Slashdot Mirror
Mozilla Branches For 1.0 RC1
At the end of last month, the Mozilla Project closed the tree for what will become Mozilla 1.0. Now jkeiser writes "Mozilla has branched for 1.0 RC1, which is the first last step to a final Mozilla 1.0! Mozilla has spent four long years getting the browser standards-compliant, fast and solid. Cross your fingers for a rockin' final release around the corner." Reader whovian points to the just-modified roadmap, too.
The release schedule on mozilla.org shows a release of 1.0 RC1, but no 1.0. When is 1.0 scheduled to come out?
It's a funny coincidence that this ended up as a story, as I was peering around the Mozilla website tonight trying to figure out an answer to one question:
When now can we expect an official 1.0 release?
I'm not a programmer in any way, so I don't know much of anything about development schedules or whatnot. And all the FAQs seemed to tiptoe around a definitive answer.
Awfully convenient that this became a story; I didn't want to ask in any of the other stories' comment sections, 'cause I didn't want to be offtopic.
We see that with a lot of open-source software - as the version numbers get higher, they change by less, never reaching 1.0. Go look at all the .99s on sourceforge for many laughs.
At least the last 0.9.9 RPM packages were rubbish; Mozilla threw Segmentation fault immediately at startup. Same both with my home comp (i586 with Mdk Linux) and my work laptop (i686 with RH 7.1), so this can't be just a random problem.
Amazingly, the nightly build RPMs seem to work just fine.
This is definitely not the first time this happens. I don't know who compiles Mozilla for packaging, but it's obviously not done very well. Problems this obvious (Segfault at startup) don't give a very fancy image of Mozilla.
All packages should be tested at least somehow before distribution.
Release Candidates are just for final bugfixing before the actual release, correct? Like the freezing process in Debian. So they'll just be stabilizing the code until they think it's ready. This is exciting because it means the 1.0 release can't be more than a few weeks away. It's a big deal because most people don't think of a project as actualized until it hits 1.0, which, let's face, many open source projects don't. If commercial product versions are any measure, Mozilla could probably somewhere above 4 or 5. It's nice to know that the developers have such a high standard for quality that even though it's been an excellent browser for several months now, they just now think it's 1.0 quality. Just think of how much better it can and will get from here. Props to all the Moz developers for such great work. Keep it up!
--
Promoting critical thinking since 1994.
There are so many nice things about mozilla that make it so much easier and enjoyable to use than any other browser ... probably the most significant thing for me is tabbed browsing man one window multiple web pages, where has this feature been, must admit it makes older versions of netscape and ie seem almost impossible to use.
Another really sweet features of mozilla is UI pleasing to the eye and intuitive to boot, if you don't like it download an alternative theme, don't like any of these roll your own. I know, I know not a new idea but it has been done well.
All in all a fantastic product. Much thanks and much respect to all involved in producing such a great product, and one thats free too :)
Too bad that the Linux version has much lower priority than Windows. Some bugs I reported half of year (or even more) still show off - just becouse they're Linux specific. Some others, mostly all platform bugs, have been fixed in a few weeks or months.
[plea for help]
Now is the time to increase the testing effort. Everybody out there, please download the latest nightly build. Get out there and test and submit bugs to Bugzilla.
You can poke fun as much as you want about the release timeline, but these Mozilla guys really work their asses off to get this product out to you at no charge. The least we can do as part of the open source community is help out by testing.
[/plea for help]
[/rant]
PK
P.S. Posted using April 9th Mozilla nightly build. A testament to how well it works and the stability of the nightly builds. I install a nightly build almost every morning and never had to revert back to using an older build because something major was broken. I always install the Linux tarballs, but of course YMMV for other platforms and installation methods. But I don't expect anything would be different for the Windoze and Mac builds.
Now, I'm not saying that Mozilla does not do BiDi - it's just that the bugs in the BiDi implementation are so severe as to make Mozilla completly unsuable to anyone who reads and writes a complex script language. most notable are the "text selction" bugs which makes copying and pasting from pages that contain complex scripts impossible, and worse - the BiDi text input bug which causes Mozilla to spontaneusly crash when entering text in (for example) the text area boxes of weblogs.
The most infuriating thing about this, is that the serious BiDi bugs resolution dates have been postponed to later and later milestones, and now, as those are marked nsBeta1 (meaning - fixes to be submitted before 1.0 released), the source tree still has no fixes in sight, and I'm starting to doubt if we will see Mozilla as a competing browser to IE on the 'end user's' desktop - even after 1.0.
see bugs :
95228
82352
125546
112101
75011
On linux/i386 it`s a case of copying the libflashplugin.so to mozilla/plugins, wherever you installed mozilla. On windows its a case of running the installer and it works out where to put it. Netscape plugins seem to work just fine in mozilla.. However it would be usefull to have flash plugins for some of the os/hardware combinations not supported by macromedia.
http://spamdecoy.net - free throwaway anonymous email - avoid spam!
just out of curiosity, what compielr do they use for the Win32 builds? I mean I would guess VC++ 6.0.
.NET (quite nice i might add) and the new optimizing technologies in there are amazing. My NES emulator gained an extra 100 FPS just from a simple recompile with new compiler. Could Mozilla for Win32 gain even better performance if they compiled it with VS .NET C++ compiler?
The reason i ask is that i recently upgraded to Visual Studio
One of the things that really bugs me is when people look at the bug cound and say hay there were 21 000 bugs in X version but there are now 22 000 bugs in Y version so X version must be buggier than Y version.
Generally most of the bugs in that were found in version Y were already in X but they weren't found. That is there aren't more bugs just more that are found.
Another thing is have you read some of the bugs submitted?
Check out these(5 new bugs picked at random):
*Bugzilla Bug 78633 [console] photon port should not print to console for opt builds (maybe)
*Bugzilla Bug 35419 solaris/gcc should use -shared instead of -G in configure.in DSO_LDOPTS
*Bugzilla Bug 108476 Error with XML
*Bugzilla Bug 56179 Broken mozilla.org links
*Bugzilla Bug 9185 Gtk command-line args crash viewer
It may just be me but none of these are show stopper bugs in my mind. The truth is if the bug database wasn't open then people would be talking about how much more stable the new mozilla is instead of how many more bugs it has.
It a couple of people went through the 22 000 bugs and removed the redundant bugs and fixed the trival bugs that most people don't care about chancers are that after one or two months the bug cound would be down to something more like 3 000 bugs BUT mozilla would be almost exactly the same.
Unicode is poorly designed it the spec.
.... #)
Apple and other corporations have created far more usable script languages supporting arabic and hebrew and asian sets > 64K symbols, etc.
I was horrified to read the unicode spec when it hit the store shelves years ago.... I wanted to see if they called and octothorpe and octothorpe was correctly listed as an entry for the symbol commonly called "hash sign" and "sharp" and "crosshatch" and "number sign" and "octothorpe". (all the same suymbol
But I found out unicode was impossible to implement in pascal and Modula-2 or any case sensitive language!!
Why?
Because Unicode morons added a header definition for Pound_Sign (the british unit of currency and then retyped pound_Sign with a lowercase letter in the c language header file for the octothorpe.
octothorpe is an undisputed non-confusing name and obviously british people take great offense in americans renaming their currency name for antoher symbol (the octothope with its many other names).
I vowed that day that I would NEVER EVER support unicode and in fact I vowed a lifetime goal of another promise to myself... I vowed to sabotage all unicode progress wherever I could by making it defective, slow, incoplete or impossible to retrofit into source code. I wanted unicode to be associated with feces. I hope I suceded in my own little ways.
I celebrate BiDi bugs in all their glory. Longe live "case insensitive symbol language definitions"
death to unicode hypocrisy and mediocrity!
That roadmap isn't actually new. Just look at the revision date, the last thing on the document. Mitchell Baker has indeed promised a new roadmap, but that aint it yet.
The link's still good, though - that's where it'll be when it's done, but don't get confused because that's been there about three weeks now...
I have been using Mozilla forever (well, since the alpha 9 release), and have been amazed by it's progress. I downloaded CVS snapshots regularly, and found more and more reason to love Mozilla.
...
However, ever since 0.9.7, things didn't seem so peachy. The same Mozilla snaps that were brining me so much joy were crashing on a regular basis. Even the official releases were crashing. The little things that I thought were really cool, were deemed to be not so and disabled.
My gripes with Mozilla, however, are now over, now that I've installed KDE 3.0. The new Konqueror is sheer brilliance, and Kmail is as full featured as Mozilla mail was. I am finally 100% satisfied with my desktop system.
I tried to love Mozilla, and for the longest time I did
There's a reason for this apathy. OmniWeb may be small and light, but it also has no DOM to speak of, is way behind on features compared to Mozilla and is also MacOS X only as far as I'm aware.
In fact I believe the Omni crew are switching away from their own rendering engine to using Gecko, because it'd take years for them to get to the level of rendering accuracy Mozilla has. OmniWeb is currently a little like Konqueror on Linux, real nice, but can't really compete yet in terms of rendering or features.
The other thing that's a bit annoying, though has improved greatly since I first tried the 0.9.3 release, is the feeling that Mozilla is a little sluggish. I don't know if it's actually slower rendering an average page than Opera is (perhaps a tiny bit), but it feels slower. Opera seems to get everything worked out in the background before drawing a page; Mozilla seems to draw it as it goes. I know this is a crap reason to not use a browser, but it's that F5, <pause>, white screen, page-draws-down that bugs me.
There are, of course, other minor annoyances, like the rather slow loadup time (but I have my browser open nearly 24/7 anyway), but those two things are probably what I still find the worst. Oh yeah, and I'm sure Mozilla supports them, but there doesn't seem to be a way to turn on mouse gestures through the preferences.
Please note that I'm speaking purely from the point of view of someone who is using Opera, and before that IE. I find Opera's keyboard shortcuts and the ability to turn off Javascript, images etc with a single pulldown menu (F12) to be really great; I imagine you can do similar things in Mozilla, but they're not as easy to find in my experience.
On the other hand, Mozilla has a fabulous preferences system that is much easier to use than Opera's. It has a prettier interface too, although Opera certainly isn't ugly. And while it doesn't have mouse-wheel window switching, it also doesn't keep focus on the old window tab because of it. Don't think I'm bashing Mozilla because I'm not. I imagine that if you were someone accustomed to Netscape, Mozilla would seem far better than Opera. Opera seems to try to be more like IE. If Opera wasn't around, I'd use Mozilla, and I'm pleased there's a really decent alternative to Opera--both because competition promotes innovation, and because if Opera ever goes under or their browser just goes to shit, I can switch to Mozilla. I'd like to make a completely redundant statement now, and say kudos to everyone involved with the Mozilla project. Awsome work guys; I may not use your browser, but I'm still behind you 100%.
I use 3 browsers on a daily basis: Mozilla 0.99, Netscape 4.72 and IE 6.0. When displaying very large html pages (500K) Mozilla waits until it has read all the information until it displays it, whereas Netscape and IE will render it to screen as it becomes available. The net effect is that you can starting reading the text much sooner in non-Mozilla browsers. Admittedly, Netscape is by far the worst browser when dealing with tables whose elements are of unknown size. IE is faster than the other two browsers in almost all areas.
Is the incremental rendering of uncomplex but large HTML pages with next to no formatting impossible to do in Mozilla's current framework?
AA mozilla is here...a .html
http://www.stanford.edu/~satyakid/mozill
Mozilla has been my browser of choice for a while now, but it still has some serious bugs. So consider this criticism based in love. It's also encouraging that all these bugs have a real chance of being fixed. Even I could theoretically fix them.
There is a huge bug with bookmarks:
51683: Unable to have 2 differently named bookmarks for the same url.
This is more than a bit ridiculous, since the bug was submitted September 2000.
Another, less serious bookmark bug:
85469: Bookmark select/cut/paste operation is sensitive to order of selection
This is a major meta-bug:
73812: Browser doesn't fit with Mac OS X UI Specs
Anyone who uses a Mac uses it because of the user interface--having a program that doesn't comply with the guidelines is extraordinarily frustrating. But they're definitely getting closer.
128658: Typing in textarea really slow
Large textareas overwhelm Mozilla. This makes editing in WP, for example, very frustrating. Totally unacceptable.
However, it's great watching bugs get steadily fixed. So vote for the above bugs, get them fixed, submit patches, hooray. The rendering engine really is marvelous.
--
Make mine methylphenidate.