Slashdot Mirror
Mozilla 1.8 Alpha Released
AllMightyPaul writes "Last Friday, the Mozilla Organization announced Mozilla 1.8a. You can download Mozilla 1.8 alpha (with torrents available) from the Mozilla public FTP server. Features include a basic upload FTP UI, improved junk mail filtering, and the number of cookies that Mozilla can hold has also increased 'dramatically.' What's amazing is that they haven't even released Mozilla 1.7 yet. Here I thought that Mozilla was going to standardize on 1.7."
I'm just glad that they will develop the Mozilla package next to the firefox/etc packages.
I use the Mozilla package at home and Firefox at work (since I have to use Outlook here).
They haven't let me down yet.
This is the sig that says NI (again)
I have been running the new alpha of Mozilla for a little time now and I can definitely say that this is the best browser I have ever used.
It's faster, more responsive, uses less memory and overall is just one great piece of code.
I'm looking forward to the final release, but to those who are sceptical to running an alpha release I recommend that you give it a try anyway - it's that great!
Internet Explorer will have a hard time keeping up with the great folks at Mozilla. In my book, the browser war has already been won.
I thought Firefox was scheduled to be *the* browser in the suite (with Thunderbird the equivalent in the mail space). How does that work if Firefox is on a branch and the suite ploughs ahead?
I hope bugfixes (217527 for example which affects Slashdot) are consistantly and promptly backported to 1.7 (and thus to Firefox) or the impetus could be there to reverse the flow back to the suite- up until now I have tended to think of Firefox as "the best of Mozilla"...
--Murray Barton
Because they are implementing FTP upload. Read the darn post. And it may be the new "default download format" for people trading warez, moviez, anime and possibly linux distros, but I certainly use FTP way more than I do BT, because I actually upload stuff to a site where people see it, not leech off others. :-)
There doesn't seem to be a version compiled against Xft or Gtk+2.0. Is this a regression?
You know, starting FireFox and Thunderbird takes longer in total than starting Mozilla.
And together they use more memory than Mozilla does, or at least no less memory.
As far as usage goes there's no perceptible difference in browsing speed between Mozilla and Firebird.
I think people like to say Mozilla is "bloatware" because it's the trendy thing to do, but I don't think it deserves the title.
The interface used to be fairly slow in pre 1.0 versions, particularly in the Mail/News component...but that really didn't have a hell of a lot to do with "bloat".
Now I don't notice any difference between the speed of Mozilla's interface or any other Windows Program.
Advanced users are users too!
Wouldn't a better architecture have been to go for a plug-in model? That way I could choose what extras I wanted, and either uninstall, or never install, the crud I don't care about.
The problems with the Mozilla monolith are:
And separating the dev' environment (browser) from the users environment as you suggest only makes life harder.
Bad analogies are like waxing a monkey with a rainbow.
That and much of the spam I've been getting recently is very simple. Say, two links in an e-mail and that's all (plain text). They're not being picked up as spam by spamassassin any more, it has about 50% accuracy at the moment.
I have a trouble, I can't make them co-exist!
Firefox is using the pref files that Mozilla uses BUT the new Mozilla hangs at the older version's pref.
Can someone tell me how to move Firefox preferences so I can make them both work.
If SpamBayes filtered a legit message in with the spam, how would you know about it?
:)
Short answer: because I am very distrustful of technology, and I do actually skim through my spam-box every so often just in case.
However, one reason it never came up is that SpamBayes has a very graceful response to email it's not quite sure of. It puts emails below a certain (user-adjustable) confidence threshold in a "suspected spam" bin, for you to review at your convenience. After the first few days actual spam almost never ended up in here, but it was very good for catching email from, e.g. new mailing lists that I'd just signed up for (which Thunderbird 0.6 simply dismisses as junk).
To be fair, yes, I should have taken this into account when making the claim you quote. But y'know.
Rarely have I seen developers so resitant to change as on the Mozilla bugzilla forums. It seems the core developers fight every little attempt to improve the interface, fought the new website (and thankfully lost), fought adding a new splash screen (and apparently threw in that nice new orange "thing" as a big "fuck you" to everyone who posted on that thread). Hell, If I was running the show every new release would have a new splash screen ala the GIMP. Because, really, who gives a shit about some minor bugfixes, but the GIMP splashscreens rock and are genuinely funny in the beta builds, so people upgrade anyway, the builds get more testing and everyone is happy.
Basically, everything should be open for change. Every UI pixel spacing issue should be open for improvement, every 1px border in the interface needs to be justified. All text that is presented to the user needs to be constantly reviewed for easy of use, and so on... Of course, these things are only essential if you care at all about people actually using your software... The Thunderbird logo will convert more users than any single feature X you can name. If you can't see that you really don't understand the end user market and their need to download spyware infested wallpaper changers.
Only slightly related is this:
What I never understood, though, is why with the X version of Mozilla (Linux in my case) clicking the middle mouse button on a tab by default tries to load the current selection as an URL.
Why? First thing with all Mozilla installs on Linux I do is to disable middlemouse.ContentLoadURL. Why on earth do they set it to true on Linux? Just to make life harder for people whop use both Win32 and Linux? Or do they track this silently somehow, trying to figure out how many people know how to change settings "back to normal" via about:config?
It seems the core developers fight every little attempt to improve the interface
- Not everyone has the same definition of "improve"
- The suite is now in maintenance mode, and so there will be no big UI changes
If I was running the show every new release would have a new splash screen ala the GIMP
This sounds like a great way to get people to spend their time arguing instead of hacking.
Gerv
- You get everything or nothing: I can't decide just to have the web browser and html editor: but I'd rather use my existing email app so I don't want that taking up resources on my machine.
The good thing about the mozilla browser is you -can- do this, all you need to do is pull down the source and compile your own version of mozilla with the various features that you do/dont want. Pretty much everything can be disabled/turned on and off in the main build by simply editing the- Regression testing. This is more an issue for the Mozilla developers, but a change in one component (email say) could break another part (html editing say).
The HTML editing component of Mozilla is in no way dependant on anything at all in the mail component. (although mail is dependant on the editor to create HTML emails etc). Thunderbird still pulls in the same dependant compoents just like standard Mozilla does. It's similar to using libGtk+ libraries in an app, it has the possibility to break things further down the chain (usually it put thru more testing anyway).Isn't this all standard software engineering practise which makes development alot simpler rather than having to have a similar thing coded in 10 different places.
Mozilla isn't bloaty though, I've been using it since 'milestone 18' back in the mid-nineties when it was a bit pokey and broken.
Have you done a quantitative ascessment of this feeling that Moz is big or slow? I think Mozilla is quite fast, certainly faster than IE. Also, I think that if you could un-marry windows and IE and get a full grasp of how much RAM IE was using (even when it's not loaded, mshtml.dll and friends are in RAM) you'd change your story.
Every web browser is going to use a fair amount of RAM because it needs at least a window-sized buffer to composite on. Safari and IE are tricky because they use the OS libraries for that, so it's not as easy to see the footprint, but Moz does it inside itself, so the footprint looks somewhat massive.
"Sometimes, I think Trent just needs a cup of hot chocolate and a blankie." -Tori Amos on Nine Inch Nails
Does this new release restore any support for the MNG/JNG graphics formats, or is GIF still the only animated format supported?
I was recently "forced" to get IE running on the kids' dual boot Linux/Win98SE machine. My son needed to use a certain college web site, and rejected Mozilla, being "IE-only". So I visited Windows Update, since IE on the box had never been used for web browsing. Many, many updates and reboots later, he was able to do what he needed to do. After I have paid the first tuition bill and become a member of the in-crowd, I'm going to write to the college about their IE-only site, about how they're aiding and abetting a convicted monopolist with a site like that, and how they should be using w3c, webwasher, and the like to generate portable content.
My kids tend to keep the machine on Windows, largely because they can do what they need there, plus play games. After this experience, I cautioned my son to avoid IE because of future security problems, even if it is currently fully patched. His response... IE is a *pig* compared to Mozilla.
The living have better things to do than to continue hating the dead.
Hi, I install cable modems for a local cable internet service provider. Before I go any futher, let me just say this:
Geeks don't get to see how the other half lives (fixing mom's computer doens't count).
I am required to configured the customer's computer and setup their e-mail. Part of the install process requires me to hit the cable company's web page to allow the customer to chose their e-mail. Every day I get to see 20 or more fscked up customer computers that have so many spyware programs, viruses, trojans and other assorted crap gumming up their desktops. It's not uncommon to see 15+ instances of IE load up with ads before I can get a usable browser. More often than not, the browser's spyware add-ins have the customer's computer so fscked, that I have to ftp to mozilla and pull down an clean, standards compliant browser that blocks pop-ups. Only when I load the same web page back to back between IE and mozilla does the customer begin to understand just how fscked microsoft software is.
So, even though I don't have the money to contribute to the Mozilla project, I would just like to thank the hard working folks who put that fine browser alternative togeter.
Thank you so very much. Without Mozilla, my install time would increase from an average of 20-25 minutes to well over an hour.
And to Microsoft: Shame on you, your shoddy code and your market share. If there's anybody headed for a fall, it's you.
I started using Thunderbird a long time ago, much like you did, and found myself in a similar situation to yours.
In the beginning, the included spam filtering worked wonders, but after time more and more spam began to leak through no matter how much "training" I did.
Instead of moving to a different email program as you did, however, I simply kept Thunderbird and used POPFile as a spam-filtering proxy. Because of this, I can actually directly compare the in-program filtering of Thunderbird to an outside bayesian client. Right now, according to the built-in statistics of POPFile, it's at a 99.36% accuracy rate, even with the large number of random-word spam attacks I get daily, yet Thunderbird only catches about half of them.
So I have no doubt that you are correct in your argument that SpamBayes isn't being caught by the same random-word techniques that are currently ruining the effectiveness of Spamassassin or Thunderbird's built in filtering.
To me, it seems that it's just tailing on the 'zilla name, and no real relation.
Bang on.
Gerv
And no, "Sunbird" isn't even close to a suitable answer. Neither is "Thunderbird" or "Firefox."
Corporate users can barely grok "Mozilla" but they certainly understand "Oh, no functional calendar? I'll just stay on Outlook..."
Agreed. I like the "UNIX way" of doing things, and it can work very easily for this sort of thing.
The desktops should provide a standard way of providing inter-app communication (is that what message bus is attempting?), so that clicking on a link in my e-mail client of choice it sends it to my browser of choice...
Although it might be nice, I don't think we really need a desktop standard for this right now. My email client already does what you describe. In a similar fashion, the browser should run the email client when you click a mailto: it's simply "sylpheed --compose [address]". A web browser can allow you to enter the appropriate command to run, or it can have a list of popular email clients to select from (including any that are built-in).
That kind of simple process invocation is the Right Way to do this. It's trivially simple, and it can do 95% of the things you'd ordinarily want inter-app message processing to do. Or 100% if you extend it a little where appropriate, using pipes and so on. The more complex it gets, the more it needs to be standardized; but the dead simple stuff that everybody uses all the time doesn't require much thought at all.
That there is no obvious way to configure Firefox to run that command when I click an email link is an inexcusable omission. Even if they are trying to build a all-powerful development platform which will eventually replace every other app, there's no reason they shouldn't play nicely with the rest of the system in the meantime.
That's exactly what I'm finding too. There are simply so many ways to misspell words that these filters are ineffective.
:(
This may simply create another round of technological escalation, but... if you search Google for "v!agra", it will ask you if you meant "viagra". That's not really difficult to implement.
SO- If instead of storing "v!agra" in the filter we stored say, $typo$__viagra or some such thing, we might get filters working well like before.
Likely someone has already implemented this somewhere, but if so I haven't heard of it
Information: "I want to be anthropomorphized"
This is not the unix way of doing things.
Agreed. But it's time to start working towards some unification and integration on desktop apps because the 'UNIX Way' has failed to capture the desktop market.
Mozilla is OSS, so improvements to any part of it wil ripple through the different products automatically. FireFox, ThunderBird, Mozilla and Camino are all coming from the same base code, and improvement to that code improves all the products. Continuing to develop the 'monolithic' mozilla is vital to the rest of the projects, because the monolithic app showcases and tests the ground for features that may or may not dribble down to the 'birds.
Thinking about it like 'if you write code for Mozilla, you DIDN'T write for FireFox" is backwards, if you improved Mozilla you improved ALL of the mozilla.org offerings.
If you add code to Mozilla that does AOL mail or AIM protocol, that would be fscking AWESOME! Someone else will modularize it and make it a plugin for FireFox later, and we'll have a better offering, and it won't be shoved down anyone's throat.
Personally, I just moved from Mozilla (for mail and web) to FireFox and ThunderBird, I'm not at all impressed. I saved a few MB of RAM, but overall I was happier with the monolithic app. I switched so that I could file bugs and make the new apps better.
"Sometimes, I think Trent just needs a cup of hot chocolate and a blankie." -Tori Amos on Nine Inch Nails
Mozilla 1.7 has been pending for a couple of months now, and I reckon something specific must be holding everything up for 1.8 to be in alpha before 1.7 is finalised. Does anyone know what these things are?
I've been following Bug 18574 (no links allowed from Slashdot), which concerns restoring support for MNG and JNG image formats. The debate about this bug has been long and arduous, and heated at times: essentially, it's the same old nobody-supports-it and it's-just-code-bloat arguments versus the same old it's-a-good-format and Web-pages-will-use-them-if-browsers-do arguments. However, people are starting to get it to work on their own builds, with some crashing still on some systems (eg. OS/2) with some image files.
If there are specific issues holding up 1.7, I'm starting to suspect this is one of them. Officially, there is no target milestone for Bug 18574. Of course, if it doesn't make it into 1.7, it may end up as an XPI (eg. Mngzilla) and all will be well. Does any one know for sure?
Attack its weak point for massive damage!