Slashdot Mirror
Mozilla's Major New Roadmap
kerz writes "mozilla.org today released a new version of it's famed roadmap, this time with some pretty major changes. First and foremost, they plan on ditching the large Mozilla suite in favor of Phoenix and Minotaur. Secondly, they have plans to change the milestone cycle to allow for more time to fix the Gecko layout engine to be smaller and more efficient. MozillaZine has the scoop..."
Wonder if this will help mozilla's memory footprint? Even for people like me who like to have the mail and browser open at the same time (hopefully they won't each take up X-ram)
Ad Majorem Dei Gloriam
Interested in AI? MACR
Switching to this model will mean Pheonix is directly competing with Camino on Mac OS X, how could they possibly beat a Mac OS X native attempt?
They're finally going to support Phoenix on OSX!
This is a big win for the Mac community imho. Camino is great, but there are barely enough developers to cover the front end, the main body of the Mozilla project being behind a cross platform Phoenix project is a Good Thing?.
"The worst tyrannies were the ones where a governance required its own logic on every embedded node." - Vernor Vinge
I love Mozilla - it's my primary browser, gotta love pop-up-blocking! What I wish they would focus more energy on though is the mail client. I primarily use Netscape Messenger (netscape 4.79) for mail, and I know a lot of other people that do as well. The reason I can't/won't use Mozilla for mail yet is bugs. Basic bugs too - things I reported over 2 years ago, and they still aren't fixed yet. What kind of bugs am I talking about? For example, when you switch between IMAP mail servers, netscape messenger used to remember the last selected message from one mailbox to the other. Mozilla has never done this, but I keep getting updates that this bug is being worked on, or passed on to the next person. The other major bug I notice is that when I type in nicknames in the To and CC fields - 50% of the time, they get translated into the right email addresses, but other times they don't. My other major gripe about mozilla mail is the lack of an option to send just plain old plain text messages again. I don't want the headers of replies and forwards being turned into little graphics. I don't want symbols like ;) being turned into little smiley faces. I want to type in courier just like I can in Pine, or netscape messenger. I think more options with mozilla mail would make a lot of people happy...
I find it facinating that it at least appears that Mozilla is leading Netscape rather than the other way around. But I am left wondering how this will fit in with Netscape's future strategy. Will they continue with tradition and continue to release an all in one Internet suite, or will they begin to follow Mozila's path?
-- Kircle
Or perhaps this is just a way of disposing of the outstanding Mozilla bugs that no one is willing to fix? Just start a new product instead?
sPh
I have not used mozilla browser ever since phoenix 0.5. And I have been using the phoenix nightly builds.
I use phoenix on linux/windows/solaris, I haven't restarted phoenix on my solaris box for days/weeks. Its fast, sleek, and has a very small memory foot print as compared to the lizzard.
Some of my concerns with phoenix though are
for the last time people, I am "frodo from middle eaRTH", not "middle eaST".
Is this an attempt to battle the smaller, faster KHTML engine that Apple picked? Sounds like the Mozilla gang is a more miffed than previously believed.
If I had something intelligent to say, I would have said it.
I just wish they'd also separate out Mozilla Composer and make the basic no frills standalone HTML editor the world needs.
The Mozilla team really need to break Mozilla up into smaller, more focused parts. That is one area I will give Microsoft credit for - they made IE and Outlook seperate programs.
The ideal for Mozilla would be (IMHO) a browser, a mail client, a download client, an IM client, and a composer. Each should be replacable - I should be able to tie the browser into whatever download agent I want, have whatever email client I want be pulled up when I click on a mailto: link, etc.
I'd even go so far as to have a caching program that the browser and downloader could talk to (to unify the disk cache system), but then I already run Squid on my systems.
Of course, all the Moz bits could and should access the same DLLs (.so's) to keep the disk and memory footprint down.
www.eFax.com are spammers
Especially when the system (Mozilla here) is completely customizable, and side projects are indeed developing quickly.
Why bloat is good, you may ask? The machines are reasonably fast these days, and users (ahem, let me say, just I) tend to spend significant time on the browser window, and mail window (and calendar, may be composer, etc). There is a common interface, one can upgrade everything at one shot (kinda like Redhat 8 and 9... you get everything updated). Makes life a little easier.
I would have complained had M$FT done the same thing. Why? because there would have been no choice then. With Mozilla, someone can customize and contribute a build that does exactly what a group of people need. Not so with the big company from Redmond.
Imagine being able to have a bunch of utilities (from mozdev.org) tightly integrated into one "application suite" (with options for people to run Phoenix, etc.) -- it is almost like having the cake and eating it too.
S
These are not buzzwords. 'Strawman' is a common name for a typical logical fallacy -- quite CS and unbuzzwordly for me. For a great explanation of the difference between buzzwords and terms which serve a purpose look at this list, from tunes.org:
These are not buzzwords. Now, for comparison, here's a bunch of buzzwords:
The filesystem is the package manager
Hm, I wonder to what extent this new roadmap is a reaction to Apple's decision to use khtml instead of mozilla as the basis for safari.
...maintainable, performant, and extensible...
I have to wonder what "performant" means. It is proof that marketing types were involved in the writing of this roadmap document.
Phoenix is a fabulous browser component. I like Mozilla, but I think that this is just what they need. They can start by skimming off each peice and the users can integrate each part of the suite at their own discretion. Mail programs can be built off of Mozilla's XUL interface, as can chat programs, etc. It's ideal for UNIX machines where libraries are most often shared. Phoenix is fast and stable. It's the future of the Mozilla browser, and I'm glad that they've made this decision. Why reinvent the wheel when you can just improve it. Mozilla is getting to be pretty good on its own, but still isn't nearly as practical as it *could* potentially be. Phoenix takes Mozilla and really strips down the crud; It even implements cool, new features along the way.
Reading TBL's autobiography, I thought this idea of bundling the editor and the browser somewhat of a stretch.
I think he had in mind a world in which anybody could write and publish "annotations" to every web page (I think the old mosaic tried to support this). Right now there are several "server-side" applications that let you do this (blogs, wikis, etc...). Most people don't need or want an html editor, and with modern technology an html editor is mostly unnecessary for people to publish on the web.
This is completely freaking me out.
I can only wonder at how a radical re-design this is going to turn out to be, from both a developer's and end-user's standpoint. The Mozilla project has, by all accounts, been an incredible success, and has been adopted by some major entities, eg. Sun, HP, IBM, Red Hat. By making this radical a change this soon after 1.0, do we risk alienating users and developers? I mean, now that people have gotten used to Mozilla, we turn around and dump something hugely different in their laps?
My fear is that commercial entities, along with the pro-Mozilla-the suite camp, will continue development on Mozilla Classic (the 1.4 branch), while the Phoenix folks work on NGMozilla...a fork.
Hold onto your hats, folks.
If all the world's a stage, anyone who says they want better lighting spends far too much time in a dark theatre.
I have not seen anything in Mozilla for a "grid" widget (kind of mini-spreadsheet in which you can change the cell contents). It has a "ListBox" widget, but it does not seem to allow direct editing of content.
Call them all the names you want, but it seems only MS and Windows products seem to offer decent grid widgets. Java's grids sucked eggs or had "unnatural" conventions from users' perspectives. (Perhaps their convention impressions are shaped by Windows, but that is the reality of the work force, for good or bad.)
Table-ized A.I.
Interesting.
;)
Not so long after this article about Minotaur came out on Slashdot, along with a bunch of comments talking about how cool it would be if Mozilla were to move away from the monolithic-bloat model towards the lean-mean-module approach, the Mozilla team says "hey, let's do that!"
They must read Slashdot! Well, of course they do, but they must actually take it/us seriously.
hey, anyone wanna hire me?
- Murphy's Corollary: - It is impossible to make things foolproof because fools are so ingenious.
Mozilla for all it failing is starting to get real recognition as a product in the market, it is being seen a real alternative to IE and now just as it starting to get traction it gets KIA.
I find it hard to believe that that a Phoenix based browser is going to reach a level of stability and adoption, any time soon.
(Someone has to say this)
This smack of developers looking at the technology and saying you know I could do better (yeah I'm guilt here as well). You know lets rebuild this and provide no migration path for existing applications and users. There must be a better way to do this with less risk and disruption.
Remember it's not always the product with the most technical merit that wins it the wins just look at IE.
Otherwise Microsoft is going to be very happy they will not have to worry about Mozilla any more.
I disagree. They took some of the ideas from Communicator, but dropped some of the others which I found important. For instance, communicator was very fast in it's own right, communicator's editor is better than Mozilla's, etc, and the interface was quite clean and elegant.
You've got to be kidding. KDE is farther gone than Mozilla. The bloat and lack of performance in KDE is incredible. In KDE, you have a base and libraries that are absolutely huge, and all the programs depend on them (somewhat like Mozilla). Just about anything you want KDE to be able to do, has to be compiled into the base system, and by god the KDE developers want every feature imaginable able to be compiled-in.
You've got it backwards... to use Mail you DO have to load Konq, the same way that, using outlook, you have to load IE. Besides, I don't know where you got the idea that Mail is loaded when you start Mozilla. They certainly share a good deal of their code, but Mail remains out of memory until you lauch it.
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
Well, I'm thinking "I wish I could use phoenix, but I need Mozilla for the mail client." So anything which puts pressure on Minotaur to emerge sounds good to me.
I also like the newer Phoenix nightlies (using 3-9-3 at work now) but I have one big complaint, and this showed up in Mozilla 1.3, too: I frequently click and drag a link to another tab, but the recent Phoenix nightlies and Mozilla 1.3 seem to frequently not pick up the link and I have to try two or three times to drag the link and it's driving me nuts.
Does this have anything to do with mouse gestures? Can I fix it? Lately I seem to have better luck if I drag the link to the right before dragging it up to the tab; this is why I suspect it may have something to do with gestures even though I've never tried or even read about gestures.
The Phoenix nightlies are MUCH slicker looking than both Phoenix 0.5 and Mozilla 1.3. Very nice. And I love having the home button back where it belongs (as opposed to where Mozilla has it): in line with the back, forward, stop and reload buttons.
their vision on this platform thing is off the target.
they mozilla is actual bloated.
But they still deny that mozilla is slow as evident on the
traffics in mozilla performance news group.
Maybe we *do* want the other components, but don't want the browser. Currently there is no option to just download the Composer and Mail portions. Much as I love mozilla, (and I'm using it right now) some of my clients would be unwilling to replace their browser (undoubtedly IE) but would be happy to install Composer to get an easy to use HTML editor or Minotaur to get a better mail application than Outlook. That is why *I* would like to see this change. And who knows, maybe after they get comfortable with the composer or mail app, they'll be easier to convince to move to a browser with a "consistant" interface.
This is SO educational! -- Kintaro Oe
If you want the new release following this new roadmap to be called Mozilla 2.0 rather than 1.5 vote for the bug here
4 54
http://bugzilla.mozilla.org/show_bug.cgi?id=200
Mr. Smoove
Apple chose KHTML because it was easier to own, if necessary, than Gecko -- that's true. But that fact doesn't make KHTML the better open source anything.
Two points:
1.) I believe the post's author meant that KHTML was better for Apple in this situation and wasn't trying to propose that KHTML is always better than Gecko, which is the interpretation you use in your reply. KHTML was better than Gecko for Apple's needs, almost by defn at this point. Gecko renders, in general, web pages and web page standards more exactly, and would be better by that benchmark, and some other that you list, etc. It's all situational.
The original post just says, "Look, you don't always use Gecko! There is an alternative! And Apple thought it was better!" And because of this, Gecko is getting "better" from the lessons it's now learned from current contrasts in comparisons (comparisions given new weight and credibility thanks to Safari) with KHTML.
2.) Since we're talking Gecko & KHTML, there's really nothing about the MPL that makes Gecko horribly much more difficult to "own" than KHTML. So when you tie any of this thread to Apple not "checking back in" all of its "tine", this isn't an arguement for Gecko. Heck, what's Netscape doing (they've had a spellchecker for quite a while, even with the Mozilla foundation, for example)? And Netscape is why we have Gecko in the first place.