platform from which > technologies are developed and others grab to produce their own product. > But Mozilla is so good that it can be used by end users.
You're confusing Mozilla-the-platform with Mozilla-the-browser. Take a look at <http://books.mozdev.org/index.html> for more information about the distinction between the two.
I reaffirm what i mentioned previously. I also add that Phoenix is a subset of Mozilla with optimizations and features that are particular and relevant to it's design goals (a browser). In no way Phoenix takes away the importance and relevance of Mozilla (THE development suite).
2 additional cents .. make it four cents now ;-)
#20 Mozilla-the-browser is not a platform
Tuesday December 31st, 2002 4:13 PM
You're still confused... but now I see that you don't really understand what the Mozilla framework really is. It's not the browser at all -- it can be used to write all kinds of non-browser software. The browser called Mozilla is one application that is built with the Mozilla framework.
It is confusing, because the term "Mozilla" is used to describe two different things. Read further through that link I gave you, and you'll see what I'm talking about. There's a whole book about creating applications with Mozilla, and at least up to chapter three, there is no browser described at all!
Bloat happens. Time is too valuable to spend it reworking code for efficiency; if it works you keep it, if you're Microsoft you keep it, regardless.
The biggest contribution to code bloat is object oriented programming languages. Objects carry baggage, lots of baggage. Still, who wants to code in assembler?
The Phoenix project appears to tackle the bloat problem by modularizing functionality. Install the components you need and run light, but realize that if you install all the features you will be back where you started (it's kind of like a diet, that way).
True, a better design under the hood of Mozilla would be good. I really do not buy that myth about performance: even though i have an old computer, I do not feel much of a difference between Mozilla and Phoenix. But a cleaner design will certainly help with maintainability and future extendability and that indeed is very important. Note that from a performance perspective, it is not very relevant whether or not mail is part of the main app or a distinct app. Both approaches have advantages and disadvantages, but the main issue gain is maintainability.
I will gladly use any browser that gives me the features of Mozilla (well, there is still a lot missing too) packed in a better UI combined with better performance and maintainabilty, but Phoenix is still quite far from even coming close to this.
And I really did not have the impression that the Phoenix developers even want this to happen: there are people who dont want the preferences UI to allow them to configure everything, who want only a short context menu to the expense of certain power features and so on. If Phoenix is the browser that makes those people happy, I am all for it - at least they will not try to get useful features out of Mozilla. But they should not bash Mozilla for keeping what many users desire.
So lets finally get all those nice features of tabbrowser, multizilla, mozgest, enigmail and other into the Mozilla trunk! :)
It's hard to say on the performance issue. Certainly Phoenix feels snappier to me, but I have no idea why. Maybe some subjective thing about how the menus redraw or some such.
Agreed that Phoenix doesn't have all the features of Moz, and I don't see Phoenix morphing into Mozilla either. My point is that the successful elements of Phoenix can be applied to Mozilla without necessarily eliminating features. Separate the apps. Hide features by default that few people use, but provide a way for the user to enable them. Steal parts of the GUI that Phoenix has done particularly well.
Tuesday December 31st, 2002 10:25 AM
Pheonix has InitialPaintTimeout set to 250ms, while Mozilla has it at 1200ms. I think this is going to be changed in 1.3b. This may affect perceived performance a lot.
#14 Re: I partly agree ...
Tuesday December 31st, 2002 1:09 PM
Phoenix is faster because its XUL is written much more efficiently than that of Mozilla. It's really that simple. I guess this is a point we didn't explain well enough, because I often see Mozilla users try to claim that Phoenix is only faster because it's "missing features", and that a browser-only install of Mozilla is just as fast. This is completely incorrect. Mozilla's XUL is an absolute mess and that slows it down considerably.
"Regardless of whether you discern a difference on your machine, there is indeed a vast performance difference; it is not a myth, as you say. There is nothing to "buy." It has been quantified time and again with the same performance tests used to measure Mozilla." I don't think anyone is sayin that Phoenx speed inprovement are mythical. Speed perception (as opposed to quantified one) is also an important quality. If in my machine my perceived reactions is that boith have similar performance, the is logical for me to saty with the product I get more from. No need to take it personal. Second, as a statistian, I have come to distrust many quantifiable measurements. Don't forget that the difference between 1 and 2 is 100% but no one would say that the diffrence between and millio and 2 millions (which i also 100%) are the same, eventhough both are of the same magnitude. So imagine what happens when measuring things at the millisecoind level. How the speed inprovement is perceived by the user is also important, irrespective of quantified data. Where I worked, we have tested both Mozilla and Phoenix and at least with the PC we are used, we don't see any difference in performance. We do like a lot the configurability of Phoenix toolbars. Keep up with the godd work
#18 Re: Sorry Blake
Tuesday December 31st, 2002 3:07 PM
We do not only go by percentages (which, I agree, can be misleading) but by the tangible difference in time. We also have plenty of qualitative accounts. If Mozilla is already fast for you then you likely won't notice any difference, but I'm surprised if Mozilla is slow and Phoenix is equally slow.
#13 Re: I partly agree ...
Tuesday December 31st, 2002 1:06 PM
Regardless of whether you discern a difference on your machine, there is indeed a vast performance difference; it is not a myth, as you say. There is nothing to "buy." It has been quantified time and again with the same performance tests used to measure Mozilla.
Second, please post the features that you think Phoenix is missing (from Mozilla) and we will gladly consider including them. People are constantly complaining about this, and then I say "well, ok: what features?" and then they choke and disappear. Many people don't seem to realize that Phoenix does contain most of Mozilla's features, either built-in or available via extensions, but the trimmed-down interface makes people think that Phoenix has almost nothing. It seems very psychological.
Third, this review, while kind to Phoenix, is absolute junk. The same author who praises Phoenix for its light interface and lack of bloat goes on to praise Opera in the article's discussion board...
Phoenix is very nice. I'm still using the 0.5 release.
You asked, so here are some features that Phoenix should have that Mozilla has: Password Manager, ability to clear location bar history, ability to show only autocomplete matches I type, ability to remove download manager entries from downloads.rdf, help files, option to disable form manager, compare page in cache to page on network only once per session, and HTTP pipelining.
Assuming that users will figure out how to do any of these with user.js is not consistent with aiming for a user base of non-experts. There needs to be UI for more preferences.
And Talkback agent. Phoenix is really stable, but it's frustrating the 2 or 3 times it's crashed because I feel like I can't do anything to help find that bug.
#22 Re: features Phoenix needs
Tuesday December 31st, 2002 7:34 PM
Phoenix has password manager. You can clear the location bar history by just clearing history (we considered a separate option, like in Mozilla, but that seemed rather pointless. We'd love to hear arguments why it's useful). You can remove download manager entries...just select them in the list and hit delete (or right click and choose from the context menu). Help files, option to disable form prefill are coming.
#25 Re: Re: features Phoenix needs
Thursday January 2nd, 2003 6:55 AM
Concerning an argument for a separate Clear Location Bar and Clear History option... I like being able to clear it with one click, but the problem I'm having is that I can't clear the Location Bar without clearing History, and in doing so my Favicons are removed. If there's a way to keep the Favicons from disappearing everytime I clear the History, I'd love to know about it.
Also... I've been looking at designing a new skin for Phoenix, but I haven't found any simple ways to do so. Any suggestions on where to look?