Firefox 2 Alpha 2 Reviewed

pcabello writes "Firefox 2 Alpha 2 was released yesterday. Check what's new in this review at mozillalinks.org with screenshots."

Memory

[siplus]

Anyone know if the memory problems that everyone was complaining about in v1.5 is fixed for 2.0?

Re:Couple of questions

[aussersterne]

Yesssss! The I-didn't-mean-to-drag thing drives me nuts. In fact, general UI slowness is the thing that keeps me from using Firefox instead of Konqueror a lot of the time.

I know that my processor is "only" 1.3 GHz, but I swear there was a time when a gigahertz-plus CPU was enough to operate a GUI smoothly. But maybe I'm remembering incorrectly...

Here's something to fix

[handelaar]

Early versions of FF allowed me to Find text anywhere on a page, including inside textareas.

That's been broken for years now. I don't care about how it renders RSS, I want basic functions to unsuck.

Re:winter release

[n0-0p]

Scheduled for sometime this winter (or summer in the North Hemisphere)

Don't forget that your summer is somebody's winter.

Good Work

[Nicolay77]

Although I vastly prefer Opera, anything that can help decrease market share of IE and its broken everything is good.

Re:Memory

[jockm]

I'm sorry the fact that Firefox will consume memory until it becomes unusably slow and/or crashes and has no way for me to free up that memory; that is a memory problem.

Download manager still broken?

[edxwelch]

Fixfox and mozilla are unable to resume downloads across sessions. In other words if you have to reboot the PC for any reason, you will have to start that 300mb download from scratch.
This bug has been outstanding for several years.
There are numerous other missing features in the download manager, just compare to the download manager in Opera.

Re:Questions . Features.

[Anonymous Coward]

2. Why does firefox need XUL gui ? Why not use gtk or something else ?
There is no other toolkit besides XUL that works well cross-platform? XUL is very easy to write for new extension developers? XUL allows for more rapid development?

Re:Memory

[Wm_K]

So install Leak Monitor. Then you can see the cause of the most severe memory leaks: poorly coded extentions.

Can anyone explain how this happens. All(?) extensions are written in XUL using javascript right? How is it possible to create a memory leak using javascript. The garbage collector of the javascript engine should get rid of unused allocated memory right? Is it the javascript engine that has leaks? Or is it just the definition of memory leak taken very broadly?

I just went trhough the changelog...

[giorgiofr]

... and there's not a single feature in FF 2 that hasn't been in Opera for ages. The FF team is slacking - they're not innovating anymore. Not that they OWE me anything, of course. Just saying.

Re:I just went trhough the changelog...

[chrisgeleven]

Question is, does Opera do these features better or will Firefox?

It isn't so much who had what feature first, it is who does it best. How hard is it to understand that?

use a permalink...

[Val314]

like this one
http://mozillalinks.blogspot.com/2006/05/bon-echo- aka-firefox-2-alpha-2-review.html [blogspot.com]
if you want to link to an article of a blog and not just point to the main page...

Search plugin order

[cpt kangarooski]

Fucking dammit.

Why the hell are there buttons ('Move Up' and 'Move Down') for reordering the search plugins. They should be able to be dragged and dropped. It's not like the developers can't do this; the bookmarks can be. Why not this?

(It would also be nice for Firefox and Mozilla to understand URL files generated by IE. Safari seems to manage.)

Re:Memory

[acm]

Sorry, but poorly designed caching is a memory leak [msdn.com]. I shouldn't have to restart my browser because it is taking 700mb of memory (no lie). Especially when I only have one window open.

Everytime a Firefox article gets posted, I see someone post a hack to fix the memory leak problem. I've tried every one of them and none of them fix it on my end. The only externsion I'm running is Google's Toolbar. Regardless though, no one except the most hardcore Firefox users would ever know to look in about:config to turn off this "feature". And they shouldn't have to either.

Re:Firefox with extensions

[Surt]

I think the problem with that philosophy is that downloading the 'right' set of plugins to get a good experience is too challenging (for most users). You really do want your users to download, once, a package of stuff that yeilds a great experience, so that your reviews will be nice an glowingly positive. Hence, we'll always want to see the best features of the most popular plugins make their way into the core browser.

Re:Memory

[SeaFox]

As reported before, Firefox does not have memory problems - it has a feature that is very memory intensive.

And as mentioned before [slashdot.org] there are bugs for memory leaks that predate the fast back-forward feature. And to say that memory probelms are all becuase of this feature is revisionist history.

Re:I just went trhough the changelog...

[Anonymous Brave Guy]

If history is anything to go by, then probably Opera will. Sometimes, you do get what you pay for, and while Firefox is a great improvement over IE in many respects, it's been trailing Opera for several years IMHO.

Re:Slashdot will never be the same

[kfg]

vim, an application commonly run from the commandline, has gotten spellchecking as well. This actually bugs me a bit. I already have shell level spellchecking. Now I've got an extra wheel to haul around.

I suppose the next version will have an embedded MTA.

Hey guys, remember that "Unix Way(tm)" thingy? There was a reason for it. How about a little cooperation between the wheel makers?

KFG

Re:Close button at same tab

[Anonymous Coward]

"I suspect they made that particular change because a large majority of users wanted it"

Ok, source for this "large" majority claim please. Considering the number of people that are already up in arms about this here, I'd say that majority isn't anywhere as large as you'd like it to seem.

", so although I was actually speaking in general terms, yes, the rest are indeed "some people"."

In other words, "I make an assertion based on nothing, and thus it is so, and everone else is clearly in a utterly small minority". Sorry I don't buy that.

"And BTW I prefer KDE as it happens, Mr Coward."

Oh, I'm sorry, it looks like you picked the wrong DE if you like arbitray annoying decisions being made about your tools with the thinnest of excuses. :-)

And btw: Ffs, if they could come up with the idea that an RSS reader should be part of the browser, rather than an extension, you'll have a hard time to sell me on the idea that where the close button go should be handled by one.

Firefox has long ago lost track of it's original goals, and it appears to have been infiltrated to boot by gnome-heads as well. Too bad, it used to be a good browser, now it's just "tolerable", apparently heading for "plain sucks".

Mod Parent Up

[Vidar Leathershod]

The parent was unfairly labelled a troll. I love the Firefox browser, and use it all the time. But it and Thunderbird have a lot of annoying, quick to fix problems that could have been fixed, and are often actively ignored. If you want a real list of stuff that is broken, we could start with the parent poster's list, which seems somewhat valid, and continue from there...

1. Under the XP home theme (reduced functionality without reason) - No "Block images from this server" in context menu - available in Mozilla forever, this prevents kids from seeing the constant AdultFriendFinder crap that comes up on some non-pornographic sites.

2. On my system, it does seem to be smaller and faster than Mozilla, though I am not sure about the new Seamonkey developments. I tried it when they first started, and their first task was apparently to introduce lots of bugs and change the icon to something they created in Microsoft Paint. Not impressed with their priorities.

3. Renamed to Firefox - Wow. This was a bad move. I get a questioning look almost every time I bring up the "better browser to use" argument to businesses. Plus, everyone ends up calling it Foxfire. There are too many "cool" names involved. Mozilla was hard enough to explain, but at least I could connect it to Netscape's mascot (since people still remember Netscape). But Firefox, Firebird, Phoenix, Thunderbird and SeaMonkey? Surely someone came up with something better, but it was turned down as too practical. Think about the words "Internet Explorer" or "Netscape". The title describes the function.

4. Memory leaks - Running latest Firefox Stable build for Win32, one window, no tabs, no extensions, haven't visited any sites with Java, one Live Bookmark (default BBC World News thing). Browsing around for a few hours, memory use creeps up by several megs. Even as I type this (watching Task manager, memory has gone from 37,??? to 39,132. Weird.

5. Incomplete, annoying interface - Well, I would call a "resume button that has not ever apparently worked an annoying interface feature. I would also say that losing favicons for no apparent reason is annoying. No built-in function for removing or re-ordering search engines (you shouldn't need an extension for this simple task.

6. Offtopic Thunderbird complaint - Signatures now have a stupid "--" in grey that cannot be turned off, and the signature is in grey too (no option to disable) which has annoyed countless customers. Some people don't feel like typing their own name 50 times a day. Email is not Newsgroups. Don't try to make it that way.

7. Memory usage is now up to 40,648. Eventually, Firefox will crash on me. Not a huge deal for me (I used Mozilla M9, M10, etc. all the time). But pretty lame for a browser that has had this much development time. No, it's not just this machine either. 40,860 now.

So stop modding people as troll, just because they didn't feel like they should have to type all this junk out, when the accusations hold water.

Vidar

Re:Close button at same tab

[Ramze]

I open about 50 windows at a time, and I like being able to close them all as i read through them with one button in a static position. I don't know where you get the assumption the developers let everyone vote and the majority of the users picked a close button on each tab. No one ever asked my opinion or gave me the option to vote, and I think having a close button for each tab is horrendous. I'm only guessing, but I bet the developers looked at other tab implimentations and went with this because it's similar to other implimentations and works for people who only use a few tabs at a time. I'll be using an extension to turn this off and maybe if enough people use the extension, they'll build the option in to change the interface. I doubt it'd be a huge amount of code to include in the release, but you go ahead and flame on w/ your opponent poster if you like.

Where are the rants?

[suv4x4]

Firefox 2's key new features:

1. the close button was moved to each tab
2. you can spell check forms .. and various other minor features

I'm still excited about it and realistic regarding the fact that getting a product of that importance out the door ain't easy.

Something is missing though. When IE7 was announced, we had hordes of Slashdotters rant how the upgrade is totally trivial just adding tabs and skinning the interface, and how Microsoft is idiotic and IE7 will be just the same piece of shit.

Where are those hordes of Slashdotters now when FF2.0 doesn't seem to live up to what was initially announced, with major features delayed or cut forever?

Or is being objective too hard for most of you, immature ranting pests :)?

Haha, hope some of you have their filter set at -1 to read this one ;)

Right Click Printing Here Yet?

[jammindice]

I would personally like a friggin r-click option to print. I mean every other thing i use has a r-click option to print. Not just web browsers but regular applications have options to print or print preview. This is a pretty damn ignorant thing to do, this is supposed to be open source so that you can configure it "your way" and they refuse to add it. Bug 204519 [mozilla.org]

What might be better though is an entire context menu options preferences page that allows you to select what options/dividers you want and where. They already have this for the bookmark toolbar folder and bookmarks in general. INASD (i'm not a Software Developer) so i wouldn't know the first thing about wrighting something like that, though i would if i could. I just think it would make a lot of people happier, hell they could even leave off the print option for a default install.

Well enough ranting, if your like me, they made an extension for printing from the r-click context menu here: Right Click Firefox Extension [mozdev.org]

Spell: Firefox 2 and Vim 7 uses OpenOffice Dicts

[herodiade42]

Well in fact that's not many "differents wheels", rather the contrary !

Vim 7.0 uses the OpenOffice.org dictionnaries (and OOo algos to take the power of them).
That's why we had 40 supported languages yet at the release.

And that's exactly what happened to FF 2: it took the well done work from OOo and based he speelcheck feature on that strong base (see http://dictionaries.mozdev.org/ [mozdev.org]), again, that why he support yet 40 languages.

So the morality ?

I would like that reviewers put this common OOoDict heritage more in perspective, so people willing to contribute on a dictionnary could know where to start. There many more languages needed to be supported (so: any contribution will benefit at least OOo, vim 7 and firefox).
Amen.

WebDav?

[CypherOz]

Any idea of a WebDav (WebFolders) capability? Then I would not need IE at all!

So why version 2.0?

[Trogre]

Where are the major changes to warrant a full major version increment?

We've just had a massive jump from 1.1 to 1.5 with little improvement. Why aren't they calling this version 1.6?

Re:Browser Speed

[Cal Paterson]

Have you read how limited account works in Vista? I leave it up to you to assess the security implications of this on IE7.

There are a number of things you haven't considered. Firstly, that when you say "limited account" what you are actually refering to is a chroot. This new feature you mention is simply that the web browser does not run as the super user; now it runs inside a chroot. This has no affect on the security of Internet Explorer whatsoever! It just means that now, malicious code can only take control of IE and IE's own files.

IE can still be broken into (probably in much the same way it can be broken into now), but with only a localised effect. Are you saying that putting a program into a chroot means that all the bugs and holes magically disappear? You are mistaken as to the importance of this change.

This has been a point of play on every other operating system that runs a webbrowser since webbrowsers were first written. Do you think Tim Berners-Lee ran WorldWideWeb as root on his NeXTSTEP box when he invented the internet? No other Operating System will run webbrowser as root for one simple reason; it's a totally stupid idea; get hacked and the whole machine will go down.

This is hardly a feature worth trumpeting; it should have been this way by design. If your "assessment" of the security of IE7 has lead you to this piteous conclusion, then you really have a problem.


Also whether it has "W3C Support" I hope you realize the phraze is non-sense, however it has plenty of CSS fixes and ehnancements which I've studied in detail and had the chance to test myself in the Ie7 beta 2.

1. You may be interested to note the correct spelling [answers.com] of the word "phrase".
2. The word nonsense [answers.com] is not hyphenated.
3. If you we to have written that sentence correctly, it would look something like this;
   "Also, (COMMA) as to whether it has "W3C Support,"(COMMA) I hope you realize,(COMMA) that phrase is nonsense.(FULL-STOP) However,(COMMA) it has plenty of CSS fixes and enhancements,(COMMA) which I've studied in detail,(COMMA) and have had the chance to test myself in the IE7 beta 2.

The sentence where you accuse me of nonsense actually is nonsense in itself. You can barely manage English, let alone HTML.

You note that it has fixes to rendering, but you fail to mention that it is not an attempt to actually bring IE in line with W3C specifications. It is still lacking W3C standardisation.


At least I don't have my head up my ass, repeating anti Microsoft cliches like a fanatic without putting any thought into it.

Instead, you prefer to sound off about new features when you have no actual understanding of their effect.

Re:So why version 2.0?

[Gogo Dodo]

Because version numbers don't mean anything nowadays. Back in the day, they used to mean something. Now the version numbers are just marketing.

Re:Mod Parent Up

[complete loony]

"... quick to fix problems ..."
Right, written any complex software lately?

Something that might look easy to fix from the user's perspective may break assumptions that have been made in the code and require a significant amount of rework / refactoring to change.

If you really think it's that easy to fix the problem, hunt down the bug in the code and fix it. While I know this sounds like the typical linux developer reaction, I can sympathise with the sentiment. Most open source devs are not paid to fix bugs. While they might fix things that directly affect them, any other issues will go onto the end of a long list of other priorities.

Thankfully firefox seems to have the support of developers who will eventually get around to fixing such issues, but it's still a question of priorities.

If the bugs are not fixed, why roll out 2.0?

[Wolfier]

Here are the important ones.

1. plugins should have their own thread priorities.  Ever wonder why a lot of Flash applets can slow down Firefox but not IE?  IE runs flash applets in a lower priority thread than the UI.

2. actions on file types should not have anything greyed out.  people should be able to choose custom actions based on MIME type, extensions, or both, and there must be a text box to type the application path, plus its parameters.

3. cancelling a save of a file over something with the same name should take you back to the dialog to rename the file, not cancelling the action altogether.

4. Find toolbar closes on its own after a *hardcoded* 5 second timeout.

If you check the conversations on bugzilla, the developers don't seem to like to listen at all.