Slashdot Mirror
Mozilla M4 is Out
Greg Johnson
writes "Mozilla
Milestone 4 is out. I highly recomend you go snag it at
the Mozilla Ftp site. "
mozilla.org does not yet have a description of what exactly
is new-and-improved from M3, but what the heck.
← Back to Stories (view on slashdot.org)
That's Great!
That's Excellent!
That's.... what the hell is M4?
will it work on a i586?
:|
Identical results to M3 for me: Linux version loads a blank window; if you invoke it with a URL it crashes and dumps core. The M3 Win32 version at least loaded Slashdot and worked for a while.
But hey, maybe it's just me.
-demona
who only logs in at home
jeeze and I thought windows users were bad with their pointless web browser upgrades... here is a typicle conversation (rather onesided)..
"my neighbor says I should download the New IE5 [or communicator]"
shudder
ACK! Finally, someone who has the same problems as me! I like netscape (for "political" reasons) but I've been hearing some good things about IE 5.0 *shudder*! It seems that Netscape does NOT like me closing/exiting it and will reside as a running app unless I CTL-ALT-DEL and forcibly kill it! I really hope Mozilla really developes into something, or else, I might sell my browsing soul for IE, or wait until Opera has all the same features and just learn to live with it's (IMHO) unwieldly interface.
Respectfully,
Kevin Christie
kwchri@maila.wm.edu
Gtk-WARNING **: /usr/local/lib/libgdk_imlib.so.1: undefined symbol: gdk_root_parent :/
what do i need to get? i have gtk and glib 1.2.1
Hm.. well I don't see how it's fair to compare anything to the speed of Netscape Communicator 4.x, which is DOG slow P.O.S.
Yeah, I'm not too fond of the MicroSlut stuff either, but fair's fair.. IE blows the socks off Communicator as far as page rendering speed is concerned.
Gecko? Uhh. yeah, I'll believe it when I see it in a non-alpha/beta product.
Of course it munged this enter-comment page entirely and also wouldn't let me click the 'Submit' button so it obviously has a lot of work yet to be done, but I'm very impressed over even just a couple of weeks ago at how much more functional and stable it's behaving.
Pro: That zooming lizard thing looks much better than anything I've ever seen in its place (in the top right, showing you that the page is loading). Cons: They've implemented support pop-up ads :( COME ON! Implement the stuff I like first!
So I'm not the only one who can't find source to this project that's supposed to be Free Software...how lame... :( I'd really like to try it sometime...
- RF (dfelker@cnu.edu)
so, why are the other platforms ~3 megs, and the linux one double that? is the thing linked statically or something?
'./run-mozilla.sh &'
no?
package:$> du
3663
...
26992 .
package:$> strip viewer
package:$> strip apprunner
package:$> du
3663
...
20080 .
From the look of it I'm willing to guess not.
Even better: strip *.so
Strip everything that unzips, and the file size
shrinks to roughly 3 MB for the whole package
Debug info. You can do a:
...} ) The debug version should give better info in the case of crashes and errors, though.
find . -exec strip {} \; (from the package dir)
to strip all the binaries and libs. (Expect lots of error messages for non-shared-object files and there are better ways to do this so, anyone, please feel free to chime in with a correction {involving find -perm
-Steve
Crashes instantly on my Win/NT box. Sigh.
I think I read somewhere that (for win32) you can copy the prefs.hs accross to the current working directory (usually the one with the executables in this case) when you run applauncher and it (including proxy configs) is supposed to work.
:(
However, my milage has varied... proxies don't work for me... and so I can't get out of my department
i am also running Debian potato and glibc 2.1.1, and this is the first mozilla release that's worked on my linux box! sure, there's a bunch of bugs and it's pretty primitive, but the rest is downhill from here.
according to this:6
http://bugzilla.mozilla.org/show_bug.cgi?id=451
you might have installed JRE 1.2
Oh, did you pay for it?
were you paid to use it?
did you help to write it?
well, then, SHUTUP!
subject says it all
Guess what? It is pre-beta. It's even pre-alpha...
As someone else mentioned, strip the .so files in the components dir. viewer and apprunner are small compared to the components...
If you'd been following the development, you'd know that several hundred bugs have been fixed between M3 and M4. And many of the remaining bugs are related to missing features...
What sluggishness? On my machine it's faster than Communicator 4.51, _with_ debugging info in it.
Not only that, but if the GTK code is compiled with debugging, it does type checks on all casts etc. between GTK object types. That costs a lot in performance...
That was the point... sarcasm.
Having a robust implementation of XSL in Netscape Communicator 5.0 is going to be critical. Are Netscape and Sun really just going to sit back and hope an independent developer will dig them out of a hole for $30,000 ? I don't see MS doing this, they just got on with the job. Its attitudes like this that got Netscape into the mess they are in. I don't like MS but you have to give them credit, when they finally get it they push the resources into the problem until they get it right - not sit around waiting for the results of some vague competion.
I did the same scientific benchmark.
Netscape 4.51: 7 s.
Mozilla M4: 26 s.
This is a PII Linux 2.2.5 box. For some reason Mozilla is incredibly sluggish. Just moving the mouse over the back and reload buttons back and forth causes them to highlight with a delay of several seconds, and a 100% load.
--i;
So when are they going to offer KDE along with GTK binaries, so the other half of Linux developers and users will be interested?
Debugging symbols do not affect run time perfomance, because the symbols are simply not loaded into memory. The missing optimization is a slow down, though (and needed to make debugging 100% reliable). Furthermore, debugging symbols increase the binary size.
marcus.brinkmann@ruhr-uni-bochum.de
Does it take a KDE front-end to interest KDE users? They must be very shallow, you sure are
Is it faster than IE5 ?
It's not a STANDARD, that's why. Of course IE5 would support WebDAV -- it's basically just another Microsoft toy to dazzle the clueless masses.
Get W3C or ECMA to recognize it, then it'll have a place in Moz. Until then, there's no reason to.
PLEASE POST THIS LINK...
Okay, I'm a sucker. 1) Javascript is cool. HTML needs if then else and function calls. Javascript does it. 2) Sometimes you need more, like actual running programs. Yes, Java's slow; it's still practical and the speed problems will become less of an issue with the ever increasing CPU speeds and higher bandwidtch connections to the net. Java's secure, understandable, portable and the damn closest thing to "C" I ever saw. How else can you instantaeously distribute runnable programs to the world in a secure manner. In sum: Java Rocks!
Following the 80/20 rule, my guess is that a lot of the complaining about M4 comes from the 21% who write 1000 lines of code/yr.
As someone else has already said, milestone code is only slight above "Holy #$@! It actually compiled!" I think that parts of M4 need work (obviously), but overall it is a big step in the right direction.
gdk_imlib.so.1
This library comes from Raster's Imlib. The latest version appears to be imlib-1.9.4.t ar.gz
SegFault
"Those who desire to give up Freedom in order to gain Security, will
not have, nor do they deserve, either one." - Thomas Jefferson
From the Mozilla "Issues" list:
:)
> "javascript:" is not working.
> Vcards in mail are not fully functional and may cause failures when clicked.
> Java is not yet implemented.
Those all sound like features to me... can we keep them?
Posted by FascDot Killed My Previous Use:
The Win32 version is slow to work but quick to render. Only problem is, I can't configure Preferences so I can't get out my firewall.
Love the interface. The the order of magnitude size difference from IE5.
Posted by shaver@netscape.com:
Here's my offer: if someone writes a Qt-based front end, I will _personally_ check it in, and make Qt binaries available on the FTP site for the subsequent milestone.
The reasons for our switch to GTK from Motif have been discussed at length in the various mozilla newsgroups, so I'll not bore the literate by repeating them here, but you might find this interesting: virtually all of the current GTK front end was written by people outside Netscape (Stuart Parmenter, Chris Blizzard and others). There's no reason that people couldn't do the same thing with Qt.
Froze up totally on Slashdot. ;b I'm glad that the idea of platform parity is still being pushed hard.
The browser situation here doesn't look too good to me. NS 4.5 is too big/slow/memory-hoggish, IE 4.5 isn't much but an excuse for MS to delay as long as possible on Mac 5.0, and the more ideological Mac users (like me) won't use it anyway.
iCab still isn't really usable for everyday work--it renders as fast as Seamonkey does (when I can get SM to work), but everything else is way too slow, as if all the non-rendering parts had been done in RealBasic. (Scrolling is especially ugly.)
It's ugly and it's slow (well duh, debugging symbols and no optimization) but this is the first build in approximately forever that has worked for me. I was going to use the New and Improved build process to make my own but, well, it just *worked*. GtkStep makes it dump core, though.
Pretty cool... client side XSL would make this thing rule beyond all. Too bad I have no idea how to do that on the browser side.
Remember that what's inside of you doesn't matter because nobody can see it.
Rather than polished software like Windows 98...
Remember that what's inside of you doesn't matter because nobody can see it.
Reading through the list of "Known Issues" tells
me it's still very very very beta. I'm anxiously
awaiting for Mozilla to be closer to the
Communicator functionality, such as mail/news
client fully functional, menus and toolbars
enabled.
Suggestion:
Someone in the Windows crowd (which
doesn't include me anymore, thank God) should
whip out InstallSheild Express and make a nice
install package for the Windoze people -- they
seem to having some trouble installing and running
the Mozilla betas. Help take a big bite out of
IE market share!
...although at least the Windows people have
the Gecko-based NeoMagic browser to look forward
to soon.
All I can say is, it had damn well better work -- unlike M3.
find . -type f | xargs file | grep 'not stripped' | awk -F: '{print $1'} | xargs strip
--
W.A.S.T.E.
W.A.S.T.E.
It was "Javascript:" - not javascript. Try typing javascript: into the url box and you get a javascript console.
Matt. Want XML + Apache + Stylesheets? Get AxKit.
They've finally enabled expat as their XML parser, so it now dislays well formedness errors. Yippee. Finally a good XML development platform for Linux and Win32.
I'm now just hoping someone will develop an XSL system for Mozilla. Seems doubtful though...
Matt. Want XML + Apache + Stylesheets? Get AxKit.
XSL is indeed a styling language, but because XML is a way to show data instead of a way to make a page, XSL is a styling language which can change those datastructures.
E.g.:
If you have a piece of XML code, you could easily pull out the piece and use it in the part of your code.
Also see www.xml.org for more
-- signed for your pleasure --
Unfortunately, there isn't a source release, so
those of us running glibc2.1 won't be able to run
this. M3 compiled pretty cleanly, so I'm hoping
M4 won't be disappointing.
I just wish M3 wasn't so damn slow. Is it possible
to compile it without all the debugging code?
----------------- "I have a bone to pick, and a few to break." - Refused -------------------
How do you launch the windows version? I'm at work right now, so I downloaded the windows version. But I see a lot of .exe 's and no readme files and most of the .exe 's I click on are dos programs that just kill themselves or guis that don't seem to do anything and/or crash.
Joseph Elwell
BACK TO GOPHER!!!
I'm running a current Debian 2.2 (potato) wich uses glibc2.1 and it works fine. Er, no, it's buggy and slow but it works.
Its a shame that Mozilla 5.0 final wont support
(client side) WebDAV. IE5 does it today.
With DAV we could view/modify our files without
having to ftp them back and forth. Version control
is also a plus.
More info at www.webdav.org
"...I'm not one of those who think Bill Gates is the devil. I simply suspect that if Microsoft ever met up with the devil, it wouldn't need an interpreter." --
Who/where is the source of that quote? I love it!
sPh
I'm usomng it to edit this comment. It's faster rendering ut when I press the spacebar
it pops me to the BOTTOM of the page.. Wierd, it's also NOT word-wrapping! ARRRGH!
I am not merely a "consumer" or a "taxpayer". I am a Citizen of the State of Texas
I HATE it when I'm typing blind!
I am not merely a "consumer" or a "taxpayer". I am a Citizen of the State of Texas
Why? Because until recently, there hasn't been any browsers that properly supported CSS in any way shape or form. So what do you use to make things "look right" if you don't have CSS-1 at your disposal?
Tables.
So there's going to be a LOT of sites out there that don't use CSS and use tables- so you've got to get both working correctly or else you might as well have not bothered...
I am not merely a "consumer" or a "taxpayer". I am a Citizen of the State of Texas
Despite what I've heard dozens of people say, Mozilla isn't slow because it has debugging info compiled in. My builds with -O3 are still slow.
Even with -O6 -mpentium and anything related to debugging removed, including your printfs, it's still slow.
People are explaining away it's sluggishness because of that, and that's not the problem.
Stanislav Meduna posted a workaround on the linux-kernel mailing list the 23rd of January, look for the "nspipepatch".
It solved 95% of my problems (Communicator still "freezes" at times 'till I move the mouse, but it occurs much less often now; It still randomly crashes in java/javascript stuff, though).
debugging info != debugging code.
There's a lot of printf()s there. Watch your
console.
--Dan
This release even has a messenger under Linux that works, and the browser works for hte most part!
It's got some debug code in it so it's a little bit slower, but it's not that much slower than hte regular navigator, and it seems like it will be a much-needed improvement to Netscape.
This certainly brings my hopes up a lot for what Mozilla will turn out to be!
I'm sure some enterprising person who knows more than I do will remove pop-up "functionality" to make a nicer browser. I don't think I've ever seen a window pop-up with any useful content, so that's a "feature" that can be removed without adversely affecting anything.
you mean like neoplanet which is coming out really soon
---
i'm getting tired of all these blokes who complain that this has bugs. Of course it has bugs-it's pre-beta for crying out loud! It's as if none of you "linux developers" never saw pre-beta code before. So instead of complaining and whining let's be a little bit more constructive here, ok? Mozilla is everyone's browser now
---
btw I'm not knocking off the _real_ linux developers-just the linux wannabes who think they are just because they were able to download compile and play around with M4 which is only meant for real developers.
---
Yeap.. http://neoplanet works better
Stan "Myconid" Brinkerhoff
SB.
Couldn't have put it better myself!
--
.-.--
I can't get out either. Does anyone know how to configure it? I already tried manually editing the prefs file. Didn't work.
I'm amazed, except for small font disease, it actually works quite well. I'm not seeing any of the slowness others have mentioned, it's definitely faster than 4.5 and probably as fast, if not faster, than 3.0x.
Went to their page. Visualy, it looks great, but is there no Linux port?? =:(
yea i notied that too...does anyone know if gtkstep is any faster or slower than any other theme engine? i.e. the default one, redmond95, or are they all just based off the same thing?
-- adraken
Slow as molassas on my pII300(Win32). I really don't see it as "much faster"
Did I miss something?
apprunner.exe :)
----------
"They misunderestimated me." --George W Bush, Nov. 6, 2000
I've heard very little about XSL.
What are the benefits to including it?
----------
"They misunderestimated me." --George W Bush, Nov. 6, 2000
..but are still rooting for Netscape:
:)
Download M4 and the other builds. You have a chance to really impact a product by running it, and submitting bugs. Try and make it crash, it's fun
The whole point of these Milestones packaged with the FullCircle software in them is to get people to give some general feedback on it. It's essential to the development of the product.
Or you CAN just sit back and wait, doing nothing - Bill Gates will be smiling down on you.
----------
"They misunderestimated me." --George W Bush, Nov. 6, 2000
ftp://ftp.mozilla.org/pub/mozilla/releases/m4
Overall, while M4 is not a very fantastic release, it does show that the Mozilla project is still promising despite what some people may say in light of recent political problems. I say, three cheers for Mozilla, may you one day be quicker and more stable than MSIE!
The mozilla as you call it or the Gecko engine is a complete rewrite. The mozilla classic was discarded months ago. This isn't even alpha so get over it.
Suddenly, the hairy finger of a familiar monkey tapped me on the shoulder. It was time.--G. T.
M4= Milestone 4 of the Mozilla project. It means a semi-major update of the project, usually includes new features, is more stable, etc.
To quote a friend of mine, "it's just more better." =)
-- sudo.ca
when are they finally going to fix things? I mean... it seems to be they're dragging their ass (so to speak) and not fixing anything... the tables still SUCK hardcore, the UI still needs a LOT of touching up, and all it seems to me that they are doing is trying to get all the "features" of HTML into it (and the table stuff really pisses me off still)
now some of you will probably flame the crap out of me but why are they pounding so hard at new code and features when they should be fixing whats already there? load up slashdot on it - you'll see that the tables are nowhere near right for the titles... i loaded up www.loackergnome.com and it didn't even show me half of the page cause the tables couldnt load right.
if i were working there i beat the crap out of someone so that they'd fix what is the MOST important feature of HTML... without tables the web would look like crap (or everyone would make their pages as client-side image maps just so they'd look right)
8Complex
This is the first version of Mozilla I have actually tried. I'm impressed. It is quite stable and fast considering it has soooo much bebug code in it. Good work guys!
Romans 10:9-10
What do you mean Lynx doesn't have progressive rendering? Do you mean loading a page as it's being downloaded? This was actually added in lynx2.8.1. Check it out. It's _very_ nice.
And it is awfully buggy when IZtry to type in forms...
<^>_<(ô ô)>_<^>
Trying all manner of possibilities, I (still) can't seem to get it running through a proxy (necessary on this @#$!@ ISP).
(tried preferences.js, prefs.js, prefs50.js in various directories with no joy-joy feelings)
Wake me when it can. Then I'll be happy. (I suspect it's the same static initializer problem, but hey, do I _look_ like a C expert?)
--
The Internet is the Suppository of All Knowledge. You get it in the end.
My bad.
:\
I forgot one combination.
From the working Netscape installation, I copied the preferences.js file to the package directory (i.e. the directory where the apprunner/viewer executables are in.
Now, why didn't I try that first? Well... sleep deprivation perhaps, or that I let someone else's hint ruin a logical jump.
--
The Internet is the Suppository of All Knowledge. You get it in the end.
1) Launch apprunner.exe
:)
2) Read the release notes
Apprunner will launch the GUI and continue running in the background as a DOS box. Check out all the unresolved references as their logged
NeoPlant
"Bugs are harder to cope with than features, because they are less well defined and less well designed."
"It may be remarked in passing that success is an ugly thing. Men are deceived by its false resemblences to merit."
Actually, it's a more complex style sheet language, based on DSSSL. (I think I have the right number of S's in that!) The goal is to have both CSS and XSL available. CSS for simple formatting (since it's easier to use) and XSL for complex formatting of XML documents.
"Bugs are harder to cope with than features, because they are less well defined and less well designed."
"It may be remarked in passing that success is an ugly thing. Men are deceived by its false resemblences to merit."
Here are some interesting notes on what's included/bugs/etc...
s e-notes/m4.html
http://www.mozilla.org/projects/seamonkey/relea
æeee!
Hi.
Does the word "development" mean anything to you? Guess not. Gee, development, do you think it has bugs?
Don't even talk about the UI, because it's not even beta yet. You complain because it dosen't behave right on a complex page like slashdot, I'm amazed it runs slashdot at all (technically slashdot is a table abusive page, one day it should be done with CSS)
So yes, you're a Halfwit. Stop whining and contribute if you want to see improvments. I personally think the people who worked on this are code.demigods for getting a cross platform app like this working.
Oh, and as a note. You don't use tables for formatting, that is an old hack. Use CSS1, almost all browsers have support enough to do most of your formatting in it.
Slashdot. It's Not For Common Sense
Im running it fine on a 120 laptop, is a tad slow though, I'm not complaining. Cool, they changed the little netscape thingie to a lizard...:)
Release notes
If you have ever used or seen someone using Neoplanet you would realise that it is nothing more than an advertising tool. Animated ads constantly flashing in the peripheral and you have to use their start page.
The only thing Neoplanet offers is a customizable UI...for a cost.
The specification suggests that you do not use tables for your layout. That isn't what they were intended for. Do a little research and learn yo'self some CSS. It's much better at layout than tables (understatement) and much more advanced. One of the points of Gecko is that it will hopefully support CSS-1 in full.
Actually, yes, M3 and M4 both have large amounts of debugging code in them, and yes, this does slow them down tremendously. The rendering engine itself I believe is running pretty much at full speed though, so actual page loading is about what you'll see in the "final" product from what I understand.
Jeff
I just wonder if you have bothered to RTFRN or to run Mozilla from the daily builds to see the progress? I have been trying them for a while already and there is progress.
A month ago almost no web pages were rendered correclty. Banners were here and there, text running in wrong places etc. But after each and every new version it started rendering better. That's progress.
Also I'd like to point out that apprunner is really pre-code. People really haven't been working on it for long. Most work has been done to create the rendering engine and not for the bells and whistles in the user interface. Also remember that they are creating cross-platform code so it required a lot of planning and programming to even get this far.
I usually use the viewer-program. It's the one that has been used to develop the rendering engine. No bookmarks or anything but it has been pretty stable and fast for me. To get an idea of Mozilla's speed, go visit sites with a lot of tables and resize the window. On my PII it's about ten times faster than Communicatore 4.51.
So if you don't mind it crashing 10 times a day, rendering pages wrong and not giving you everything a finished product does, go and try it. But if you don't understand what M4 is, do find out before whining.
31337 = ELEET = elite
just like d00dz = DOODZ = dudes.
********* sig: If you don't like the law, get filthy stinking rich, and buy a better one.
Well, to be fair, your run-of-the-mill Netscape Communicator and IE aren't spouting debugging info. Isn't M4? I don't know, I haven't gotten it yet, but I think M3 did spout a lot, which would slow it down.
I take it back, there's no bebugging info. But it definitely seems faster than NS Com 4.5 to me. I would be posting with it, but it's form handling is not so good.
Deffinately faster.
Using 4.5 freshmeat.net causes the browser to hang for a few seconds as it renders all the tables and other stuff. By-the-seat-of-my-pants benchmark: 23 seconds.
Using M4 freshmeat.net pops up much sooner after the data comes in. By-the-seat-of-my-pants benchmark: 14 seconds.
M4 kicks ass, nuff said... can't wait till NeoPlanet puts out their version.
Sweeeeeheeeeheeeeeeet!
Hmm..
I already have that problem witn Netscape 4.5 anyhow, so M4 can't be much worse; Currently running with 4 browser windows, sometimes as much as 7... One of them will hose itself and Netscape will suck up 99% CPU cycles, or Netscape will slow to a crawl, and if I kill one of them, all of them lock up and have to be forcibly killed.
And sometimes Netscape will hose itself and none of the links will work, unless you right click and open in a new window, and then when you close all the netscape windows, a little dangly bit is left in memory, which you have to forcibly remove before you can run netscape successfully again...
AS
-AS
*Pikachu*
Me Too!!!
I'm currently running NT4.0SP4, but the system has been up for like a whole week, and thought the error may be just some little dangly bit that never erased itself properly and that the problem would go away after I rebooted, say this weekend.
I guess it may be a more serious problem, perhaps an NT specific compatibility issue?
I hope this post gets moderated up some so more NT users will see this...
In my case I get
"The instruction at "0x503371b0" referenced memory at "0x013751a8".
From M$ Developer Studio I get
"Unhandled exception in apprunner.exe(npjava32.dll)0xC0000005: Access violation"
Whatever that means...
AS
-AS
*Pikachu*
And this is part of the power of the Open Source model; free communication, open exchange of data...
I'll try the above, and see perhaps if some funky path problems perhaps may have screwed up Netscape or something...
I'm using Netscape 4.5, if that makes any difference...
AS
-AS
*Pikachu*
However, it is a big step up from M3 -- at least in the short time I've been here. I feel confident that something at least as good as IE4 will be out by 2000, which is more than I would've said after M3.
. . .
OK, that's enough -- back to IE for me. :^)
--
..!!in an intastella burst i am back to save the universe!!
Under Windows NT 4, I get a crash:
ToolTipText: apprunner.exe - Application Error
The instruction at "0x503371b0" referenced memory at "0x014e51a8". The memory could not be "written".
The dosbox with the debug messages shows:
nsComponentManager: Using components dir: C:\mozilla\bin\components
width was not set
height was not set
Reading file...
Reading file...Done
The Messenger component is available. Initializing...
Messenger has been bootstrapped!
The Composer component is available. Initializing...
Composer has been bootstrapped!
Reading file...
Reading file...Done
Reading file...
Reading file...Done
runtime error R6016
- not enough space for thread data
Is anyone else seeing this?
I'm trying to work out how to submit it as a bug report right now, but if nobody else sees it maybe it's just my system...
Cyberfox!
Greetings,
Exactly! NPJAVA32.DLL. That was the key eventually for me.
I got it... Same thing you did, went into MSDEV, and poked around. If you look at your debug window under MSDEV, you'll find that it's loading the DLLs from your Netscape directory instead of its own directory. I moved all the DLLs out of the normal Netscape directory (and the Plugins directory) to an 'ouch' directory under each location respectively, and re-ran apprunner, and it worked to some approximation of fine.
Unfortunately, I can't figure out for the life of me how to make Apprunner NOT look at the Netscape DLLs to run, which means I can't run Netscape on the same system as Mozilla, a Very Bad Thing(tm) if I want to test it and yet be able to really browse when I need to.
Damn...
Cyberfox!
I don't hang with d00dZ, so can somebody tell me
:)
what 31337 means?
PS. About to try on Mac. For those interested in whether it's slow or not,
the HTML test with all the tables:
IE 4.5: 1 second or so
NS 4.5: 10 seconds (!)
M3 : 1 second
so it's the same speed on tables as IE 4.5, but PLEASE get a release
Pope
It doesn't mean much now, it's built for the future.
On a PII-300 using Linux:
... BUT,
On the bad side, it was as slow as mollasses, the sidebar on apprunner couldn't be dragged out (it had to be nudged out), lots of sticky graphics clutter (have I said slow already?)
It hasn't crashed on me, yet; XUL works; GTK themes worked (I cycled throught 6 different GTK themes and it didn't crash the browser)
Would really love to use the different modules in my own programs but they're too complicated to learn!
mozilla = netscape 4.x + old bugs*2 + new bugs
That's from the release notes link above.
LOL! No thanks!
And what exactly is the problem?
Just wanted to remind everyone that this is not even considered a "beta test" release yet. M4 is "Development Milestone 4". This just means the Mozilla folks have reached a certain stage in their development of the program. A bit better then "Wow, it compiled!", but not by much. Do not expect production-quality code out of a development release, let alone beta.
It also appears the release notes page contains a list of problems reported. That does not make them universal. I suspect some things (e.g., JavaScript) work poorly for some, better for others. But that is pure speculation on my part.
dragonhawk@iname.microsoft.com
I do not like Microsoft. Remove them from my email address.
M4 works better than M3 did ... no crashes yet ... but re-draw is really ugly when dragging the side scroll bar up or down
Accually your problem is that you clicked somewhere on the page!
for some reason the arrow keys control the scroll bar and not the cursor, meaning if you clicked somewhere placing the cursor where you clicked when that cursor gets pushed off the page when scolling with the arrow keys it immediently jerks the page back to a point where the cursor is on the s5creen where you last left it.
Whats really freaky is when your cursor is in a input box like this using the up/down arrow keys goes up and down the text and up and down the scroll bar.. really freaky effect
The pop-up ads are done through basic html and javascript. If you removed the javascript command to create a new window with the no file menu/scroll bar/explorer stuff that would effectivly be deviating from the DOM pertaining to the creation on windows. While I abhor pop-ups like everyone does, when we start picking and choicing which standards to incorperate and which ones to leave out, we become no better than the guys creating proprietary standards in IE and Netscape. Now one thing that can be left out safly I'm sure if that onexit command or whatever its called that allows a site to dictate a final command when you leave a site. This command was created at the request of advertisers, and is generally only used to keep people from leaving a site unless you purposfully crash your browser. Though luckly very few sites, but the PrOn 31337 sites sink to this level.
This is not a pointless browser upgrade. This is following the phylosophy of release quickly and release often. And if you'd take the time to look at the release notes provided you'd see everything that has been done and what is planned to be done for the next milestone.
Go Gusties
apprunner.exe
I've had the same problem for ages. Some days it doesn't bite, some days it does.
Most often, I get the "dangly bit" error: most of Netscape works except linktext, and then all windows have to be closed and the dangly bit process killed off.
Three times today, even.
- The preferences dialog is almost completely nonfunctional. One would seem to be expected to muss about with the prefs50.js file to do typical things like set proxies, cache size, and so forth (*sigh*... I run that squid for a reason!) Seems like a lot of that interface was there in the last build I tried... But that was pre-gecko.
- Unlike as documented, the prefs file, cache, and other stuff that should be in
...\Netscape\users\username\ are actually created in the current directory. Scripting or shortcut foolishness must be resorted to. Worse than the last time I tried a mozilla build. - No, it's only pretending to hang. Blur and focus the window and it will wake up. This is an improvement.
- Shrinking a chromebar with its handle has somewhere between little and no effect. This behavior also seems more broken than the last mozilla I tried.
Some things I noticed but aren't really brokenness:- There's some kind of ugly spinning barbershop thing at the bottom of the window. Its purpose seems roughly equivalent to the load animation, except it's even uglier and more distracting.
- cellpadding is taken very (way too?) literally. Check out
/. under mozilla and marvel at the slashboxes. - The chrome feels really sluggish. In fact, the whole app feels sluggish except for the blazing rendering. This lends it a kind of IE feel, where pages render fast, but the app feels unresponsive.
- URLs aren't "corrected" in the location box. This is good in one way (it doesn't seem to want to take the focus away from me to do things to the widget -- an IE behavior I abhor) and in other ways bad (there can be bare hostnames in there with no protocol specified).
It's more functional than your average Amiga browser, but less functional than anything I would want to use on a regular basis. I ought to contribute something because it really feels like mozilla is making jack squat progress in terms of delivering a usable app, but I'm in no position to be picking up enormous projects, or parts of them. It's all I can do right now to keep up with my rinky-dink stuff!I guess it's back into the peanut gallery for me...
bad ass.
I'm running NT 4_and_SP3_with_Communicator4_and_noh ence_
crashes_(yet).
If_I_hit_the_spacebar_my_windows_scrolls_down,_
the_underscores_instead.
Not_bad_for_prebeta,_I_guess...
> I'm now just hoping someone will develop an XSL system for Mozilla. Seems doubtful though...
Don't count it...
" Sun will put up $30,000 for implementations of XSL to be added to the Mozilla.org open source effort, developing the source code to Netscape Communications' Communicator browser. This implementation would be a plug-in that would provide XSL formatting apabilities for the Mozilla browser and would fall under the Mozilla public license."
see: http://www.news.com/News/Item/0,4,33534,00.html
for more details...
I hate Javascript. I've had more of it than I can take.
I agree with the concept - something like it should exist. But in practice it is HORRIBLY overused. All too often I see javascript used again and again to perform functions that are exactly the same as what plain html can do - except it makes it harder to actually get the information from the site that you want. Javascript is obfuscation of information, and that's why I hate it. Gamasutra, for example, often makes you click on stupid Javascript popup windows just to get a picture in an article. What for? So that you HAVE TO view the site interactively, presumably so that they can shover adverts down your throat. I can't spider the sites with wget. In my country phone billing is done according to how long your phone call is - so for every second I'm online I'm getting charged - and it pisses me off that some lame site is forcing me to sit for hours online, clicking link after link after link just to get one silly little article. Just give me the damn *information* that I want, without all the damn frills and crap getting in the way. Ideally I want articles etc to be made available as zip or tarball; this is what I try do on my webpage.
Javascript should be used extremely sparingly, and only when necessary. The way companies use Javascript to artificially keep you clicking away at their sites while they shove ads down your throat goes against all my principles. The way amateur web designers obfuscate their web pages by throwing in as much "keWl stuff" like Javascript also annoys me.
Java by itself is alright, as long it is used sparingly and in a useful way, for example to demonstrate algorithms. Java doesn't screw over 'wget' either.
For an arbitrary example of a site made totally useless to me by all the ^$%#^$ frills, have a look at http://www.geocities.com/CapeCanaveral/9589/ , an electronics website. (Quote: "ActiveX - please use Internet Explorer".) And those damn geocities popup ads. Bah humbug.
Three or four years ago the web was pretty damn cool, and useful too. Now you can't go two hops without having ads pushed at you, and it's become increasingly difficult to wade through all the useless commercial blurbs when searching for real nuggets of info. The current move towards search engines that return links of companies that pay the search engine is going to kill the Web. It will become almost impossible to find information without being steered towards someone selling a related product.
It's looking pretty good, but things get real exciting if you try to scroll using the arrow keys. (At least under Linux) Doesn't draw the page correctly and scrolls uncontrollably. I like it though. It's going to be some excellent software when it's all finished.
blah grow a brain. in case you didn't know a lot of stuff is slow because of debugging symbols and no optimizations have been done, remember this is not even alpha.
Well, I have M3 installed right now, and it feels way faster than both Communicator 4.5 and IE 3. Lynx is faster, though, but lacks progressive rendering :-P
(those are the only browsers installed in this machine, which is an NT 4 box, PII 266, 64 Mb RAM)
Can't wait for M4 to download, but the bits are flowing slowly... I hope we don't get mozilla.org slashdotted this time.
No kiddin? Didn't know. Thanks for the tip, this is something really wanted to see.
I see someone pointed you at mozilla.org's ftp server already. But even if there's not an 'M4' source tarball, that only means it hasn't been packaged yet.
You can get the up-to-the-second source with CVS anytime you like. Afaik, there's an M4 branch you can use, if you don't want to get the code people is hacking at right now. Ask around in the mozilla newsgroups about this.
Something's broken in your NT (or Communicator, or Mozilla) installation. I'm running both Communicator 4.5 and Mozilla M4 at the same time here. No problem at all. The system is NT4, SP3 (soon these bozos will have to let me run linux in the office.. :-)
Anyway, I have no idea about what could be wrong. Sorry. I mean, is your Communicator directory in your path, or something? Are the Netscape DLLs in \WINNT?
Anyone know why the linux binary size if 2-3x the size of the win32? Static gtk?
You Like Science?
You Like bottomquark.
I think its CSS for XML, but don't quote me on that.
What's NeoPlanet, a code fork?
iCab is nice, but my largest problem with it at the moment is the lack of stylesheet support. As I've shifted to doing most of my pages using simple stylesheets, this just won't work for me...
eh, unfortunatly, i've used 5.0, its a lil buggy...
"The world only exists in your eyes. You can make it as big or as small as you want." - F Scott Fitzgerald
Well I built a Windows installer using my companies software. I'd like to post it but just want to make sure I'm not violating the license agreement (the legal document confused me). Do people want a link to this?
i can't ever get in to load on my machine. M3 worked at least, crashed after about two sites, but it worked.
i agree with matt, ie4.5 is from microsoft so i won't use it and comm4.5 is huge and slow, and iCab is far from ready.
maybe the mac guys need some help......