After 19 months of development, two name changes and more than 8 million downloads of its preview release, the Firefox browser is finally turning 1.0. Firefox is scheduled to be available for free download at 1 a.m. PDT Tuesday. Our Take: Sky’s the limit. After Thunderbird (and third-party NVU & Mozilla-Calendar), I would love to see the Mozilla Project build a multi-IM application (ICQ, AIM/iChat, MSN, Y!, Jabber, etc) with the same design/usability goals as Firefox, coupled with good voice and video support (support which lacks on all third party multi-protocol IM solutions today). UPDATE: It’s out.
Mozilla readies Firefox 1.0 release
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
There *are* multiple ways to defeat IE’s incorrect rendering of the box model, but that’s not what this discussion thread is about..
GO FIREFOX! my only qualm is that I don’t think the team took the time to REALLY prepare the servers for the massive load. They were expecting millions of downloads, yet their servers obviously aren’t up to handle the load. I would have xpected better
Note that I’m not complaining (much), just wish they’d have set up mirrors to help with the /. effect.
“Is there a way to get the Location Bar and visited URLs to behave the
way it does in Mozilla? I like the feature that the most recently
visited sites are listed at the top and the fact that the order of
visited sites change with the frequency of visits.
Firefox just keeps them stagnant with the most visited listed at the bottom.
Thanks!”
> I think, perhaps, you miss the point of a reusable browser
> component…
>
> Architecturally, IE is no different to KDE/konquerer’s khtml.
There’s one major difference: on a Linux system, KDE/Konqueror is an optional component. On a Windows system, IE is not.
I’ve been using Firefox on the PC for a while now… Using the new release to write this.
I have to put the new version on the Mac, I put a .9 release on there a while back, but ended up liking Safari better, should try the new version there too…
Does anyone know what this Cross platform COM is? Is there a link to where I can have a look at this.
I’ve been developing Win apps using COM and really really hate it, so was wondering if this was a nicer alt.
Just downloaded the Mac version (using it for this post), very nice. Didn’t know Google had a Firefox specific page till I ran this one.
Didn’t seem to have a import from Safari feature, but I’m sure they’ll add that soon…
The customize toolbar works, but the drag icon is just a large rectangle, could confuse a few people maybe…
Other than that, seems really nice. Will use it for a while and see if it comes close to Safari…
Reports still say that M$IE has 93% of the market, but in fact it does not have 93% of the browser market, please follow the link below and look at November 2004!!!:
http://www.w3schools.com/browsers/browsers_stats.asp
As you can see it only has 69% of the market with Mozilla (including Firefox) at almost 18% of the market! I happen to use both M$IE and Firefox, but my preference is for Firefox!
mv * > /dev/null
How odd – my Mandrake, that poor heavily patched unstable distribution, doesn’t seem to have any problem remembering that I’ve checked the “close download window when all downloads finished” option. With 1.0 final. Sorry, couldn’t resist.
“Reports still say that M$IE has 93% of the market, but in fact it does not have 93% of the browser market, please follow the link below and look at November 2004!!!:
http://www.w3schools.com/browsers/browsers_stats.asp
As you can see it only has 69% of the market with Mozilla (including Firefox) at almost 18% of the market! I happen to use both M$IE and Firefox, but my preference is for Firefox!
mv * > /dev/null”
I just wish that were true.
I am sorry to tell you this dude, but those stats are just for that web site which is a site that is visted by web developers who are more likely to use Firefox/Mozilla than most people.
IE, like it or not, is still dominating. Take a look at Web Side Story. There stats are much more global for the net. They are ranking IE at around 93%.
Not to mention that /dev/null is a doofus who can’t read. That site quotes IE 6 and IE 5 figures separately. IE’s total figures currently are actually 75.1%. Also, as Jason says, those figures only come from that one site, which may have a higher proportion of non-IE users for whatever reason.
…they’ll stop breaking all my plugins every upgrade?
Great, but now that we’re dealing with a 1.0 release
of a ‘finished’ product can we finally have back the
support for offline installation of themes and extensions
for those of us who use the browser offline? Especially
for those of us who like to download at work or the
cafe with that fast pipe and don’t have a connection
at home? (Yes some of us /still/ don’t have even a
dial-up account at home…yet…)
Except KDE is not tied to the OS underneath […]
Neither is IE in any sense except a marketing one.
[…] and do you are not required to use Konqueror to upgrade the goddamn OS.
You don’t _have_ to use IE to upgrade Windows either – it just makes it a hell of a lot easier.
Jabberzilla: a plugin for mozilla browsers contains a limmited functional Jabber client, could expanding on this result in the desirable multi-platform IM client, by either the use of Jabber gateways/transports or the inclusion of a more powerful backend (perhaps libgaim)
> Does anyone know what this Cross platform COM is?
I don’t know how similar they are, but google for XPCOM. That is the cross-platform COM stuff the Mozilla project made.
Sorry, but mozilla/firefox etc. are crap.
There are so many old bugs:
Example: https://bugzilla.mozilla.org/show_bug.cgi?id=115174
Mozilla/firefox reloads/resubmitts the page when I want to save the page;
This bug is 3 Years old and still unfixed:
from the bug-report:
I’ve seen this behavior when making payments to Discover Card and Citicards … The Citicards web server gives me an error message saying essentially that my payment has already been submitted. This indicates to me that the form may be being sumbitted again when “Save Page As, HTML Only” is selected…. What really concerns me about this is that, on a less smart web page, a user’s payment may be submitted twice, when all the user wanted to do was save
a copy of the payment receipt. This is more than just annoyance, it could cause people’s checks to bounce unexpectedly.
There’s one major difference: on a Linux system, KDE/Konqueror is an optional component. On a Windows system, IE is not.
Where’s the option in, say, Redhat’s installer to remove khtml – and *only* khtml ?
IE’s “unremovability” is a marketing/business issue, not a technical one.
You should not confuse the lack of *alternatives* with the lack of *possibility*. IE not just a standalone web browser app like Firefox.
This is very good news, Firefox is a great and important piece of software.
However, I don’t like them promoting Firefox as some fast, lightweight browser. This is not true at all. Yes, the speed is relatively acceptable on MS Windows. But it’s a cross-platform app, and it’s terribly slow on anything else than Windows.
I’m running 1.0 on Linux right now, and I was using the older versions, too for a long time, in Linux and NetBSD, and I’d say it’s about 10-20 times slower than on Windows. The GUI feels native on MS Windows, while on *NIX with X it’s much more heavy, with extremely big overhead for most basic GUI things. Opening a dialog on MS Windows is instant, while on Linux it takes 1-3 seconds, I can even see the windows being gradually drawn to the screen. Pressing the back button brings the previous page immediately on screen in Windows, while it takes 1-2 seconds in Unix. Resizing the main window with a webpage open is perfectly smooth and fast on Windows, while there’s noticeable resize lag on X. Clicking on a tab shows the page immediately in Windows, while there’s a small delay in *NIX. When I have several tabs open with pages that contain animated GIFs in MS Windows, CPU is at 0 %. The same thing in Unix/X eats 10-20 % CPU – even if the page with the animation is hidden, it still consumes CPU. While holding the tab key (with autorepeat) for navigating in the links eats 0-10 % CPU in a fast, native browser, in Firefox on Linux it consumes 100 % CPU and can’t even keep up with the user (when I release the tab key, the movement between the links still goes on for some time). While typing something into some text box eats 0-3 % CPU in a native toolkit, it’s 10-20 % in the superslow Firefox GUI. Basic GUI things like opening menus and working with them are much more CPU-intensive in Unix.
I like Firefox and wish it the best of luck, but I’d be extremely happy if someone could look at these things and find out why it’s 20 times slower on Unix than on Windows – is it really necessary, is it really necessary for the big cross-platform apps to be always 20 times slower than on Windows? Is there something fundamental about the architecture of X, GTK+ or whatever that prevents this? Or is it just that the programmers of these big cross-platform apps (OpenOffice.org etc.) consider anything else than Windows “second-rate” and don’t optimize for other systems? And use some non-native unoptimized abstract layers-on-layers of slower-than-slow stuff like the XUL UI?
”
Where’s the option in, say, Redhat’s installer to remove khtml – and *only* khtml ?
IE’s “unremovability” is a marketing/business issue, not a technical one.
”
tell that to Microsoft. they claim that the browser is part of the system and cannot be removed technically. they told that to the court. prove them wrong and you win the case againt MS
Firefox (v1.0) and Opera(6&7) crash too much when trying to print documents.
“There’s one major difference: on a Linux system, KDE/Konqueror is an optional component. On a Windows system, IE is not.
Where’s the option in, say, Redhat’s installer to remove khtml – and *only* khtml?”
He didn’t say KHTML, he said KDE/Konqueror – I suspect he meant that you could avoid it totally if you wished.
You could try setting DO_NOT_COMPILE=”khtml” and compile KDE; in theory it might work, although you’d probably be lucky for it to compile okay, let alone run. And that’s hardly going to show up in any installer…
If you have your theme not working and don’t whant when it will be available via update.mozilla.org you can use the ‘hack’ described here: http://www.livejournal.com/community/mozilla/201452.html?thread=124…
I can’t replicate what you see here at all. On my fast HTPC box under Linux, Firefox is perfectly snappy, as it should be. On my ancient P2/400 laptop with 4200RPM hard disk and 128MB RAM, it loads in about 20 seconds, feels fine browsing and, to take one of your measurements, needs one second to open the preferences screen, which is perfectly reasonable.
Yes, I think many Linux users simply don’t notice how slow their GUIs are, because they’re used to it, and they don’t really compare it with MS Windows, at least not thoroughly.
Have you really tried various aspects of the Firefox/Mozilla UI on Linux and Windows? Did you really compare how long it takes them to do various things (go back, open an Open dialog, switch tabs etc.)? It uses native speedy Open dialog on Windows, while it’s the superslow XUL-based Open dialog on Unix, which is typical for what I’m talking about. Did you really compare how much CPU it takes to do various thing like using the menus, holding tab with autorepeat, displaying GIF animations etc., carefully observing the CPU meter? Did you really try resizing the window (which is nowhere near as smooth as it is in Windows)? And so on and so on.
I did.
Your post indicates that you didn’t, which is what I’m talking about – saying it’s “snappy” and “fine” doesn’t say much, it’s just your subjective feeling, you probably didn’t notice the heaviness or don’t interact with the UI too much. Or don’t watch your CPU meter at all.
In a native GUI, a “saving…” dialog opens instantly, in the Moxilla/Firefox abstract cross-platform UI, the superheavy “saving…” dialog pops up slowly, I can see it being slowly drawn on the screen. Etc.
Yes, just loading webpages is perfectly fine and fast, Gecko is quite OK, but anything else, especially the GUI (GUIs tend to be much slower on *NIX) is slow. If it feels fast to you, it’s OK, but unfortunately it still consumes way much more CPU cycles on *NIX than on Windows.
IE is a disaster and I avoid it like the plague, but the IE renderer is faster than Mozilla/Firefox. Firefox isn’t “slow”, but not the fastest kid on the block. Dillo is the fastest browser I’ve ever used, but can’t really be compared to IE or Firefox because of its limited scope.
Excelent in Linux
Firefox rocks!
that sounds reasonable. But really…I don’t care. Yes, I can see the window redraw as I resize it. But it still gets resized. And besides, I don’t resize my browser windows, and neither does anyone else I know – everyone I’ve ever seen browse the web, literally, uses a full-screen window. The open file window comes up immediately on this machine (the fast one); it takes (tests) 1 second on the super-slow laptop. Sure, on Windows I guess maybe it’d be 0.5 seconds. I really, really, really don’t care. Nor do I care about CPU usage; my CPU usage is never anywhere near 100% unless I’m ripping a CD or something, and I think for the vast majority of people this holds true. If Firefox needs to use more of the capacity to do its thing, eh, I don’t care. It still doesn’t get close to maxxing out either of my systems, so it’s just irrelevant to me.
Basically what I’m saying is, I can believe your points are true, but I honestly don’t see why they should ever stop anyone using Firefox. They’re practically irrelevant. ISTR you said you didn’t agree with it being marketed as a lightweight, fast browser – well, remember the marketing is mainly aimed at Windows users, and by your own admission, it IS very fast and lightweight on Windows.
I haven’t compared Firefox on Linux and Windows fairly, no. I have it on my partner’s XP machine and it’s bloody slow, then everything on that machine is. I’ve given up on it as a lost cause; he won’t let me use the only reliable measure, which would be fire. My laptop has Win98 on it but I wouldn’t ever boot into that with the network connected, for obvious reasons. And we’re not allowed to install Firefox on the Windows machines I use at work, for whatever reason. I do get to use IE on them all day long, and even if Firefox were five times slower I’d swap in a heartbeat if allowed to.
AdamW – I think we can agree on this. Yes, for many people, this is a non-issue. For me, it’s the main reason why I don’t use Firefox, but that’s just me. It’s not that important. I just don’t like when people say the Firefox GUI is fast on Linux, because it’s one of the slowest GUIs you can find on any platform.
Lumbergh – actually, Gecko is the best thing about Firefox. It is much faster in loading webpages than IE. That’s also one of Firefox’ best selling points and the main reason why many people switch from IE to Firefox. It is really noticeable. Especially pages with long complicated tables – IE has an obsolete rendering engine that must wait for the whole table to load, while Gecko shows it almost immediately, and updates the page progressively. This is something even Microsoft admitted somewhere I think – they said that for IE7 or something, they might perhaps update the rendering engine for this fast page loading, among other things (they admitted many other deficiencies in the obsolete IE rendering engine as well, like poor standards support).
It is much faster in loading webpages than IE.
That’s just not true in my case.
nice work mozilla, fixed the problems firefox had with pdfs’s it seems. I have no complaints whatsoever now.
First off, I’m not some Firefox zealot, what I say comes from experience. “Choosing” Firefox seems today as much a matter of faith as a matter of security concerns. IMO it shouldn’t be any of those. One should choose to use Firefox simply based on personal experiences on functionality, standards-compatibility, portability, ease of use, features, speed, availability of addons (extensions). Just try for a few days or even hours, and not just click and watch, but try really to explore some of the functions Firefox contains.
For me usually it’s the small things that make a point. For example open IE, load whatever page you want, then select some words from a text sample and try to drag&drop that selected text into the location input box. Won’t work. And I use this functionality in Firefox dozens of times a day: to drag a text into the Search box, or drag an unlinked url int the location bar.
I don’t even go into detail regarding window tabs, live bookmarks (bookmarks which in fact are RSS feeds, automatically updating themselves) or an extension which I highly came to admire and love, TargetAlert, which shows a little icon beside links in the text suggesting where it’s pointing at (e.g. a little e-mail icon, a pdf/doc/… icon, showing the link wishes to open a new window, etc etc).
And as a developer (no, not web designer) there is one more thing I respect: standards compatibility is the best with Firefox. Pages that don’t render well in Firefox probably won’t go through W3C’s HTML validator. Don’t believe it, try it. It’s not a fairytale, it’s a fact.
The extension set Firefox as of now has is just stunning. One just has to browse through to see how much functionality can further be added by writing some extensions, suiting even singular user’s needs.
All in all, if you are thinking about trying out Firefox, and reading people’s reviews, I suggest don’t believe any of them, even don’t believe what I just have said. Download it, it’s only about 4.5 megs, and try it. Don’t like it, don’t use it.
What have you got to loose ?
Where’s the option in, say, Redhat’s installer to remove khtml – and *only* khtml?
You can remove kdelibs, and only kdelibs. Do you claim that, say, Swing support is built directly into the OS because you can’t remove it separately from the entire java runtime?
khtml is built directly into *KDE*, and may well have hooks in a lot of places in there. That’s still a lot better than being built directly into the OS as IE’s rendering engine is, as there is no way for khtml to touch kernelspace.
Yes, I think many Linux users simply don’t notice how slow their GUIs are, because they’re used to it, and they don’t really compare it with MS Windows, at least not thoroughly.
I use both Linux and XP extensively and I can tell you that while some toolkits are slow, others are not. Redrawing on window resize/movement is not an issue for me, because I always turn on “no-fill movement/resize” in both Win and Lin.
The sluggishness of toolkits like gtk, etc. is far outweighed by the control I get over the system and improved multitasking (Linux never gets slow for me).
Both systems are great and serve different purposes…I use Windows to play my video games, Linux for everything else.
Choose your favorite and have fun.
Went to amazon.co.uk to buy something proceeded to checkout and got a ‘Firefox and amazon’ don’t share an encryptation key ‘ or something like that. Retried and got ‘This form contains no data’. Tried 3 more times and the same. Against my own will had to use IE… Darn!
that’s weird, I’ve bought several things from amazon.co.uk with firefox on Mandrake…
Surprisingly, this extension hasn’t yet been included on the new extension list pages that the current incarnation of Firefox points to. However, if you’re willing to brave an ugly page layout, you can download a current version of it that will work.
http://piro.sakura.ne.jp/xul/_tabextensions.html.en (the primary page is in Japanese; this link is the English version)
FAR FAR superior to anything else I’ve found that enhances the tabbed browsing experience. Tab Sessions, Tab Groups, remember recently closed tabs in case of accidents, FAR more granular control of when and how to open a link in a tab, a separate Tab menu, and contextual menus… the list goes on and on.
WON’T use firefox without this.
You can remove kdelibs, and only kdelibs.
1. Where’s the easy, obvious install-time tickbox ?
2. What does doing this break ?
khtml is built directly into *KDE*, and may well have hooks in a lot of places in there. That’s still a lot better than being built directly into the OS as IE’s rendering engine is, as there is no way for khtml to touch kernelspace.
IE has nothing to do with the Windows kernel. It’s a reusable component heavily used by the shell, just as khtml is.
“Windows the OS” includes the shell and, by extension, IE. Since Windows is only sold retail to the public as a complete package, the shell is, for practical (ie: market) purposes “part of the OS’. This does not mean it’s “part of the OS” in the academic “the OS is the kernel” sense.
This is why IE cannot be removed. It’s not that the basic kernel functionality won’t work (embedded NT demonstrates this), it’s that the *product* “Windows” will no longer function completely and fully. The same applies to KDE – take out khtml and KDE no longer works fully.
The situations are conceptually identical. The *difference* is on unix replacements for KDE are common and supplied by the retailer, whereas no reseller provides Windows shell replacements (and probably isn’t allowed to by licensing restrictions anyway).
I suspect you are already well aware of this and are just be facetious, but with the lack of basic business acumen and critical analysis ability so common on this forum, you really might not understand that from a market perspective, “the OS” != “the kernel”.