A Linux 'Browser War' in the Making?

We all know about Mozilla. It's one of the major 'poster children' for the entire Open Source concept. Build 9, Build 10, Build 11, and so on. Someday there will be a Mozilla that will run without crashing, and we'll love it. Someday. Then there's Opera's Linux - and BeOS, Mac, EPOC and OS/2 - ports. Over the last week we've had 100+ readers submit the news that (proprietary shareware) Opera for Linux is close to beta release. But there's another potentially exciting Linux browser in the works that has hardly gotten any publicity. (continued)

I learned about the new KDE browser project almost by accident. The concept is only a few months old, and active work on it only started a few days ago. Konqueror - "Konq" for short - is not the spotty KFM utility included in the current KDE release, but a whole new code base.

The people working on Konqueror are worried about getting too many people's hopes up too hard and fast; if they do, and if they run into Mozilla-type slowdowns, they'll end up with plenty of egg on their faces. All they're willing to show the world at this point is this screenshot.

It's amazing how far Konqueror has come in almost no time, especially when you realize that this is a purely volunteer project with just a few members, not a big deal with big money from a big company like AOL/Netscape behind it.

Are there other Linux browsers in the works? Good question; if you know of one, please tell us about it.

Another question: Would more volunteers help Konqueror? Perhaps, perhaps not; the KDE developers aren't sure that more bodies would necessarily help.

Should we all get behind Mozilla and push? Yet another good question - and one that's been hashed to death all over the place but hasn't been fully answered yet.

Whatever the answers, I believe most Slashdot readers agree on one thing: that a better Linux browser would be a Good Thing(tm).

We have a little poll about Linux browsers to the right of this story. And, as always, your thoughts on the subject are more than welcome.

Comments

Competition is good

[slk]

What the browser space really needs is competition, on all fronts. Companies and projects should be competing to make the fastest, most stable, most compliant browser with the best user interface. Mnemonic and several other free browser projects were effectively killed by Mozilla.

We shouldn't have all our browser eggs in one basket, any more than we should all be using the same operating system or text editor. Especially with open source browsers (but even with closed source ones), competition brings about innovation, as well as better code and, in the end, a better browser.

If one browser supports PNG, then they all will feel the need to support PNG. If another one is 100% compliant with the HTML 4.0 spec, then they all will feel the need to be compliatn. This competition is going to be the best thing that's happend to browsers, on any platform.

Personally, I look forward to trying out all the new browsers (konquerer, opera, and mozilla), as well as the old favorites (w3m, lynx, and netscape), and using whatever's best. Especially if it's open source, I'd also look forward to contributing bug fixes and new code. However, this means a relatively small and clean open source project, not that 120MB of C++ monstrosity called Mozilla.

Re:Competition is good

[Darth Maul]

However, what about the browsers that feel they can "win" by adding "new features" that are NOT HTML standards? That's what Netscape did to win over Mosaic, wasn't it? I don't think that was necessarily a good move. In the fight to "win", the browsers will not compete by "being more compliant". I can only see the opposite happening.

P.S. - this was posted through Mozilla. GoMozilla!

Re:Competition is good

[Anonymous Coward]

The reports of Mnemonic's death are greatly exzaggerated.

Mnemonic is alive and well. It's coming along nicely. The reasona the website has not been updated much is because people are too busy coding, and they do not want to generate premature hype, and only want to attract "serious" developers at this stage - the core architecture is still subject ot revision. If you check out the developer mailing list archives, you'll see they're quite busy, with SSL support about to be introduced into the main tree.

Mnemonic is, arguably, a better architecture than Mozilla, though simply less far along the development process.

Re:Competition is good

[richnut]

What browser companies need to do is to meld the browser with the UI, like (dare I say it) Microsoft does. You have to admit there's some value in being able to type c:\mp3 or http://slashdot.org into the same window in a MS operating system and receiving the results you want. KDE has an advantage with Konquerer in that they have already spent a good amount of time learning how to handle files and applications and all they need to do is drop in a few more components to meet the level that IE is at. Netscape/Mozilla dont have that. They're designed as standalone utilities not integrated components. The web browser has become the consistent pervasive UI for computers, and the KDE folks are looking to capitalize on it. People on /. will always seek out niche utilities to do what we want. People who are not hackers will be able to flock to something like a KDE/Browser two-headed monster.


-Rich

Re:Competition is good

[twit]

That's an excellent idea if you want to dump the entire unix philosophy of application design: one task, one application, one application, one task.

Admittedly, you might feel that a browser already does that (image viewer, html renderer, a plethora of plug-ins). However, that's not a reason to have a browser do more. In fact, that's an excellent reason to have a browser do less and do it better.

--

Re:Competition is good

[bmetzler]

umm if you try typing ///home/luke or file:/home/luke (or whatever the directory/file is,..) in netscape,.. (GASP).. whats this? Im looking at the directory... be it,.. in a ftp type style

And it's pretty simple to change how it displays the files too.

-Brent
--

Re:Competition is good

[twit]

Well, I can see a few things that gave Netscape the "advantage" over Mosaic (although I can hardly see a free application distributed by a government research facility as being subject to the normal rules of economic competition - it's not like NCSA gained or lost funding because of Mosaic).

Primarily, the ability to display inline jpg's and then the ability to render tables. These were both rather useful as extensions go.

On the other hand, Mosaic seems to have gained the decisive advantage in the browser wars at last; the NCSA codebase was licensed by Spyglass and then that was relicensed or bought by Microsoft. Most MS users use Mosaic without knowing it: Internet Explorer.

--

Re:Competition is good

[richnut]

That's an excellent idea if you want to dump the entire unix philosophy of application design: one task, one application, one application, one task.

Personally, I like UNIX. I like being able to hack my stuff. Part of the reason I like UNIX is it's always fun to guess which parts of the "UNIX philosophy" have been appiled to a package and which havent. :-)

I guess it depends on your definition of browser. To me a browser is a tool for viewing things. If it's source code, /., or pictures of my vacation it doesn't matter. It should be viewed by the browser because stuff is stuff (IMHO). To me that's consistent with the UNIX philosophy. Do you consider Emacs to be a part of the Unix philosophy?

Interesting argument....

-Rich

Re:Competition is good

[Daniel]

You have to admit there's some value in being able to type c:\mp3 or http://slashdot.org into the same window in a MS operating system and receiving the results you want.

Uhhh, no, I don't.

But if you like that sort of thing, I believe the Gnome mini-commander interprets URLs..

Daniel

Re:Competition is good

[twit]

Hmm. I don't know. I would consider the Unix philosophy of design to have two parts: modularity and interoperability. An application should be able to do one thing well, an application should be able to operate with other applications.

(No doubt Kernighan or Ritchie, if they read Slashdot, would consider this a gross oversimplification, but it's good enough for my purposes).

I would consider the bare bones of a browser to fit well within the unix paradigm: they implement a single standard, HTML, as defined by the W3C.

The ability to pair the application with other applications, for purposes to which the former application is not suited, as Konqueror is said to, satisfies the latter criterion.

Emacs, on the other hand, tends to be the kitchen sink - but, as I understand emacs, the core functionality is quite small when compared to that implemented by emacs users. I might be wrong - I'm not an emacs user - but an extensible framework is perfectly consistent.



--

Re:Competition is good

[Ed Avis]

You should be able to type in URLs at the command line, eg 'cat http://slashdot.org/'. That really would be cool.

For many years now there has been a set of kernel patches called userfs which allows you to write filesystems in user space. One of the examples was an 'ftpfs', allowing you to mount FTP sites as if they were NFS servers. I don't know if it has rotted away or whether there's any prospect of it getting into the main kernel.

Re:Competition is good

[twit]

Ah, the fabled red herring, second cousin to the well-battered straw man. A utility should do one thing, but not just one, atomic, thing. For example, there shouldn't be a million versions of ps, one for each option or combination of.

As for the unix philosophy or not: having modular and interoperable utilities is much more powerful than having one thing that does many things but works with nothing else.

More to the point, it's easier to develop something which does one thing well rather than everything and the kitchen sink. Everything and the kitchen sink applications usually suck heartily, no doubt because they're so hard to develop and maintain - I can't think of one that doesn't.

Perhaps that's what would attract someone now using Windows: the chance to discover instability and flakiness on another platform. As for me, no thanks.

--

Konqueror should use the Mozilla Layout Library

[hattig]

Is it just me, but the layout engine in Mozilla is pretty damned good by all accounts, it is the stuff surrounding it that cacks up all the time.

So wouldn't it make sense to use the Mozilla layout engine inside of Konqueror, and also to use that layout engine as a standard html widget for all of the different programs that display html to some extent? That way, all (bug hunting and fixing) resources will be focussed of one code base, instead of having loads and loads of different code bases around?

Or maybe it is just me being hopeful!

Just the idea of a standard libhtml widget would be great for Linux and other Unix variants. Why reinvent the wheel indeed!

Oh well... there will always be two or three competing things in the Linux world it seems (gtk vs. qt, KDE vs. Gnome, Mozilla vs. Konqueror, etc)... it is when they are merged that the trouble occurs... look at gcc.

Re:Konqueror should use the Mozilla Layout Library

[Christopher B. Brown]

That certainly sounds like a good idea; one not-so-minor problem; MLL uses the Mozilla Public License, which is not compatible with GPLed code as documented in the MPL FAQ:

18.How can GPL code be incorporated into the Communicator code base?

Under our reading of the GPL, it will not be possible to incorporate code covered by the GPL into the Communicator source code base. It is also not possible to use GPLed code and NPLed code together in a Larger Work. This is different for LGPL code. It is possible to create a larger work using LGPLed code that can then be used in conjunction with NPLed code through an API..

Re:Konqueror should use the Mozilla Layout Library

[HunterD]

Mozilla has an Embeddable browser. So really if Konq wanted to embed the browser it shouldn't conflict with any of the liscenses. All you would need to do is call the embeddable browser from a GPL program. The program at that point is in two distinct code bases.

For that matter, a program like neoplanet could be created for linux - it allows multiple rendering engines to be used (IE or Mozilla), but it is a seperate & distinct program.

If Neoplanet can incorperate mozilla's rendering engine without breaching the MPL - or releaseing the neoplanet browser under the MPL then there should really be no problem doing it with a GPL program as well.

Re:Konqueror should use the Mozilla Layout Library

[Ledge Kindred]

But doesn't that paragraph imply that Konqueror couldn't be folded into Mozilla, and not vice-versa? If the MPL really means "this is freely available software" then the layout engine should be able to be embedded into other code with little restriction.

-=-=-=-=-

Other browsers that work with linux

[scrutty]

I can think of

• mnemonic
• gzilla
• hotjava
• arena
• amaya
• lynx
• Star Office
• w3

Right off the top of my head. I am only too sure there are others.

Note that this list does not seek to compare or champion any of the browsers mentioned, a few of them are very sucky indeed. I just thought I'd mention them. Certainly in the linux world there hardly ever seems to be a lack of choice. Celebrate Diversity !

Re:Other browsers that work with linux

[msphil]

gzilla is now known as Armadillo, due, in part, to confusion with Mozilla.

There's also "Armadillo" (formerly gzilla)

[Booker]

Not to fan the widget wars or anything.... :)

There's another one out there - haven't played with it for a while, but I don't see it get much press, either. It's Armadillo - a GTK+ based browser written in C.

----

Zawinski's Law, Redux

[kzinti]

I think it was Jamie Zawinski who said that every application seeks to expand until it can read e-mail. I would add the corollary that the really bloated applications expand until they can browse HTML.

For example: there is Emacs/W3, which just released version 4.0. To quote the Freshmeat entry: Emacs/W3 is a full-featured web browser, written entirely in Emacs-Lisp, that supports all the bells and whistles you will find in use on the web today, including frames, tables, stylesheets, and much more. Emacs/W3

Now, I happen to use XEmacs. It's my favorite editor. I couldn't code without it, debug without it, or even read e-mail without it. But I can browse the web without it, and I think building an emacs-based browser is just way over the edge.

As the wise man said, though, Your Mileage May Vary.

--JT

Re:Zawinski's Law, Redux

[Zoltar]

I tried the latest version of Emacs/W3 a couple weeks ago with Emacs ( I'm not a big fan of XEmacs either) and (Insert-> I might be a moron disclaimer here) unless I did something wrong I found it to be the most horrible browser that I've ever seen. You should have seen the hatchet job it did while rendering slashdot. Again... I just did a simple configure, make, make install and then fired it up, I didn't try to look for a .rc file or anything to configure it, but it was nasty. Butt slow and nasty.

Someone mentioned the browser that ships with Star Office too. I would recommend that they use it for a couple weeks before they even mention it as a possibilty for anyone else. I tried to live with it for a couple days(just for fun) and found it to be pretty sad. Bloated slow and sad.

console web browser -- w3m

[vkulkarn]

For those of us who don't fancy a GUI, check out http://apps.freshmeat.net/homepage/928951047/ it's a console web browser that supports frames (in converts them into a table, and then renders the table)... it even supports SSL...

I have used the new BeOS browser and loved it.

[Pengo]

Well, if the Linux version of the browser is anything like the new BeOS browser expect the following.

1. A NON-MDI Browser!!
Yes, you got it. I have been long used to the Netscape/MS modal (each browser window apears to be its own running program/window).

2. Fast.
I was VERY surprised on how fast the browser rendered. Again, this is on the BeOS. Seems like everything on that OS is just damn fast.

3. Facelift on browser.
All of the buttons, graphics etc. have gotten a major facelift. The new browser really has a 'next-gen' feel to it.

4. More compatible.
I have gotten on different sites that gave my windows version of 3.6 problems, ran fine with the browser I played with.



5.. (And the strongest point in my opinion..)

IT IS SIMPLY A BROWSER... and it does that well!


If you want to use it to read your email , better have a hotmail account. I can't speak for anyone but myself.. but I -love- the idea of a lightweight browser that does nothing more than what its intended to do.


Honestly I am surprised that Opera has not moved into this market earlier.



But.. as with the good, must come the bad.


1. Crashed a few times.
Duh, its a beta.. and BeOS has not proven to be the most stable OS. (Though, crashes no LESS stable than Netscape..)

2. Nagware.
Yeah, won't kill me to pay for it.

3. Closed source..
(Doesn't bother me that much, I have been working hard to learn C++ .. and would love to contribute to an exciting project like this as a learning experience.. Mozilla (from the outside..) appears to be just a little to complicated to me.. oh well, if they don't want the eyes in their code. Their loss.


Anyways, all of the points where based off what I saw of the BeOS release of the next gen Opera browser. Hopefully I will be pleasantly surprised with the linux beta.

Also, this was not intended as a MOZILLA vs OPERA flame bait.. etc.

Cheers!

My thouhts

[killmeplease]

The real solution to the browser wars is to make all browsers run around a java renderer. It makes no sense reinventing the whel for every platform, as a lot of the code requires poring, ifdefs and the like. A Java browser solves many problems and would be feasable with the JDK 1.2.2 release from Blackdown in the future, and the JIT compiler from Inprise. This way, the only upgrade to your browser you would need is a download of the new JDK from sun or Blackdown. Writing a functional browser in Java is very simple, and can be done in a few days by a skilled Java programmer. With this sort of system, we could worry less about who builds the better renderer. "Opera loads each page, .1% faster", "mozilla loads pages with applets .5 seconds faster" would be heard no more. If this was a sucessful mission, programmers could focus on actually putting something interesting into your browser. Like a page that went to slashdot in the morning for you and stripped away the articles and ads that do not fit your interest, or did other cool things.

Live Free or Die Trying

Re:My thouhts

[jilles]

I'd agree with you if it weren't for one thing: java's graphics performance is very poor. If you look at Hotjava for instance you see that it loads and parses pages nice and fast. It even manages to render them within a reasonable amount of time. But when you start scrolling down you notice that it's a bit slow with updating the screen. Smooth scrolling like ie does is impossible with Java at this moment (I exepect this to improve in future JDK's).

Otherwise Java would probably be a very nice choice:
- it is very suitable for parsing (just check out the XML parsers for Java, they're probably the best (speed & quality) parsers you can find at this moment)
- Porting is nearly trivial to platforms with a good java implementation. Unfortunately linux, apple and beos all have pretty lousy support for Java at this moment (in terms of performance, compatibility and stability).
- a great deal has been invested in allowing java programs to integrate with the native platform without sacrificing platform independence (drag & drop, printing, GUI, cut/copy/paste, sound, 3d, ...). Probably it is really hard to duplicate this effort in a C/C++ project.
- The java API provides many things that can be reused in the browser.

The main arguments in favor of a Java browser are:
- reuse of lots of existing stuff
- platform independence
- development speed
- easy integration with different platforms

The main arguments against its usage would be:
- poor graphics performance
- lack of good Java implementations on some platforms
- high memory usage, Java programs tend to use a lot of memory (despite the small code base)

it's really hard

[arielb]

It's hard enough to support HTML 4, CSS 1 and 2, XML, DOM etc. But a real browser also has to emulate the bugs on IE and Netscape too. The only browser that comes close is Opera and it still chokes on many popular websites. Therefore I agree with the layout engine "gold standard" idea. The work should be getting CSS 2 and CSS 3, emulating IE/NS proprietary stuff -not doing the whole thing from scratch. Because unless you want a help file browser, you're not going to browse the real web.

There are even more browsers

[Fufie]

hi,

Actually there are a couple of other web-browsers out there, and some of them are written in Java. IMHO the best of these is currently the ICEBrowser by ICEsoft at http://www.icesoft.com/. What I like best about it is that it is actually quite fast, despite being written in Java. And as for following standards it is pretty up-to-date and comparable to Netscape, Mozilla and Internet Explorer.

Some not-so-nice things though, it is a commercial product and not a free browser at the moment. The ICEsoft focus is also on making an embeddable browser for other products and as such doesn't have a glossy interface.

It is currently in (downloadable) beta right now and is supposedly going to be released first thing in december which might be a place to stop by then..

But still it is an important browsing-alternative for those keen on Java.

Other browsers...

[lar3ry]

There are lots of other browsers, if one just looks.

There's W3C's Amaya, and their orphaned Arena browser. Sun has HotJava. There's Netscape and Mozilla (of course). Lynx. Mosaic is still around. I remember something called "dozer" (or was that an HTML editor?) as well. Star Office also contains a browser within it.

There are probably a few projects that I haven't heard about.

It makes sense for KDE and Gnome to incorporate browser technology into their desktop environments.

What would be bad for everybody would be if every one of these started to extend HTML with their own proprietary features (Netscape's CENTER tag comes immediately to mind, but there are plenty of other culprits here).

If I can sit down at any of these browsers, and once I figure out how navigation works (click on links for graphical browser, TAB and ENTER for lynx, etc.), I should be able to surf on any and all of these. And all pages should render as best as possible on every one of these browsers.

This means that JVM's need to be standardized, JavaScript implementations need to be compliant to some standard, or else we will be in a tower of Babel where everybody can talk, but nobody can hear what anybody else is saying.

To see what I'm talking about, just browse some of the comp.infosystems.www.* newsgroups, and listen to all the complaints about how Netscape and IE disagree on the rendering of this-and-that, and multiply it by the sheer number of browsers that are still coming out.

I welcome all browsers, even newcomers. But this isn't 1993... there are standards that they are expected to adhere to.

I know the complaints...

HTML (what version? 3.2, 4.0? hmmm?). CSS (1 or 2?). JavaScript. Java (1.02? 1.1.x? 1.1.x + Swing? 1.2?). HTTP (1.0? 1.1?). You name it.

Well, the best thing would be to support as many of these as you can; usually the newer versions are backward compatible. If not, there is usually some way to specify which version something is written to; support as many and as much as you can.

This is a tall order for a web browser nowadays, and the weight of these requirements has been very apparent in the Mozilla development. But your users will love you for it.
--

Why Mozilla is so badly broken

[Otto]

They're trying to support too much.

Mozilla has support for everything. Everything from email to the kitchen sink is part of the codebase.

It's gotten so badly bloated now that I hesitate to call it a browser anymore. All I want my browser to do is to display web pages, run some java/javascript, and support plugins for objects in a page. That's it. No more, please.

Look at the mozilla modules list:
-E-mail/news? I'd really prefer that to be a separate program that can open my web browser if needed, thanks.
-Dialup? I've already set that up, thanks again.
-Embeddable Web Browser? What the heck are you embedding that in?
-HTML to Text/PostScript Translation? Wouldn't this really be better as a separate program? How about saving using HTML? Simpler, eh?
-PerlConnect (Perl and JavaScript connection )? What the heck is this for?

And all the other stuff there. It just seems to me that they're not developing a browser anymore, they're developing an application suite.

Start with a simple architecture, and work up from there. Ahhhhh, would be nice.


---

Re:Why Mozilla is so badly broken

[Chris Siegler]

E-mail/news?

From a recent /. poll, over twenty percent of /. users use communicator for their mail client. Now what do you think that is for the less knowledgable browsing public?

HTML to Text/PostScript Translation?

I convert to text all the time, and conversion to PS is nice when there are many embedded images and you would rather not fetch them all with wget.

Embeddable Web Browser?

If you mean the GTK+ mozilla widget, it's small and useful and was contributed code.

Re:Why Mozilla is so badly broken

[Anonymous Coward]

Maybe you should look at the mozilla code and/or newsgroups before whether you decide if the mozilla team is forward thinking.

Mozilla is a very heavily Object oriented project which is very modularized.

Most people who actually know anything about the Mozilla project look upon 'bloat' differently than idle commentators such as seen often in slashdot.

The fact is that the market requires certain features from their 'internet experience'. This includes mail/news, Postscript output, etc.

Communicator was built as a monolothic application which had all these modules inextricably linked in one binary. In mozilla, these modules are seperated. If you want, you could rip out the mail/news DLL, and you would not have mail support any more.

I don't understand what you consider 'bloat'. You don't like large executables? Fine, we split it up into multiple DLLs. You don't like large download time? Fine, we make the distribution download size a fifth of what is was previously.

Maybe you're just whining that you think we'd get a better product quicker if you had all the mail/news engineers working on core browser stuff? Well, anyone who's done any s/w engineering knows that you can't just throw more engineers at a project to get it done quicker/better.

Mail/news is an application that sits on top of the new mozilla framework. If mozilla was just a browser, it could not compete with IE. The core of Mozilla is a framework for building applications such as a web browser, or mail client.

The importance of an HTML mail client cannot be underestimated in a corporate environment. Many millions of seats of Communicator and outlook have been sold to corporations, and they love and demand HTML mail.

So, we must provide and HTML mail client. Do we make them download it seperately? What do you do about the shared components such as core layout? do we make people download it twice?

Here's some real numbers from a recent mozilla status report:

Footprint
Estimate of compressed total download size on Win32: 1266k

Uncompressed DLL sizes:
Win32: 1174k, Mac: 2382k, Linux: 2331k


Now compare to Microsoft IE 5.

Minimum installation (without Outlook) is about 50MB. With Outlook, you're talking about another 20MB.

THAT'S bloat. We can get mail/news for around (guess) 500kb compressed download.


My main point is that adding these features don't significantly impact the size of the project. If you don't want a feature, remove the DLL. But the mozilla team know what the market demands. Just because you don't want something in the browser doesn't mean nobody else does.

Re:Why Mozilla is so badly broken

[Otto]

From a recent /. poll, over twenty percent of /. users use communicator for their mail client. Now what do you think that is for the less knowledgable browsing public?

Adding up the rest, that's 77% that do NOT use a built-in e-mail client.

Okay, I admit, I didn't think about conversion to PS as also being known as "printing", but this is because I don't use PostScript in any way, shape, or form, so it didn't occur to me.

If they had built the system as being a basic browser with advanced plug-in capabilities, they'd probably be further along in the project. Build the base, then build the addons.

For shits and grins today, I downloaded the latest version I could find at mozilla.org, which was M10. It works, to an extent. It'll display a web page, but that's about all I could get it to do without crashing.

Frankly, this does not look like quality code. I haven't downloaded the source yet to have a look, however.

When I'm building software, the first thing I prefer to do is to get some form of functional state. Preferably one that doesn't crash. Then add features as needed.

For a web browser, I agree that it seems like an object model is a good idea. Build your basic display code, then make everything else a plugin to that code. Perhaps this is what is meant by embeddable web browser, I don't know. All I know is that it's been well over a year, and the code crashes when I click a menu option. (Yes, don't click that menu, I know :-) Even debug level code shouldn't crash at the slightest provocation...

Hrmph.. I'll reserve judgement until I see a finished product, of course, but the outlook looks grim to me. Naturally, I hope they prove me wrong, because I want a good browser, damnit.

---

Re:Why Mozilla is so badly broken

[Daniel]

(a) I (and many other people, I suspect) don't particularly care what Netscape Corporation thinks "the market" demands, I just want a working way to view Web pages: ie, HTML and image rendering.
(b) Clearly, adding more developers screws things up..that explains why my kernel crashes every few minutes..er..wait.. [1]
(c) I'm not sure what Mozilla you're talking about, but M9 takes 13007K of
disk space on my computer and appears to eat even MORE memory and CPU than
Communicator.

Daniel

[1] Actually, I'm willing to concede this point, just because I'm probably an ignorant moron in this area :)

Re:Why Mozilla is so badly broken

[extrasolar]

(a) I (and many other people, I suspect) don't particularly care what Netscape Corporation thinks "the market" demands, I just want a working way to view Web pages: ie, HTML and image rendering.

You contradict yourself! YOU are part of the market! Thinking that way is a ME ME ME attitude and can't be healthy.

***Beginning*of*Signiture***
Linux? That's GNU/Linux to you mister!

Re:Why Mozilla is so badly broken

[Daniel]

Err, you missed the irony in that comment, didn't you? In any event, I don't see how I can be part of "the market" for something I don't want; if I *am*, which seems to be the position Netscape is taking, I think they are very very out of touch with reality. This is an approximation to my point, but having to explain a succint statement never succeeds ;-)

Daniel

Re:Why Mozilla is so badly broken

[Otto]

Are you saying that with a straight face? What a pompous ass you are. It is a very complex project. You have no right to diss these people when they have clearly stated that this is not even a beta release! Do you really think you could come up to speed with the code that fast to make a judgement as to it's quality?

What a pompous ass you are for telling another person what they have the RIGHT to say. I have every damn right to say any damn thing I damn well please. Damnit. :)

All I'm saying, in any case, is that after a year, ANY code *should* run without crashing so easily. That's it. Since it doesn't, IT DOES NOT LOOK LIKE IT'S QUALITY CODE. Okay? Deal with it.

---

IE for linux

[_martini_]

we'll just have to ask microsoft to make an IE client for linux. We'll just have to tell them it's for "winlinux", I'm sure they'll understand.

My personal opinion is that the mozilla project needs to focus more on fixing the browser before they add things like a html editor/news/mail reader. but since they've already added those things...mozilla should enforce a feature freeze on the current milestone.

but thats just my opinion

Once there was Mnemonic

[heroine]

When the Gimp 1.0 came out in early 1998 that spurred a frenzy of students to produce open source applications to equal The Gimp both in size and complexity, AudioTechQue, AbiWord, GnoMoney, Gnuotes, etc. One of these mega apps was Mnemonic, intended to be the world's first completely open source web browser. Mnemonic was mentioned on Slashdot every day and it was even predicted to kill Internet Explorer 4. Well a lot changes in 2 years. Hardly anyone using Linux today has ever heard of Mnemonic and most of those Gimp spinoffs have either died off or slowed to geological rates of progression.

Re:Once there was Mnemonic

[NettRom]

Hardly anyone using Linux today has ever heard of Mnemonic...

I think that's sad. Partly of course because Mnemonic has a derivative of my C code in it, and partly because I believe in the availability of choice (I guess this link should say enough about just that).

I don't think everyone should push behind one browser, at least not all of the time. Push to get it released, yes, but not push to make it become a "standard" of it's own. Netscape used to be the "standard" browser, now it's IE4/5. The web designers/authors follow that "standard" and create content looking its best in that browser. How it looks in other browsers they often do not care much about.

And I think that's sad. Sad because people seem to like it that way, and they end up using the browser the creators use. Secondly because it leaves me with less options. I recently had a harddrive crash, and my old P75 isn't too keen on running X. Mail & news is done with emacs, so I had no trouble with that, but when I wanted to check out some web sites for my daily doze of news: forget it! There's no good structural markup so Lynx is badly choking.

I surely hope that Opera releases the text browser they mentioned before, so that Lynx gets good competition. From the screenshot it looked like a really nice browser, and a useful one for when I only have console access to my system.

As others have mentioned already, why not have a browser that simply does one job, and does it well. For me that's a good idea. Others might want other things, and that's what we have helper applications, plug-ins and the like for. They can add to the already existing platform and create the browser of their choice (add c00l skinz and great sound effects for extended pleasure).

I'm all for choice. Mozilla might become my preferred browser when it's out due to its support for standards, and the fact that it so far looks to be a fast browser. Or I might choose Opera 4 (for Windows) since 3.6 is a fast and standards compliant browser, so therefore v4 should become another fast and standards compliant browser. Maybe I'll end up on a Mac with iCab... who knows... but I want the options.

Re:Once there was Mnemonic

[kas]

As one of the developers of Mnemonic, let me make a few remarks about the status of the project.

As with any project (including KDE's browser and Mozilla), Mnemonic has gone through several design stages and we have thrown away the entire codebase several times now. It's not very strange that only very few people know about it, since none of those times did we ever reach a point were the program did anything remotely useful.

At the moment, things are, however, progressing very rapidly. Apart from a very small core library (some 230 Kb), everything from network protocols to rendering engines is in separate modules. There is a rather decent generic XML parser, a completely GUI-toolkit independent rendering engine, network protocol modules (with SSL being added at this very moment) and a GTK based layer on top of that.

Just as the Konqueror team kept things relatively quiet, we have decided to work out the basics without giving too much publicity (although of course the code has always been available for anyone to look at). But as soon as the table layout algorithms have been debugged, we will start to release binaries (as that is probably the first time that the browser is really useful).

For more discussion, please join the mailing list; more info and status updates can be found at http://www.mnemonic.org.

Re:Once there was Mnemonic

[NettRom]

thanks, will try out. I don't read Japanese though, so the English version is preferred. :)

red baron

[fishbowl]

What happened to the Red Baron browser that was
bundled with RedHat 4.x? I can't believe that was just a one-off and abandoned. It worked great.
Why wasn't that open source, Redhat? It would
probably be pretty far along by now if only it
were out here.

Re:The existing KFM browser is actually pretty goo

[platypus]

I agree.
I wondered all the time when anybody will openly say that konqueror is a competition to the upcoming mozilla.
I used it most the time and actually have posted something on slashdot with it (and 1.1.2 does the cookies reliable).
One has to remember that a full slashdot commentary site (esp. with moderator functionality) is a real good hardcore-test for a html renderer.

Mozilla Lite?

[DataGrok]

I am not a C++ coder. Yet. And I know I'm totally oversimplifying the situation in my quesiton. With that in mind, maybe someone more enlightened than I could answer me this: Could it be possible for an industrious group of coders to grab the mozilla layout engine and wrap a simplistic UI around it, effectively creating what I see many people here asking for? ... A lightweight-but-functional browser-only browser, minus all the crap?

I used to be a big fan of the Mozilla project, but every screenshot that I see, I end up saying, "What is that mess over in that sidebar there? I don't want that. Can't they just finish the friggin renderer?"

Re:Why you're wrong

[asa]

Mozilla starts with a simple architecture. It uses a light-weight, quick as lightning html rendering engine. Add xml support, javascript and their blossoming child XUL and you've got the beginnings of a great product. Add on to that a simple mail/news interface (btw, the browser does not rely on mail/news code, mail/news code is pretty lightweight and sits atop the browser)and you've got a very functional product. Add an open java interface so you can plug in the version you like and it's really starting to look sharp (but still simple). Read a little more before you judge mozilla. Not all modules are part of the default product and not all modules that are part of the product are as big as the name would suggest.

Not just a web browser - and more screenshots

[Quigley]

Konqueror isn't just a web browser, konq is what Microsoft's IE intergration/COM should be. Using the new Canossa(sp?) component API, I believe Konqueror is able to browse and/or render filesystems, samba shares, web pages, postscript, dvi, pdf, and plaintext. Given the rapid pace of development there's probably even one or two more by now.

What this topic should be is "khtml", KDE's html widget code which in this case is acting as a Konqueror component. Here's some more screenshots, displaying its current rendering ability...

Konqueror displaying various webpages

Konqueror itself

Konqueror displaying postscript

We Need Netscape

[bjb]

I'm sure some people won't agree with this, but I strongly believe that we need to get Mozilla out soon so that Microsoft won't have an opportunity to 'standardize' the internet. What I mean is that if most of the general population uses Internet Explorer long enough, web developers will start developing sites that work ONLY with IE. This hasn't quite happened yet, but with some more time and some more 'features' put out with IE, this most likely will happen.

What we need is competition in the browser market to prevent this and to ensure that standards, not companies, rule the internet's content.

This is why we need Mozilla. It is Netscape. People know who Netscape is. Netscape might be in a strange state right now, but they are the underdog and people did use them at one point. Now I'm not bashing KDE or Opera (I use both browsers), but telling web designers that their new IE feature won't work under those browsers probably won't hold much weight (I hope I'm wrong there). Netscape, on the other hand, will get them to think about it.

I guess it comes down to the number of people who use a particular browser. There is still a significant number of people using Netscape, so you can argue that you need to support that browser. I just hope we can see Mozilla in force before its too late.

--

Re:We Need Netscape

[Gleepy]

I don't think we need to worry too much about "Microsoft-only" web sites. There's a big problem with them. Those "Microsoft-only" features don't mean squat to the the blind or visually impaired. As long as there is an Americans With Disabilities Act and a (U.S.) Federal requirement that their web sites remain accessible, there still will be pages renderable in a meaningful way.
Besides, Europeans don't want to waste time downloading silly and slow-rendering graphics when their phone service is being metered.
We still need to be vigilant when dealing with web browsers and company-issued "standards".
--

Re:We Need Netscape

[rhdwdg]

OK, try taking Netscape 4.x with all the modern features enabled over to www.toysrus.com and see what you get. The style sheet support is so broken that it can't recover from the designer's error. IE can, and that makes toysrus.com an IE-only site. This came up precisely *because* they are trying to use HTML 4.0 and CSS1 correctly to create a site that will work anywhere, with a GUI, a braille screen, with no more bandwidth used than needed.

Now if they would put the style sheet in where they say it is, it might work with Navigator too. But they would probably find Navigator's CSS support so awful that they'd rather have it look good under IE and not work with Netscape than try to find the microscopic middle ground where both major players support the same parts of CSS1 in their 4.x browsers. If you can get through, try the site in IE. It's fabulous work. I doubt they care that they are unwittingly helping Microsoft and ultimately hurting themselves; MSIE is simply the best way to present their e-commerce front end, period.

If Netscape Navigator 5.0 -- the finished, packaged product they can link to on NetCenter and have 50 million people install with a few clicks -- does not appear very soon, I am of the opinion that we will lose the web to Microsoft. And then we can almost give up. If Mozilla makes it, however, then we will be free to use whatever niche browser we want, because MS will submit to standards.

The more I think about it, the more I agree with the editorial featured here a couple weeks ago. Sorry, I couldn't find it through search.pl.

Re:We Need Netscape

[tgd]

A point to remember is most people still say they use Netscape when they're talking about their ISP, even if they're running MSIE.

Even stranger I know pleanty of people who if you ask them what web browser they're using, they say Internet Explorer, and if you ask how they get on the internet, they say Netscape.

You know what I want in a browser?

[Greyfox]

I want a browser with a nice fast rendering engine, capable of handling XML. I want modular language plug-in support, so I can run not only Java from web pages but also perl, scheme and python if I have the right language modules configured. I want a browser that I can point to a .tar.gz or a .zip file on a news server or a FTP site and have it automatically give me a list of files in the archive and display an index.html if there's one at the top level (Allowing entire HTML trees to be posted in one archive on news or ftp sites.) I want a browser that sticks to the HTML, XML and CSS standards defined on the W3C sites. I want a browser that lets me enable or disable cookies and modular languages on a site-by-site basis with include-all or exclude-all default policies being possible.

I do not want a browser that also reads news or mail. I don't want a browser that takes up 40 MB of RAM as soon as I start it up. I do not want a browser that makes proprietary features part of the main browser.

I think that's pretty reasonable, don't you?

Crashes...

[schon]

It can render Slashdot just as nicely as that screenshot, and I've yet to see it crash

It crashed on me last night. Twice. (using FTP.)

And the really bad thing is that after the second time, I couldn't launch it any more.. I had to log out and log back in again. (This is the latest version.)

Please note that I'm not complaining, I'm just stating a fact... KDE has a lot of good things going for it, but I wouldn't count it's stability as 100% yet (although it's good enough to be extremely usable at this point.)

(For those interested, I was using it because of it's convenience - editing HTML pages with KWrite - it's MUCH easier to deag&drop to load/edit/save than it is to manually use an FTP client to download them to a local directory, then do your editing, then upload them again.. and this is the key reason why I will continue to use KDE.. and it can only get better...)

I'm writing this from win32 opera right now

[dennisp]

While I like the ease in which I can increase max server connections to speed loading, I think I'll stick to IE (and netscape when I'm in X) for now. I tried loading a number of sites and it chokes on spacing, images behind text, no underlined links, no onmouseover underlines or highlights et al. The download speed in kB/s is nice though. Won't be switching anytime soon.
----------

Re:GPL'ed browser

[PigleT]

I generally agree with that.
I think a GPL'd browser (what's Konqueror's license like, and in KDE2?) would be a good thing, especially if it's as rock-solid as Konqueror looks at the moment (I've used 1.1.2 fairly extensively, and only found the lack of javascript a minor niggle).

To be honest, mozilla annoys me - I'm one of those who finds mozilla too slowly developed and unstable, and netscrape is getting too heavy. I can't say I'd be interested in an expensiveware browser for linux like Opera might well be, either. So Konqueror for me, whenever possible :)

Lynx? w3m is yet another GUIs-are-4-wimps browser, too :)

The REAL Law of Zawinski

[IGnatius T Foobar]

Actually, the real law of Zawinski goes something like this:

Any software package touched by Jamie Zawinski will get f**ked up until it no longer works properly, then he will make a big public show out of bailing out of the project.

Sorry, I just don't have any respect for Jamie Zawinski. He is personally responsible for a lot of the brain damage in XEmacs and Netscape, and he walked out of Netscape like a big crybaby when things weren't going his way. This person is not an asset to the community. He should go work for Microsoft.

Re: The REAL Law of Zawinski

[Pyro P]

First, JWZ's own software, i.e. xscreensaver, xkeycaps, and other things, are not "f**ked up". In fact, i consider them quite good. Additionally, you have given no real proof that JWZ caused the braindamage in stated packages. If you have any, please show it.

Second, we could have done without that gratuitous piece of MS-bashing at the end of your post. I'm not defending MS. But MS bashing is counter-productive.

Status of Konqueror and the need of helping hand

[boloni]

I really like that there is such a mystical view about Konqueror like a hidded gem. It really prepares the terrain for the upcoming Krash release. In reality basically everybody who was compiling KDE 2 was sort of aware of the possibilities and problems around the system.

What is happening is that the new KDE will have probably the most flexible and fast component model available in the Unix world and Konqueror is one of the biggest beneficiaries. I don't know if it was mentioned here that it can dynamically embed postscript, dvi or for that matter, whatever viewers. It was a real mess in the last two months while this things cleared up, but it seems that it opens new possibilities.

One of these is that it allows independent programmers to contribute parts to konqueror - while still keeping their independence, responsability etc. The same thing is true for basically all the Koffice applications.

On the other hand there are still a lot of bugs to iron out both in the html renderer - and probably more of them will surface as more people will go and use it for daily browsing.

It is true that by putting 100 people to work on a relatively small piece of code like a html renderer is not going to work. But the component model allows a nicer way of contributing meaningful parts - I really see an emerging cottage industry of kparts - from mp3 players to flash plugins etc.

Lotzi Boloni

Re:Uh, you forgot Mosaic...

[SoftwareJanitor]

There were rumors recently that someone was trying to get Mosaic development going again...

I don't remember where I saw them or who it was that was doing it, but it wasn't that long ago.

Mozilla != Netscape

[SpinyNorman]

Netscape (Navigator/Communicator) comes from Netscape/AOL, while Mozilla is the open source project from mozilla.org (originally based on Navigator, but subsequently rewritten). Netscape 5.0 is being developed by AOL, and is distinct from Mozilla.

HTML Embedding

[Indomitus]

Having a nice, embeddable browser is very exciting. Things like the GNOME help browser would be greatly helped by having a real Mozilla inside it. Help files could link to webpages and it would work inside the app, without having to call Mozilla seperately. An email program could use the embedded browser to read your HTML email (I hate html mail but I do get it), or an HTML editor could actually do WYSIWYG for once instead of having to call a seperate program or write a new browser inside the editor. Of all the things you listed, I think browser embedding is probably the most useful if you have the imagination for it.

What Linux (and all other OS'es) really need is...

[kuro5hin]

... a very lightweight (i.e., basically no-controls) standards-compliant (XML, HTML, Javascript, CSS) internet-applications renderer. I've said this before in this space, and I'll keep on saying it. "The Web" as it exists now, is a crappy place to deploy apps. But it also has incredible potential as the future of computing. Sharing resources, mobility, cooperative work environements, the list of advantages to "web-based" apps goes on and on. The main handicap right now is that it's a huge pain to support the damn things, because you might as well be writing them for an infinite number of clients. What if there was just a set of standards you needed to conform to, and everyone had, locally, their little app renderer. All they need to do is launch it, and maybe enter a URL, and then they're using your application. It doesn't need email. It doesn't need buttons of any kind. It doesn't need bookmarks or any of the content-is-king-www-dot-com GUI frosting. It doesn't need a "message center" (What IS a message center?) It's just a place where I can draw my interface, so you can use my nifty distributed application. Who wants to write this?

----
Morning gray ignites a twisted mass of foreign shapes and sounds

w3m screenshots

[KMSelf]

Nick Petreley plugged this at IWE a ways back. I posted some screenshots of a few favorite sites: IWE, Slashdot, and The Register.

As previously stated: Karsten M. Self seal of approval with five stinky herrings.

Gnus/W3

[Per Abrahamsen]

Now, I happen to use XEmacs. It's my favorite editor. I couldn't code without it, debug without it, or even read e-mail without it. But I can browse the web without it, and I think building an emacs-based browser is just way over the edge.

But W3 is what allows Gnus to read /.. Why would anyone use a web browser to read a message board, when a news reader is so much more convenient? Also, W3 allows Gnus to read mail (and news) messages in text/html format. It's so fast and tranaparent that I often forget to be angry over getting text/html. In fact, I rarely notice it.

Re:When running the nightly builds...

[Otto]

Considering I had no trace of any older version on my system, somehow I doubt this would have helped.

---

Re:Why You're Wrong Again

[Otto]

M10 is way old (pushed to ftp a month ago?). Development toward M12 has been going for nearly a week now (that's post-M11 development.) M11 is within minutes of being posted but was pretty much ready for Windows and Linux a few days ago (Mac required a few small fixes.) Look at the nightly builds, try something recent and then post more inacurate information for me to refute.

Hey, I'm talking about what's on the webpage. That's the face they're presenting to the world in general. Not the FTP, not any CVS trees, whatever is on the web is what a person will see and download.

BTW, my argument is irrefutable because I posted no information that was not my opinion. :-)

In any case, I'll give M11 a shot, the moment it's on the webpage. Until then, forget it. If you have to have more inside knowledge to attempt to use their product, then it's not worth it. You think Joe User is going to go to the webpage, see M10 under the "Latest release", and say "maybe there's a later one on the ftp".. Hell no.

Well, dang me. They put the M11 on the web page yesterday, it looks like. Never mind, then. I'll give the new one a shot. :-)

---

Re:It's Spyglass

[fishbowl]

But they COULD, for what's pocket change to them.