Slashdot is powered by your submissions, so send in your scoop

 


Forgot your password?
Close
typodupeerror
×
Mozilla The Internet

Mozilla.org Posts New Roadmap 170

Posted by Hemos from the updating-the-.plan dept.
berteag00 writes "The mozilla.org staff has posted an updated roadmap detailing the Mozilla code base's relationship to the upcoming launch of Netscape 6."
This discussion has been archived. No new comments can be posted.

Mozilla.org Posts New Roadmap More

Mozilla.org Posts New Roadmap

Comments Filter:

  • A user perspective... (Score:1)

    by Anonymous Coward

    It's a lot more fun to use elegant, bug-free, minimalist software than it is to use a bloated, buggy monolith. The Mozilla people forgot an important stage of the development process: ripping out useless crap. They should have dedicated a few milestones to it...they've got a long way to go in that direction.

    Mozilla, as it stands, is an excellent example of why you should design software to do as little as possible. Every extra feature not only takes extra time to develop in the first place, but also extra time debugging, extra time migrating code when underlying stuff changes, extra time documenting, extra CPU time spent compiling, extra CPU time loading the executable, extra memory, etc...

    Really, about the only part of Mozilla that interests me is Gecko, because Nautilus requires it. Everything else important is supplied by Nautilus in a (IMNSHO) far superior package.

  • Re:Only one thing... (Score:1)

    by Anonymous Coward
    Yeah, but the Mac IE is a totally different codebase from the Windows IE. Microsoft had to throw one bunch of engineers at the Mac version, one bunch at the Win32 version and another bunch again at the Unix version. The whole point of Mozilla is that it is cross-platform and a single code base. While this may mean problems in the short term for one particular platform, it means more robust and tested code in the long term and for less money.

    Mozilla is making accomodations for the Mac, so things are getting better. For example, one of the primary problems with Mac performance so far is how shit its multiple open file access times were. All the chrome and skins is going into a small number of JAR files so this problem will lessen.

  • Re:You don't program for a living, do you? (Score:1)

    by Anonymous Coward

    I don't mean for this to sound like a troll, but have you ever heard of the concept of time to market?

    A lot of what you propose is nice and elegant in theory, but doesn't work well in the real (Cathedral) world. It may work fairly well in the Bazaar world where time to market isn't as much a factor, but if you're on a project that scratches a major itch, people are going to clamor if the project drags on.

    Mozilla is one of those Bazaar projects which scratches a HUGE itch. And we already have heard the gripes about how Moz must be dead since they are "taking forever". "Realease when it's ready" is well and good, but you still have those in the community who want or need projects like this in a timely manner. All these academic Software Engineering techniques have limited benefit when timelines are tight (whether due to Cathedral TTM or Bazaar Huge Itch need it now).

  • In case this is actually ever read....

    I was actually in the same boat as you are. The latest nightly builds however, are stable enough for me to use mozilla as my browser of choice now. Yes, I'm on a pIII-550 with 256mb ram, yes, it does come down occasionally, yes I did have to install the https personal security manager mentioned in the other mozilla thread today, but it's usable and I'm happy. I will be happier when it's totally stable and fast, etc, but I can say that I am happy I held in there.

    Give the nightly's a shot, you may be impressed!
  • A lot of the "features" added (office suite, theme designer) are not mozilla features, they are other people creating something *for* mozilla. Thisis cool IMHO, as long as it doesn't slow development....
  • Excessive modularity is often the downfall of major software products.

    Didn't I already call you out once on this stupid view the other day? Aparently you have a problem with reading comprehension.
  • If I understand it correctly, their choice to release the source code, was to expand the user base.

    You do not understand it correctly. Back to boot-camp with you. Get down and give me 20 recitations of CatB.

  • Well, if if Netscape 4.7 is any indication then it will be a long time. they STILL haven't finished the last 10% of IT yet.

  • I haven't gotten a nightly to run in about 2 weeks.

    I run ./mozilla and it just sits there.. nothing happens.

    Last good one was like 083100

  • Not a feature freeze exactly, but the article says:

    Mozilla doesn't need new features, or any particular "new" or "next generation" module (there will and should be module rewrites, to be sure; there will also be new features, but I'm not about to call for any here).

    --

  • this is kinda off topic, but can any one tell me how to get the popup menus in mozilla to act like every other popup menu in x?

    in every other x application i run, to use a right click menu, i have to:
    push the right button,
    drag the mouse to the item i want to choose
    release the right mouse button

    but in mozilla, the sequence instead goes:
    push the right button
    release the right button
    drag the mouse to the item i want to choose
    push the right button (again)
    release the right button (again)


    i'm told that this is configurable, because every thing about the user interface of mozilla is configurable through xul or xsl or some tla, but i've never been able to figure out how.

    while, i appluad the mozilla developers attempts to make the browser render pages exactly the same in every single platform (good job by the way), they should have left the behavior of the application to follow the normal behavior of the platform the user is on. the popup menu thing drives me nuts, and it's probably the single biggest reason i still don't use mozilla (and probably never will if i can't figure out how to change that.)

    on a slightly related note, all of the mac os users are gonna freak out because mozilla's not going to have the menu up on the top of the screen like every other macintosh application in the world....
  • i checked this out in bugzilla for the first time today. i had always assumed that this behavior was intentional, as it is the normal behavior of a popup menu in windows (for everything ns4) which presumably is their most targeted platform.

    now that i looked into it, the bug you posted is one of 19 duplicates for bug #16766. bug 16766 was originally scheduled to be fixed for M13, and has been pushed back for every milestone since then. for a while it was marked as "WONTFIX" for the reason i had mentioned above (following windows platform behavior) despite being the most requested/duplicated bug out there. (from bug #49844: "Bug 16766 should be marked as WONTFIX, as well, given the resolution of this bug. And that's going to make more than a few people unhappy.")

    it appears to be reopened now, but from the comments later in the page it sounds like this is not going to be fixed for the release ns6 (despite several peoples comments that the non-platform standard behavior of ns4 over ie is one of the major reasons they prefer to use ns in windows) because it appears to require some major working of the way they handle events. it appears that it took them from october of last year when the bug was reported to the end of april of this year just to figure out what was happening. And, quoting one of the more recent comments for that bug: "This defect is just not worth the time and risk. In order to ship this year, which we *must* do, we cannot afford to make unnecessary changes to the codebase."

    so, in the end, it sounds like mozilla users on linux are just going to have to live with crazy ass popup menu behavior. here's hoping galeon gets popup menus soon...
  • By "helper apps" do you mean applications to handle protocols the browser isn't aware of, or applications to post-process downloaded files?

    The former has been in for quite some time. It's the latter, though, that's far more desperately needed. Is this the one you're saying has landed?
    ----------
  • It's ironic that Microsoft got into so much trouble for including a browser with the OS which they had a monopoly with while Netscape at the same time was including email and a development platform with the browser they had a monopoly with.
    No it's not.

    If all e-mail and news reader applications required the Netscape 'platform' in order to work, and the market for e-mail and news reader clients became really hot, forcing Netscape to bundle their e-mail and news reader clients with their 'platform' in order to compete, that would be ironic.

    At that point, everybody would say "How in the hell did we get ourselves into such a position that we're completely dependent on Netscape."

  • Which platform, Linux?
    You do clear out your old binary before unrolling the tarball, right?

    Have you tried trashing the contents of the .mozilla directory in your home directory?

    I download the nightlies just about every weekday, and in the last three weeks I've only gotten 2
    that were unusably broken.

    (posted w/ Sep 20th Linux nightly build)
  • Off topic, but look closely at the Mozilla branch graphic:

    http://www.mozilla.org/roadmap-im ages/branching.gif [mozilla.org]

    Look at the "Netscape 6 RTM". See it flash? Why? Is there a subliminal 'RATS' in there? Hmmm.

    The wheel is turning but the hamster is dead.

  • I tried M17 and it was really fast for me. The windows version at least. Now, I'm not running on p166. I've got p3-600 with 128MB RAM. But still, it's as fast if not faster than IE. The Linux version is a different story... But perhaps that's because they have all the debug code compiled in?
    ___
  • Have you ever tried modular development? The best way to do it is to make several modules, and make sure they all work. It makes sure that you don't have any false assumptions. Now, making it work with these doesn't PROVE that it is sufficiently modular, but without doing such development, it would be impossible to even attempt it. Personally, I'm not at all a fan of XUL. I wish the Mozilla team had focused on the browser as a component, and let other existing programs be used for mail and such. However, being as I'm not a developer (although I have done some testing and bug reporting), I have very little room to talk. I am actually quite impressed with the amount of progress Mozilla has made. I think its unfortunate that many people look at Mozilla as a failure. In fact, it is a great success! Some people are upset that it didn't ship in time. Well, as everyone knows, great products take time, and they'll ship when they are ready!
  • Yes, 4.X used less cycles. However, you had to wait for the WHOLE FREAKIN PAGE WITH ALL ITS GRAPHICS to load in order for it to display anything. However, with Mozilla, it progressively renders as it loads. Therefore, it uses more clock cycles, but it renders faster. In NS 4.x, it sat around in a wait state, but in Mozilla, it
    is actually doing something. I've been using Galeon for a while, and it loads about as fast as NS, and the page loads are much faster.
  • OK, I see what you mean.

  • Re:LOL! (Score:1)

    by PD ( 9577 )
    The reward for finding a bug in one of Knuth's books is a binary dollar - 2^8 pennies - $2.56.

    He's written a lot of checks for $2.56, but very few of hem have been cashed.
  • It's going to be interesting to see how the "management" committee handles their new job.

    I'm also curious as to how the non-Netscape developers feel about committing to a more rigid schedule of releases. This is going to become an issue once other projects release dates become threatened.

    How is Nokia going to react if their product release is delayed due to some missing functionality from the base Mozilla release?

    Just a few random thoughts...

  • Didn't you hear? They didn't think that Mozilla was late enough already, so they've now started to integrate The GIMP into the codebase. :)


    Cheers,

  • BUT I find it hard to believe mozilla is fast and stable on a 166. I've been running the builds(out of cvs) for a long time now and it chugs slowly along even on a P3 500.

    I agree. It creeps along on my P5/233, takes sometimes as much as a minute to load a browser window (spins up my CD-ROM drive for some reason) does not do JAVA, fails to work with secure sites, crashes several times a day (I've filled in that feedback full-circle wossnot a million times) looks exceedingly ugly and is, generally, the shittiest browser I've ever used.

    On my Celery500 I use Netscape 4.73 and have NO plans to replace it with M17 and have grave misgivings about downloading any later milestones any time within the next two years.

    Frankly, I think Mozilla is the worlds leading Open-Source failure and I doubt it will ever come to anything. At the rate of progress I've observed it will take them another two years to come even close to that POS from M$ and by then the world will have moved on. The number of users will probably remain (statistically) zero until the project is abandonned.

    I'd love to think I was wrong, but....
  • These days most sites require javascript and/or java to run "correctly," and email and AIM are both really popular.

    Uh-uh. Won't fly.

    Javascript support is a necessary part of the browser application, I'll grant you that. I'd even grant Java, except they don't seem to have bothered to include it. At least not yet.

    E-mail? AIM? Those are totally separate applications that should be produced separately, with their own separate executables, and their own separate release schedules. The browser should not be delayed so that an e-mail client can be released as part of the base package.

    In case you have not noticed there are a few dozen e-mail packages already available, and I doubt the majority of people who currently use e-mail under UNIX/Linux are in any hurry to switch! I've used Netscape since 1.0 and I've never used any integrated e-mail client.

  • Mozilla M17 runs fast on my PII-400 both in windows and linux.

    The only thing I can't do is crypto stuff.
  • maybe what needs to happen is to replace the Konqueror rendering engine with the mozilla rendering engine. Since it is being dually licensed this may be possible. This would give Konqueror a better rendering engine.
    *SIGH* This was just covered in a thread today on Slashdot. Check it out.
    ----Woodrow
  • I've been running a few nightly builds since M17 on my P3-500 and it's smokin'. On my P200MMX at home I run a nightly build from three weeks ago, much improvement and much faster than NS4.7, maybe even a tad faster than IE5, but the bugs show a little faster too. But then NS4.7 is "buggy" at home as well - and got buggier after a "Windows Update" (which I will never run again).

    It's just short of becoming my default browser in both places. I think another two months will be good enough for me.
    --


  • I want "Just a Browser" to view "Just Sites" I frequently visit, not those screwed portals that don't work without JavaScript

    So run Lynx, or Netscape 1.0.

    Seriously, I don't understand (and never have) why people who say things like this have any interest in the Mozilla project at all (even enough interest to bitch about it.)

    You already have browsers that are ``just browsers'' and don't work on those evil sites that use Java and JavaScript and Flash and all those other modern inventions. What are you asking for that you don't already have in a dozen different forms?

    Most people want browsers that are capable of displaying bleeding-edge web sites. If you don't, then you don't need Mozilla at all. That's not what Mozilla is for. So why complain?

  • Are the Mozilla project investing too much effort into the irrelevant stuff, or too little?
    --

  • live in society and retire someday. Programming is something I'd do if I wasn't paid for it. The only problem then is how to pay rent, food, clothing, etc.

    Well, to some extent, but I find that after spending 8 hours a day at a programing job I'm sick of comptuers, and no longer want to do the codeing I want to do. To put it anouther way, I love programing, but I don't want to spend all my time programing, I also want to spend time fishing, playing mandolin, reading (I love history, bioragiphes, fiction amoung others) playing with the neightbor kids, building something physical, and like many geeks I harbor this desire to meet a girl worth dating.

    Programing fits in my life, if I wasn't paid to program I'd probably do it. However I'm paid to program, and that overfills my life's need for programing. Hence I'd love to retire or find a job doing something else.

  • Sure it's great to code an elegant design the first time that "does the job right". But what do you do when the definition of the job changes 75% of the way through coding? Doing the new job "right" would take more time than anyone is willing to give you. So you have to fudge it as best you can given the time you have.

    Or how about when you have to deal with legacy code written by someone who didn't care (or couldn't do better)? Sure it would be nice if you could rewrite the code to be elegant and such. But... the more code you rewrite, the greater the chance of one of those typos or such. And in the real world, it doesn't fly when you say you introduced a bug while rewriting the code.

    Just some things to consider.

  • I thought that the NS 6.0 installer is supposed to only install the pieces of the application that you want. If that's not the case, I'm sure someone will correct me. Now, on the Unix side of things, there already is something called Muttzilla [telus.net] that does exactly what you describe: launch your favorite email program. In my case, it uses gnudoit to send a message telling emacs to create a mail buffer with the given email address, subject, etc. I would expect something similar will be available for Mozilla.

  • duh... (Score:2)

    by ksheff ( 2406 )

    Given that it's at best beta quality, I wouldn't think that any corporations would be letting their employees use it for anything other than experimentation or testing it to see how it deals with production web pages. I wouldn't say IE has conquered all corporate environments. Netscape is the standard browser where I work. The only way anyone uses IE is if they download and install it themselves or they access it via a COM object in bLotus Notes. Considering that several desktops use Linux and Linux is beginning to be used for an intranet station, I don't think we will be switching to IE soon, either.

  • For crypto support, you have to download the PSM from iplanet. There is a link to it in the Mozilla release notes. Once installed, crypto works just fine. I've never had any problems with M17 and cookies (using it right now to auto login to Slashdot and post this message). Also, if Mozilla is slow on your machine, you need to get more memory. I added an extra 16M of ram (48 to 64) to a P233 machine and Mozilla became very responsive, almost like it was a entirely different program. Hopefully, someone will include the ability to launch an external client for mail from Mozilla like Muttzilla can do now with netscape 4.7x.

    I've tried Galeon, but the version I had tried did not work with proxies or sites requiring user authentication. Since about 100% of the pages that I view at work fall into either of those categories, it's of no use to me. The nightly mozilla builds are supposed to consume less memory and the new mozilla skin looks much better than the original. It's far from being dead, so why do you want to carve it up?.

  • It kind of goes back to the theory that open source programmers are scratching their own itch when they contribute to a project.

    Forget the ESR theory, what open source does great is fix bugs in software that we use. If more people used Mozilla as their daily browser, those bugs would get fixed. But, at least in my case (using Galeon too), it's just not stable enough to keep me from using Netscape 4.7X for the majority of my browsing.

  • There's always a few biased posts that completely praise the named product. In this case, mozilla.

    I think this is really a Mozilla-only phenomenom. I'd say the ratio of pro/con is greater for this project than any other.

    As for speed, if you read some of the old chats at MozillaZine, especially Mike Shaver's [mozillazine.org], you'll find that Mozilla will never be faster than Netscape 4.7. As Shaver himself said

    when it comes down to how many CPU cycles are needed to display a given page, I don't think we can _ever_ get faster than 4.x
    So except for things like tables, that were a mess in Netscape, everything will render a little slower, with perhaps the perception of rendering a little faster.

    And as for stability, I've run both Moz and Galeon (using the embedded moz gtk widget) on a P133 and they are about 5X more unstable than Netscape overall.

  • On the milestones page [mozilla.org] you see that M16 is labelled "feature complete", and rest of the work after that is bugfixing and optimization. How stringently they are/will be holding to that, dunno. He (Brendan) does say in the background paragraph that "Mozilla doesn't need new features" or new modules. So I really wouldn't worry about feature creep at this point.
    --GAck
  • Maybe your right and Konqueror will end up being the best browser, but if I have to install all the KDE 2.0 libraries just to get a browser on my system my responce is no thank you.

    And we all thought that Mozilla was getting bloated with an e-mail client -- now you have to install a destop with your Browser.
  • "/. has seen many articles about all the /stuff/ that gets thrown into Mozilla because it's "k3wl." Is there any information on a Mozilla feature freeze? I don't see any on the site..."

    The only two things that I can think of that qualify as "k3wl features" are the redesigned skins (classic and modern) that were a response to the constant complaints that the interface was ugly, and BiDi support, which has been planned for quite a while, but hasn't landed yet.

    There are a number of third-party applications that have been written using Mozilla as a base, such as Chatzilla, but blaming these applications on Mozilla developers is like saying that Linux 2.4 is late because too many people are working on KDE.

    I'd also like to add my voice to that of the teeming throng - I use Mozilla as my primary browser on Windows (work) and Linux (home), I usually update to the latest nightly once a week, although I update more often if the nightly is having problems. I don't find it noticeably slow compared to the alternatives. I really don't notice anything wrong with its day-to-day running except a bit of slowness with <TEXTAREA> boxes, and the occasional need to switch back to another browser when I need SSL (I'm too lazy to keep up with PSM).

    Charles Miller
    --

  • How could you tell if you'd found a bug in TeX ? Most of it is the formatting algorithm, which is inscrutable, and the rest is the interpreter, which is ...... ummmm ...... bad (though I hasten to add that its only bad because it was written before modern parsers were invented).
  • Your suggestion that they build the browser part first and then add the other stuff sounds like a good idea, but one of the reasons they didn't do this is to make sure that they modular and xp facets of the architecture would actually work and support different types of apps.

    By writing the email, address book, etc. they were breaking/fixing the modularity of the whole system. The bugs in design were exposed as other systems tried to use it. (This goes for some of the third party applications such as MozillaIRC and MLTerm).

    Ie, the other applications were in part a debugging exercise to test the overall Mozilla engine.

    W
    -------------------
  • I can't believe how fast Mozilla's build for today is running on my P166 and so stable too. Only a few minor superficial bugs (such as when changing theme). It looks and runs great. It is already my main browser on both Linux and Windows. All I need is the Java and security tied in the rest of the way.
  • Have you tried it? In the past three days or so it's been getting dramaticly better. I run everything from 486's to dual processor PIII 900Mhz boxes and have been using Mozilla since it first became a project. And I'm telling you it is running great on a P160 (and not bad on a P100).
  • The machine in question has 64Mb RAM. Have you tried adjusting the size of swap memory on your box? If you're swapping a lot you should make it bigger. If you seem to be spending a lot of time in swap for simple tasks then you should make it smaller and think about adding more RAM. Swapping to an IDE drive especially can slow things down a lot.

    Also note I said it's been really fast/stable the past 3 days or so. This implies that it has improved greatly.
  • Did you ever use IE3? Now that was a shitty browser. Obviously you don't remember some of the early browsers. ;>

    It's a product in development so of course it will tend to be slower and buggier than a final release. I look carefully daily at the latest builds and if the comments are good I try the build on one machine and if it runs better than my previous build I update all my machines. I'd say it is better than NS 4.73 in many ways already. Crashes far less often and when it does crash it does far less damage to the system. (I keep Netscape under limits so it can't crash X).

    If you haven't tried a recent build you have no reason to give an opinion of where the project currently is. If you think it's failed then go write your own browser. I think Mozilla will be one of several very successful browsers for Linux and it is really starting to show some hints of the finished project. It really doesn't matter who wins the browser war. If I can choose a stable, flexible browser that I have source to modify as needed then I've won. The project will never be abandoned as long as at least I am using it.
  • If you really want to be technical, that was not an ad hominem argument, but instead a confusion of correlation with causation. That is, he didn't attack a person, but instead drew the erroneous conclusion that since Mozilla was happening at the same time as Netscape's market share was decreasing, it was the cause.
  • Finally, this is on-topic. :)
    I've been using Mozilla as my primary browser for nearly a month now.
    Under Windows: IE has a very bad habit of garbaging the display if you scroll up and down... so does Netscape for that matter. Mozilla is immune to this.(IE dislikes 'big' jumps in scrolling but is okay with little ones; Netscape dislikes incremental scrolling especially with the mouse-wheel but is okay with big ones; at least, in my experience.)

    Under Linux: For some unknown reason that I really don't care to research, Netscape suddenly started giving me a Bus Error every time I try to start it. Mozilla runs flawlessly.

    YMMV, but I'm liking Mozilla M17 and looking forward to Mozilla M18.


    --Parity
  • I -am- running a relatively new TNT2 card, but Netscape was not having problems with it before; and I -do- have more memory now, -but- I've run memcheck86 on it several times enabling all tests and everything; nothing; perfectly good RAM - and, as I said, Mozilla runs flawlessly. That's for the Linux install.

    For the Windows install... no way. -Only- the Big Two web browsers have this problem. Acrobat doesn't have it, Mozilla doesn't have it, nothing in Office has it, my development tools don't have it, my various graphical player things don't have it... and I've seen it on multiple platforms. Sure, maybe it'd all be fine with a standard VGA driver in 800x600-256colors, but if I have to cripple the box to make a piece of software work, I'm not going to use that software.
    And again, Mozilla runs flawlessly. Well, almost; it has a habit of causing an exception after closing it, but, whatever. Doesn't affect my experience, just an extra button to click on close. (I don't use M17 Mozilla on Windows, though; M17 for windows is pretty flaky, I use the daily builds there because it's rapidly improving.)


    --Parity
  • It's not released in 2.x fashion yet. Only the beta has all that and if I want to be a beta tester I'll use mozilla

    I don't want a lot, I just want it all!
    Flame away, I have a hose!

  • maybe what needs to happen is to replace the Konqueror rendering engine with the mozilla rendering engine. Since it is being dually licensed this may be possible. This would give Konqueror a better rendering engine.

    I don't want a lot, I just want it all!
    Flame away, I have a hose!

  • I have tried it. If by more useful you mean it crashes more then yes. Last verison I had would not do flash properly, would not do plugins, and would crash alot. This was M17. If I want a program that will crash on me all the time I'll use Windows 95/98 thank you.

    I don't want a lot, I just want it all!
    Flame away, I have a hose!

  • Yeah, they're porting it to XUL [iplanet.com].
    ----
  • There already was a Netscape 1.0, 2.0, 3.0, 4.0. There wasn't a 5.0. They just skipped that generation. Really, the source code which they were going to release as Netscape 5.0 mostly got scrapped and everything was rewritten. At least that's my limited understanding of it all.
    ----
  • The thing is, the people who actually ARE contributing to mozilla are still strongly tied to netscape (okay not all of 'em), and they have wide acceptance as a goal

    If they've got "wide acceptance" as a goal, rolling their own Mail/IM/Calendar/etc client is a big mistake.

    I want a browser that has Java2 support, up-to-date JavaScript/ECMAScript, CSS, XML, etc.

    I want a browser that when I click a maito link, launches the mail program which has all of my mail and address book(Outlook Express on my home Mac and Outlook on my work PC).

    I want to keep using my calendar software.

    I want to keep using my AIM client.

    I do NOT want a bloated piece of crap that refuses to play nicely with others. And here's a hint: neither does virtually anyone else.

    Do you want to know the sick thing? The browser that most closely matches this description is IE.

    Mozilla is dead. Luckily, the useful bits of the carcass are showing up in usable programs (yay open source). But I can't see any sort of situation where I'd switch (or my company would switch) to the entire Mozilla/Netscape 6.0 environment.

    -jon

  • I can't understand why so many people saying that Konqueror will overtake Mozilla. Have you ever compard the two? Well, I did just that recently, and this is the conclusion I came to:

    Mozilla has an excellent rendering engine. It is very fast and displays all the pages I go to without any errors. While some people have complained about instability in Mozilla, personally I have found it to be quite stable. The biggest problem with it, in my opinion, is just the amount of memory that it consumes. Hopefully later revisions will improve this and removing the debugging code and compiling with proper optimizations should help as well.

    Konqueror is the complete opposite. Sure, it is a lot lighter than Mozilla, but that's about the only thing it has going for it. The rendering engine is horrible. I recently had the displeasure of watching it make my T3 feel like a 28.8. I would load a page with Konqueror and watch in almost disbelief as it SLOWLY loaded the images one at a time. I fired up Mozilla to look at the same exact page and it loaded in an instant. Not only that, but Konqueror frequently displayed pages incorrectly, even pages that are targeted at Linux users and presumably follow proper HTML standards.

    Now which of these would you rather have? An excellent rendering engine in a bloated app or a crappy rendering engine in a not-so-bloated app? Personally, I'll take Mozilla myself, although I suppose you could make a case for Konqueror. However, what really guarantees Mozilla's success, even if Mozilla itself is never slimmed down, is the fact that you can take the Gecko rendering engine and use it as part of less-bloated apps (Galeon anyone?). It really is a great rendering engine. But hopefully that won't even be necessary if the Mozilla developers work on optimizing it some more.
  • I like this roadmap. While it will not please the nay-sayers who want something out right now, it does show that they have thought about what they are doing. Quaterly releases are perfect: the product is updated in a manner that keeps it on everyone's radar, without forcing them to rush stuff out the door.
  • Yep, ad hominem attacks are always the best when facts start to get in your way. You've learned the Open Source playbook very well.
  • If I understand it correctly, their choice to release the source code, was to expand the user base. How has this been going for them? What percentage of users actually use Netscape?

    Lookie here at some nice historical browser statistics. [uiuc.edu] Netscape / Mozilla has about 27% of the market. Netscape has around 45% of the market when Mozilla was opened. So you can see that this strategy, like most things associated with Mozilla, has been a failure.

  • What new k3wl stuff? I don't see any of it. Every feature request in the bugzilla [mozilla.org] database is being marked as "future" now, and even bugs are often marked "future" or "nsbeta3-".

    (For those who don't know the Mozilla jargon, "nsbeta3-" means "not now" and "future" means "really not now".:-)

  • Mozilla is already faaaar more useful than Konqueror. Much slower, of course, but much better in every other respect. I really don't see why it should be overtaken by it.

  • IIRC, M17 was the feature freeze milestone. Looks like M18 (due any day now) will become the point beyond which Netscape 1.0 will branch out.
  • I dunno about direct cause, but the fact that Netscape haven't done any substantial changes to their browser in over 2 years, nearly 3, is almost certainly contributing to Netscape's decline. The reason for the lack of new versions? Because all their developers are working on Mozilla.
  • Mozilla is modular. When you run the installer, you can choose which components to download. Even if you download everything, you don't load mail/news code until you start mail/news.

    I'm not sure why Slashdot people don't realize this. I guess they've never looked at the code, looked in the bin/ directory, or even run the installer.
  • My take is that Brendan's busy enough doing actual work without have to worry about getting his diagrams pixel-perfect.
  • What use is Javascript without DOM support?
  • Mozilla's support for DOM2 is very nearly complete. Mutation events are the only big thing I can think of that isn't there yet.

    XML/XSLT should definitely be turned on for Mozilla 1.0.

    SVG will not be in Mozilla 1.0.

    There is a team working on JS2, but I don't know what their schedule is.

    Note that Mozilla 1.0 will have MNG and MathML, which IE doesn't have... IE doesn't have SVG yet either.
  • I don't know how many times this has already been explained at length, but here we go again:

    The IRC client was developed by a person on his own time before he was working at Netscape. He did it because it was cool. There is no impact on the rest of the browser. It was checked into the tree so that other peole could play with it and learn from it.

    To suggest that people should be prohibited from using Mozilla in this way is absurd --- and entirely against the spirit of open-source software, to boot.
  • > If that's really the culture at netscape, it's
    > pretty sad, though it explains a lot.

    It isn't, of course. It's a Slashdot user's fantasy. Or do you believe everything you read here?
  • > XML, XSL and CSS2 are the only things left to be
    > included

    I think you forgot "the DOM".

    And these aren't just things you hack in at the last minute. The reason Mozilla's taken so long is because they had to rewrite the browser to get these things in.
  • You quote Shaver out of context. As he goes on to say, *perceived* speed of Mozilla can be much greater than 4.x, due to incremental reflow and other stuff.
  • Helper apps via Internet Config has landed. Dunno if it works.
  • They're already using a lot of contributed bug fixes and some contributed features, mostly small stuff.

    These other big chunks (MathML, XSLT, MNG, PICS, Bidi, XMLextras) aren't even part of the main Mozilla build yet. It's considered just too risky to try to get them all integrated and fully working at this stage. All the Slashdotters whining about feature creep are wrong; it is nigh impossible to get big new stuff added to the tree, and has been for some time.

    After the PR3 branch, expect to see these cool new items turned on in Mozilla, and then appear in the next major Netscape release (6.1? who knows).
  • Yes, writing those applications doesn't prove that the interfaces are right, but it does prove that they're sufficient to build some interesting applications. And since the HTML editor and a mail/news client are pretty sophisticated applications, that gives high confidence that you can build other interesting applications.
  • Take a look at your Mozilla directory and count the shared libraries. Then look at "mozilla.exe". It's tiny. If you only use the browser, that's all that's loaded.

    Actually just using shared libraries is not nearly adequate. They had to build XPCOM as well.

    It disturbs me to keep reading comments "The Mozilla people are idiots, they should just do X, Y and Z and their problem would be solved" when in fact the Mozilla people have been doing X, Y and Z for years, and even that's not enough because the armchair critic has no idea of the full extent of the problem.
  • That's good to hear, honestly. Still, I wonder how much stability could be improved, and footprint be reduced, if some of the cool stuff just got temporarily pulled... hmmm.

    Just my out-of-my-butt thoughts.
  • Or how about when you have to deal with legacy code written by someone who didn't care (or couldn't do better)? Sure it would be nice if you could rewrite the code to be elegant and such. But... the more code you rewrite, the greater the chance of one of those typos or such. And in the real world, it doesn't fly when you say you introduced a bug while rewriting the code.

    So how would your statement above relate to such things as Refactoring [refactoring.com] which aims to do what you describe in the real world? It's an integral part of Extreme Programming [martinfowler.com] methodology which seems to be somewhat popular amongst the Open Source software developers.

    Refactoring has been, according to author Martin Fowler, practiced by the best software engineers for decades (long before XP came around). Are you saying it won't fly in the real world?

  • I don't know about you, but I like to be paid only so I can afford to live in society and retire someday. Programming is something I'd do if I wasn't paid for it. The only problem then is how to pay rent, food, clothing, etc.
    --

  • Disclaimer: I haven't been following the Mozilla development process so I'm not aware of their test practices.

    By writing the email, address book, etc. they were breaking/fixing the modularity of the whole system. The bugs in design were exposed as other systems tried to use it. (This goes for some of the third party applications such as MozillaIRC and MLTerm).

    Ie, the other applications were in part a debugging exercise to test the overall Mozilla engine.

    Writing those applications and getting it to run with them does not prove the overall design is modular. It does not prove that the engine plays nicely with other email, address book and IM clients - just that it plays nicely with the ones included in Mozilla

  • ...but why will Mozilla 0.9 come off of the PR3 branch? It looks like the "trunk" (which I assume will be the code that is being worked on under the auspice of mozilla.org) will get different changes made to it than the NS branch. Does this mean NS features will end up in Moz 1?
  • no way. there's still a lot of cool shit that needs to be added.

    maybe they'll get to 1.0 in Q1 of 2002 or something.
  • Have you seen the nightlies from this or the last week? Not only is Mozilla feature-frozen (for a long time now), they are committing to a virtually bug-free, fully working beta browser by the end of October (my estimate).

    Karma Police, arrest this man, he talks in maths

  • Is there any information on a Mozilla feature freeze? I don't see any on the site

    I think they more-or-leass addressed that in the 4th paragraph of the announcement:

    Mozilla doesn't need new features, or any particular "new" or "next generation" module (there will and should be module rewrites, to be sure; there will also be new features, but I'm not about to call for any here).

    There's not a DATE given for a feature freeze, but at least they're thinking about it.
    Sean

  • I want "Just a Browser" to view "Just Sites" I frequently visit, not those screwed portals that don't work without JavaScript
  • This article [editthispage.com] argues that rewriting code from the ground up is generally a bad idea, and cites Mozilla as an example. Combustible, but it's well argued and I thought a lot of the other stuff on his site was extremely interesting.

    Actually, when this came up [kde.org] on kde-devel, instead of a flamewar it generated a patch for klipper's faulty handling of the $47 in the URL.

    ---------

  • LOL! (Score:3)

    by mosch ( 204 ) on Wednesday September 20, 2000 @01:53PM (#765721) Homepage
    elegant and bugless code... yeah... right.

    you can come close, but you'll NEVER make it. i mean yesterday i was fighting a bug in PHP of all things, and when i tried attaching gdb to apache, discovered a bug in FreeBSD's binutils, where you can't get into the internals of a shared object despite having symbols properly loaded.

    that means, without writing a single bug, i had to deal with a) a bug in PHP and b) a bug in binutils which made it hard to find the bug in PHP.
    ----------------------------

  • Re:Mozilla is running great today. (Score:3)

    by Matt Amato ( 2494 ) on Wednesday September 20, 2000 @10:52AM (#765722)
    I'm not trying to troll or anything, but I've noticed an increasing trend hear on /. Mainly, when a article gets posted dealing with a hot topic, i.e. mozilla or KDE/GNOME debate... There's always a few biased posts that completely praise the named product. In this case, mozilla. I'm all for making mozzila the best browser available, just as I am for making GNOME AND KDE great desktop environments, downing any open project is a Bad Thing(tm) BUT I find it hard to believe mozilla is fast and stable on a 166. I've been running the builds(out of cvs) for a long time now and it chugs slowly along even on a P3 500. So how about some honest comments for a change instead of just praising a project because you like it.

    Just my 2 cents..
    Matt

  • Into 2k (Score:3)

    by josepha48 ( 13953 ) on Wednesday September 20, 2000 @11:38AM (#765723) Journal
    Looking at the milestone, they are now saying that M18 is the last release using the Mx naming conventions adn now they are going with Mozilla 1.x or startting at .9. Well this looks like history repeating itself. If you look at Netscape they actually started with .9 as a browser and then moved to 1.x and so on. If history repeats itself Mozilla will ahve to wait about 2 years before Mozilla 2 or 3 get out and it has a viable (supportable) user base. By then kfm or the Konqueror will be out in 2.x or 3.x fashion with full flash, JavaScript and Java support and people will have UNIX users will have moved to another browser. I know that Konquere is looking better and better and am waiting for kde2.0.

    I wonder when Netscape 6.0 will actually be released and how stable it will be. I also wonder how the themes contest is going. No updates of that mentioned on slashdot.

    And the winner is???

    Here is agood browser timeline.http://www.dejavu.org/ [dejavu.org] I know it may be old for some but it was rather interesting to get to try these out ;-)

    I don't want a lot, I just want it all!
    Flame away, I have a hose!

  • Re:Maybe I'm just not smart... (Score:3)

    by mackman ( 19286 ) on Wednesday September 20, 2000 @10:23AM (#765724)
    As far as I can tell, the code handed over to the commercial Netscape developers is going to be called the "Mozilla 0.9" branch, and very little Mozilla development resources will be applied to it from then on.

    The reason it continues is twofold, first, because Netscape 6.01 will probably be based on the Netscape 6.00 code instead of newer Mozilla code, and second, because things like MathML are maintained by the Mozilla team but need to work with Netscape 6.00.

  • Re:That last ten percent... (Score:3)

    by Rombuu ( 22914 ) on Wednesday September 20, 2000 @10:59AM (#765725)
    Fixing bugs really sucks!

    That's why most programmers like to be paid for their work.

  • Netscape 5.0 (Score:3)

    by thales ( 32660 ) on Wednesday September 20, 2000 @02:49PM (#765726) Homepage Journal
    Netscape 5.0 was under development, when the decession was made to go opensource. A pre-alpha of NS 5.0 was the code that was turned over to Mozilla.org in March of 1998. In October of 1998 the 5.0 base was scrapped, in favor of Gecko.
    There WAS a Netscape 5.0. It never reached Beta but it exisited. The Classic skin in Mozilla is based on Mozilla Classic, which was the 5.0 code. If you want a look at it the source is still avaible at Mozilla's ftp.
    About all thats left of Netscape 5.0 now is the icons in the Classic skin and a directory called users50, So it makes perfect sense to call the new browser Netscape 6.0

  • PSM, Java, Javascript all work (Score:3)

    by tjwhaynes ( 114792 ) on Wednesday September 20, 2000 @12:05PM (#765727)

    Mozilla is still a long way from being useful. It still eats up 80Meg just for one session... Until they clean that up, Navigator and IE will be the best options.

    Guess you aren't up to speed on things (and from your comments you are still deep in the woods) the current version of Mozilla runs at about 35MB on both Linux and Windows and no longer leaks memory left right and centre. Java works on Windows, PSM is available for Windows and Linux and the NSS 3.1 beta should help fill in a lot more of the https functionality. So Netscape has been left pretty much forgotten on my Linux box as Mozilla now handles my browsing and mail needs. I also run Mozilla at work on Win NT as it now outstrips Netscape and runs neck-and-neck with IE for speed, despite the GUI independent interface, so Mozilla is my browser of choice on my two main platforms. Stability is now better than five or six hours and its getting better pretty swiftly.

    Cheers,

    Toby Haynes

  • gimp? (Score:3)

    by wishus ( 174405 ) on Wednesday September 20, 2000 @10:19AM (#765728) Journal
    gotta love that professional looking graphic illustrating the branches.
    Vote for freedom! [harrybrowne2000.org]
    ---

  • Their Strategy (Score:3)

    by empesey ( 207806 ) on Wednesday September 20, 2000 @10:27AM (#765729) Homepage
    If I understand it correctly, their choice to release the source code, was to expand the user base. How has this been going for them? What percentage of users actually use Netscape?

    I cannot imagine that the average user would be lured to using a package, simply because the source code is available. I only know two people who use it, and one of them uses IE most of the time. Do Linux users see Netscape as an attractive option (is it even available for Linux)?

    Nice picture by the way. Do I have to put up NetNanny for the youngsters?

    --

  • Only one thing... (Score:4)

    by Millennium ( 2451 ) on Wednesday September 20, 2000 @10:38AM (#765730)
    I would strongly suggest that the fork into the Mozilla and PR3 branches not occur just yet, particularly if PR3 is supposed to be a feature-freeze for NS6 (and judging from that illustration, it appears that's the case). There's still a lot to be done on Windows/Linux, and the Mac version is lagging far behind those, particularly in terms of security (no PSM, despite it being promised "soon" for many months now) and helper apps (which, last I checked, didn't even work with InternetConfig yet, though it was certainly planned).

    Besides, I liked the old Milestone system. Granted, some of the Milestones might need to be changed to reflect new developments, but it gave a clear look at where one is, how much progress has been made, and above all it wasn't a version number so there was no mistaking it for anything but a developmental release (this was Netscape's biggest problem with the preview releases; people keep seeming to think these should be fully finished browsers).
    ----------

  • That last ten percent... (Score:4)

    by devphil ( 51341 ) on Wednesday September 20, 2000 @10:25AM (#765731) Homepage

    We are near the last ten percent of the "Mozilla 1.0" project, where the going gets tough.

    We all know the quote about the last ten percent taking the other ninety percent of the time, etc, etc. I think the last ten percent of the project is going to keep expanding (always remaining at "ten percent," of couse :-) unless they impose a freeze on the nifty neato keen cool shit that keeps getting added.

    /. has seen many articles about all the /stuff/ that gets thrown into Mozilla because it's "k3wl." Is there any information on a Mozilla feature freeze? I don't see any on the site...

  • "Just a Browser"? (Score:4)

    by rkent ( 73434 ) <rkent AT post DOT harvard DOT edu> on Wednesday September 20, 2000 @11:00AM (#765732)
    A ton of people are writing saying that mozilla should have started off as "just a browser" because "that's all I really want, a stable browser for UNIX." Unfortunately, that's not what there's a market for! These days most sites require javascript and/or java to run "correctly," and email and AIM are both really popular. So, saying that those things should wait until later is basically garaunteeing that mozilla doesn't gain any marketshare until later.

    Of course, one could argue that hey, this is an OpenSource project, it's the platonic ideal of software development, we the hackers can have whatever we want. Well, yes. Go ahead and check out a mozilla build and roll your own [kmeleon.org]. You can do that. The thing is, the people who actually ARE contributing to mozilla are still strongly tied to netscape (okay not all of 'em), and they have wide acceptance as a goal. Which means it needs all that stuff. I don't think they made a poor decision by including it.

  • Re:"Just a Browser"? (Score:4)

    by donutello ( 88309 ) on Wednesday September 20, 2000 @12:04PM (#765733) Homepage
    Whoa! You start with saying the browser needs to be extended with javascript and/or java and you are right there. There's no point shipping a browser which won't work on the majority of web sites.

    Then you make a huge logical leap by saying that email and IM is important so must be part of the browser. They are completely unrelated to browsing.

    It's ironic that Microsoft got into so much trouble for including a browser with the OS which they had a monopoly with while Netscape at the same time was including email and a development platform with the browser they had a monopoly with.

  • Re:That last ten percent... (Score:5)

    by Hard_Code ( 49548 ) on Wednesday September 20, 2000 @10:52AM (#765734)
    You know what's even more satisfying that both? Creating elegant and bugless code in the first place. Sure it may sound facetious, but I really do get a lot more satisfaction upon committing rock solid code that does its job right, and is designed well, than coding up some whizbang piece of code in a frenzy then leaving it for poor bug hunters to waste thier lives trawling through for bugs. I'm of the opinion that even if it takes you twice as the first time to get something right, that still far outweighs the penalty of having to come back over and over again to fix bugs. The only "bugs" there should be are either typos, brain-blackouts, or really subtle design issues like threading and locking, etc.

  • Re:That last ten percent... (Score:5)

    by rkent ( 73434 ) <rkent AT post DOT harvard DOT edu> on Wednesday September 20, 2000 @10:47AM (#765735)
    /. has seen many articles about all the /stuff/ that gets thrown into Mozilla because it's "k3wl." Is there any information on a Mozilla feature freeze? I don't see any on the site...

    From an outside perspective, it's easy to say this. But, realistically, it's a lot more fun to code in new features than it is to fix bugs. I mean, anyone who develops professionally knows how much cooler it is when your manager says "hey, figure out a way to code in feature X" than when she says "could you take a look at bug A1.006 and check in the fixed code when you're done?" And then you have to sit there and stare at this stupid text box that sometimes pops up and sometimes doesn't for no apparent reason, and check out 4 modules you didn't even think were affected because they're not interacting properly... etc etc etc. Fixing bugs really sucks!

    It kind of goes back to the theory that open source programmers are scratching their own itch when they contribute to a project. And most of the itches have been 90% scratched on this project. So who's still going to be contributing? Fortunately, there's a lot of effort still being put in by netscape; I don't think mozilla has really been depending much on huge fleets of independent developers. And we see how quickly it has gone so far. I wouldn't expect the pace to pick up now that the fun work is over.

Slashdot Top Deals

In practice, failures in system development, like unemployment in Russia, happens a lot despite official propaganda to the contrary. -- Paul Licker

Close