Mozilla M9 Released

_m writes "The boyos at mozilla.org have dropped mozilla M9 and, from looking at the m10 nightly releases, it looks really promising. Go out and support your local developer. Still some small problems, but it looks like quite a lot of the important things have been sorted. Go alt tags! "

Comments

The good and the bad

[Anonymous Coward]

Just downloaded M9 at work and rushed home (no, I don't really have much of a life) to install it. First impressions, starting with the bad...

BAD: Dammit, it's still basically brings my machine to a halt when I run this thing on Linux. Just running the mouse over the slashdot navigation to the left can bring my system up to 80% load. I know this is still alpha, but on the same machine in Windows 98 (with M8) this was not a problem. No one else seems to be complaining about this. Is it just me? I have a PII450 with 128 Meg of RAM...

GOOD: Lots of Javascript fixes, including a bug I submitted that was marked as a "won't fix - as designed".

REALLY COOL: It uses my Gnome widget themes! For some unknown reason, I feel all warm and fuzzy inside because of this :)

I have to say to all the Mozilla coders out there that as a user of your product(s) and a web developer please keep up the great work! If any of you ever come to Boise, ID I will personally buy you a round at a local pub!

David

error: method `~` of object `~` failed
Actual Microsoft Error Message

Re:KFM!

[dirty]

I have a cable modem and I find it's a big time saver. Nothing against /. but it can be disgustingly slow at times. I really love though, how in unix clicking w/ the middle mouse button is the same as right click->open in new window. God I miss that at work...

It won't let me post without a subject.

[John Campbell]

And.... we have form widgets! And they don't even jump out of the way when I click on them! Woohoo! Mozilla has just joined the ranks of usable web browsers... On the other hand, apprunner doesn't want to work... it crashes silently at the end of its startup spiel. viewer works fine, though. I'm okay with that... I don't need all the junk they packed into apprunner anyway. I've already got a mail client and a newsreader... I just want a good, solid, standards-compliant web browser... This post finely hand-crafted using Mozilla/5.0 (X11; Linux i686).

You can fix the pointer

[dirty]

Somewhere in the KFM settings there is an option for "change pointer over links" or something along those lines. I can't check exactly where it is right now because I'm at work in nt*blech*4 but the options aren't that extensive.

Re:grrrr...

[tgd]

Just as a heads up, in the last two weeks or so, the CVS tree has not been able to build for me on any RedHat 5.2 system -- the only functioning builds I've gotten are on RedHat 6.0 systems, so downloading the source may not help you.

The configure crap makes a LOT of assumptions right now, and doesn't actually know what library versions it actually needs, so its often times a real bitch to figure out what everyone at Mozilla seems to have upgraded on their systems that you haven't.

Updating imlib, gtk+ and gdk don't seem to have fixed the problem when I upgraded those yesterday (and promptly broke Gnome...), although the problem seems to come from imlib.

There are often wierd problems running with older system libraries, and things like that.

If you're going to delve into the dark and mysterious world of Mozilla building, here's a couple tips:

1) Run the most recent versions of everything. If that means upgrading to RedHat 6.0, you're probably going to have to do it. They're having similar issues supporting VC++ 5 and 6 under Windows, and the growing trend is build with the newest, and fix it later...

2) The client.mk script, and makefiles aren't very reliable. If you keep blowing core, and aren't seeing lots of other reports about it (and tinderbox is green), blow away the tree, and repull a new copy. Dependancies don't work all the time, nor does make clean, and that seems to be the only reliable way to fix it. Sucks over a modem, huh? I'm suspecting that may be the problem here...

3) Sort of restating #1, but if you're running a stock RedHat 5.2, upgrade your glibc unless you've got a pile of ram. There's some issue with loading libraries multiple times that chows down LOTS of RAM on a stock 5.2 system.

4) When gdb asks you if you want to load all the symbols, *SAY NO* if you don't have a quarter gig of RAM or something. Trust me on this one. :)

RTFWebSite

[dirty]

Head on over to www.mozilla.org and read the damned thing yourself. I remember a day when people actually actively searched for information rather than expecting to have it handed over to them on a silver platter.

Instead of getting milestones...

[Rayban]

I know everyone loves getting the milestones, but the project could always use a few helping hands for stuff like QA (general bugfinding), development or even just suggestions.

Take a look at bugzilla (bugzilla.mozilla.org) and get an account there. Browse the bugs to see what the report looks like and see if you can submit an original bug, make a test case for an existing one or whatnot. We're all going to be using this browser in the end, so you may as well have some say in how it turns out. ;)

Enjoy M9! :)

Will Mozilla ever gain outside contributors?

[Paul Carver]

I finally got around to looking at the Mozilla source and as expected it's enourmous. I've also looked around the mozilla site and I have to ask: Do the resources exist to allow outsiders to contribute to Mozilla?

Yes, I know that the source is available, but this is a huge object oriented project. Source code alone just doesn't cut it. Where are the models? Here, where I work there are information models, state transition diagrams, interface event documents, how can anybody be expected to dive into 19+ MB compressed of source code?

Am I mistaken, or is there virtually no information available that would allow someone outside of the project to figure it out without brute force reading all of the code?

Re:Will Mozilla ever gain outside contributors?

[ajv]

Yes - I was at one stage contributing actively to the Rhapsody/MacOS X Server port. I got out of that when Apple canned the x86 port of MacOS X Server - something that is still 100% boneheaded. My Rhapsody DR2 install no longer works, so basically Apple lost a developer FOREVER. Oh yeah, I was going through a bit of a down patch in my life, and I wasn't coding much. But I did have commit privs on the CVS tree for Mozilla/Rhapsody, so it is possible.

If you have an itch, scratch it - download the source, get cvs in order and sync with the latest source.

Even if you can't code, get the latest stuff, build it (it's not hard) and run it through its paces. If you're going to debug the lizard, you'll need mucho memory (on my 96 MB dual PPro, it just swapped...). But documenters, bug testers and more than just the occasional fix really help. High quality bug reports are worth their weight in gold.

Maybe think about profiling. If memory or CPU usage is bugging you, compile with the profiler options turned on and figure out where the problem is, and fix it (or suggest a fix) based upon your research.

Re:Will Mozilla ever gain outside contributors?

[umoto]

The suggestion to make object-oriented models for all of Mozilla was in a /. discussion a while ago and I took it upon myself to investigate the idea. Well, I did it, and I had to conclude that Mozilla is moving so fast that the models would be obsolete too quickly. I like the idea that the module owners would supply the models.

Even in my own programming I find that when working on a new project, it's extremely difficult to model the code except with the code itself. Until the classes are written, the ideas exist only in my head and can't be easily translated to visual information. By the time the ideas are concrete enough to visualize, the classes are already written and there's not much reason to concretely document anything but the interfaces and the less obvious sections of code. I have a friend who says some people are "visual" learners and some are not; well, I suppose many coders think in a non-visual way. To them, the addition of a GUI to an IDE is only helpful if it means fewer keystrokes. :)

Once Mozilla 1.0 is out, though, I think we'll start seeing a lot more OO models. Then it will be easier for all of us to tack on our own little mods.

Re:Idiot slashdot moderators

[CrosseyedPainless]

Why is this flamebait? It seems pretty correct to me, and I'm running 2.3.15 24-7 on my home machine. Netscape is showing its age in a big way, Mozilla is far from catching up, IE5 works fine (dare I say that on /.?) KFM has some holes, but like cameldrv says, it's stable and going somewhere.

So, where's the flamebait?

Moderators, a post doesn't become flamebait or a troll just because you don't agree with it. If it's on-topic and contains valid opinions and/or information, try to hold back on the downmoderating.

Re:68 Freakin 6

[zuvembi]

Please don't bloat browsers by stupidly including an unnecessary DNS cache

I think it needs a buffer anyway. Maybe just for this session say. It can throw away anything that's in their when you quit. Say a hash table that you prune at 128 entries. This should take up about 10-20 KB max of memory. The code really would be trivial. I can imagine you could do it in 50-150 lines of code (assuming you already have a hash class). And it would really be a boon for a lot of different people, and I don't think it would hurt anyone. And if it does, just make it a configurable cache like the memory cache for netscrape.

Mozilla? Quick/small/IEkiller?

[nitehorse]

I don't like saying this at ALL, but if the Mozilla team only releases binaries for the current platforms they support in the future, they will have NO way of making it onto my system. I don't run Windows, I don't have a Mac, and I don't use glibc2.1. So why am I excluded? Come on, I thought it was supposed to work *better* than IE5, and if it is (I hate Windoze/IE btw) then why is it larger, more flickery, and just about as stable? I realize it's only M9, but that's supposed to be like a halfway point. When projects reach a halfway point, aren't they supposed to have either 1) some of the features that were promised (e.g., full XML capability, HTML-4.0 compliance, CSS support, speed, small size) or 2) all of the features, with lots of bugs?

I notice that with Mozilla, at least, it's halfway between. It has some of the features, and lots of bugs, with no real reason for me to download. I can use Lynx, I can use Netscape (the piece of sh*t that it is) and I can use the KFM or the Konqueror (which is MUCH more unstable, but it is SO SWEET it's not even funny); and when Opera comes out, I swear I'm gonna bite the bullet, pay the fee (and no source!!) and use a browser which works with all the neat features I need/want from the Net. Mozilla was promising, but I don't even really like the UI. Ah well, maybe I'm anti-GTK; I don't know. I guess that's probably it. Although, if it supports full GTK themes, I might take a look.... some of the themes are DAMN cool and it would be almost acceptable. I don't like the GTK as far as stability goes (which is our key hand against the Windows crowd; take it away and what have we got? Lots of useless source code for instable programs....) but it can sure look like one sexy bitch. Interesting times, eh?

Re:Mozilla? Quick/small/IEkiller?

[warmi]

Stop lying !
IE5 is much much more stable than any version of netscape ever was ( specially Unix versions)

Try to be objective for once !!!

Re:Gnome libs required: won't work on Red Hat 5.1

[Spooks]

Potato works great. I use my computer to get work done all the time. Stuff breaks every once in a while, but not that often. I've been using it for about 8 months, and the only times I had trouble were when the JDK was compiled against the wrong version of GLIBC, and when I upgraded something that broke StarOffice for about 2 weeks. Other then that, it's been great, and is a wonder to use :)

Yes, a crash...

[John Campbell]

No, this is a crash, or, at least, a premature termination. It runs the installation wizard the first time through, then terminates. I was expecting that; I've seen that behaviour in older milestones. However, on the second and subsequent runs, it gets to the point in its startup process where it would normally open the browser window, and the program exits to the bash prompt without so much as a warning message. It never even opens any sort of X window. I went back and rmed my ~/.mozilla directory and let it go through the install process again, in case the old settings were confusing it somehow, with no luck.

M8 did the same thing to me... I think I skipped M7 for other reasons. M6 was the last one I had where apprunner worked.

That's the only serious bug I've run into with this version... I'm doing my browsing in the Mozilla viewer now, and it's working fine. And the form widgets work! (Can you tell I'm excited about that? ;) )

Pork Jet, a.k.a. M1[01]

[Herka]

This stuff is reproduced from a newsgroup posting on one of the groups found on news.mozilla.org:

M10 still exists as a target for features and specific fixes. ...There will be no 10 deliverable. M11 is our feature complete beta candidate. Once all M11 feature work is done, we are in the home stretch to get the bug list down and performance and stability up.

The project is called `porkjet'; I suppose that means they're making pigs fly.

Re:I think you're confusing shared with static

[Sinner]

So, then the windows DLLs *are* position independent, no? Which brings us back to the first question -- why are the linux binaries larger, and do symbol tables have *anything* to do with it?

No no no, Linux shared libraries use position independant code (PIC). This means it works wherever. Windows DLLs use position-dependant code. This means that if you get two DLLs that have the same base address (or overlap), one has to be automagically re-linked to use a different base address. The Linux way is less kludgy but the Windows way results in smaller libraries.

However, the major issue for Mozilla seems to be that the C++ code is producing excessively large symbol tables with gcc. I haven't been following this closely, so I may have the details wrong, but it seems a way has been found to get around this. I think this is only a disk space and download-time issue, and doesn't impact run-time memory consumption.

gcc doesn't produce very compact code at the best of times. Unfortunately, the gcc developers don't seem to consider this a priority.

Dynamically linked binaries considered harmful...

[Lazy Jones]

I wonder when people will finally realize that with the huge amount of different library versions installed on Linux systems, it's pointless to distribute dynamically linked binaries, unless you're a developer building a package as part of a distribution. The correct way to build distribution-independent, nearly hassle-free binaries is to make them statically linked (but even that doesn't protect you from the braindead libc incompatibilities due to configuration files in different locations).

Those who can...

[Anonymous Coward]

Those who can, do.

Those not smart enough for that, test and submit bug reports.

Those not smart enough for that, wait patiently while others work out the bugs.

Those not smart enough for that, sit back on Slashdot and bitch about how the quality of pre-release software isn't good enough for them, make meaningless comments about momentum and market share, etc.

Those not smart enough for that are dead, since they couldn't figure out how to breathe.

the ugly

[Freshman]

As of the 19th, the engineers found a butt ugly nasty memory leak that grew at the rate of 3.6 megs an hour. As far as I can tell they haven't fixed this yet.

Because the transition of M9 to M10 is a big one (finishing implementing all the features), there has been a whole slew of problems and "breakage" in the source tree.

Real nasty stuff, but hey, it's the home stretch.

grrrr...

[miahrogers]

to the mozilla crew:
what's up with only posting a linux bianary for RH6? that's really lame, now i have to wait and download 18 megs of source. Also, I would like some bzip2 zipps, my 28.8 modem works hard enough as it is, why don't we get the processor off rc5 and into some work for a change(I'm talking about my processor, as it's easier for the modem to download smaller bzips and have them take longer to uncompress)?

TEST

[Signal 11]

This is a test of the emergency broadcast system. Had this been an actual posting, real content would have followed this announcement. This is only a test...



--

NO Gnome libs required: Clue here

[tialaramex]

Those libraries are parts of GTK+ 1.2 (except the C++ library) GTK+ is not a "Gnome library" any more than Qt is a "KDE library"

The reason old builds worked and this one doesn't was because AS IT SAYS IN PLAIN ENGLISH this build is intended for glibc2.1 You presumably don't have glibc 2.1, or if you do, your GTK+ and C++ libraries aren't built against it. To try M9, get glibc2.1 and those libraries built for the new libc.

If you don't have GTK+ 1.2 at all, I have no idea how you expected most new GUI stuff to run, seems like half of Freshmeat is GTK+ apps these days.

How to help

[Anonymous Coward]

If you want to help the Mozilla project and are unable to or don't have the time to code, there's still a lot you can do that will be a great help and won't take up much time.

1. You should read the Getting Started document:
http://www.mozilla.org/get-involved.html
This will give you some ideas in how you can help.

2. The Gecko Bug a Thon is useful to help with:
http://www.mozilla.org/newlayout/bugathon.html

3. The status page keeps you up to date with current progress:
http://www.mozilla.org/status/

4. And of course stuff also appears on the homepage!
http://www.mozilla.org/

68 Freakin 6

[chocolateboy]

It's not like I'm running on an Amiga or a Casio E-100 or a Palm Pilot or a Sinclair QL or a Dreamcast or a yawn... I mean, this is a red-hot-one-year-ago (remember how things were a year ago... a month ago... last week... kids today) laptop. i586. MMX. Quake II. WindowMaker. vi. Capisce? 1024x768 pixels of uncut attitude. Every Milestone I have to break it to my baby: the Man from Mozilla says 'No'.

Laptop: Are you on crack? You're saying I have to put up with Netscape 4.07? That heinous piece of crap.

Me: Hey, watch your language, laptop. Don't make me hafta open a can of Lynx on your ass.

Laptop: Alright, alright. Couldn't you just download the source? Pretty please? Me: 20Mb on a 33.3k connection? Buddy, your F00F bug is showing.

Laptop: But Navigator hates Java. The slightest applet brings it to its knees.

Me: I know.

Laptop: Why the hell don't browsers cache DNS lookups? There's some kind of DNS locking going on that causes it to whiteout. On DNS for chrissakes! You look at top and it's got like 96% of the memory.

Me: I know.

Laptop: Try reading Slashdot on that baby. Good luck. Back doesn't mean back on planet Netscape.

Me: I know.

Laptop: Don't get me started on Javascript. Or CSS. Bloat. Speed. Key Bindings.

Me: I'm sure it will come. One day...

Laptop: Yeah. That'd be a freakin Milestone.

Re:I think you're confusing shared with static

[mattdm]

$ ls -l apprunner
-rwxrwxr-x 1 mattdm mattdm 58133 Aug 24 20:31 apprunner
$ strip apprunner
$ ls -l apprunner
-rwxrwxr-x 1 mattdm mattdm 25384 Aug 27 08:30 apprunner

$ ls -l viewer
-rwxrwxr-x 1 mattdm mattdm 190161 Aug 24 20:31 viewer
$ strip viewer
$ ls -l viewer
-rwxrwxr-x 1 mattdm mattdm 122068 Aug 27 08:30 viewer

--

Javascript

[QuantumG]

Someone was telling me they were putting Javascript support into lynx.. and I said "man.. that's futile.. you must be bored" and he was like "yer.. but it's almost working and it's REALLY REALLY slow.." so I said "why not contribute that to Mozzilla" and he was all like "yer.. I would.. but I'm not that bored". Next thing you know they'll have tables working correctly in lynx..