What is Digg?162 Comments
- MAGZine, on 05/26/2008, -6/+199"This preview release is being made available for testing purposes only." *cough*
https://www.mozilla.com/en-US/firefox/3.0/firstrun ...
Its nice that these guys found a bug, but they're making way too big of a deal out of it. Its a RELEASE CANDIDATE, not an official release. - drakenlot, on 05/26/2008, -8/+108Does no one realize the significance of RC1 being only testing purposes?
You are supposed to find bugs so they are fixed when it's finally released. - fuzzynyanko, on 05/26/2008, -5/+37Oi veh... The complications of doing multi-platform code...
- cornflakepirate, on 05/26/2008, -2/+32Yes, yes it is. Primary because it is very small and lightweight.
- sockpuppets, on 05/26/2008, -4/+31Grammatically correct cat does not live in either of your houses.
- dawpa2000, on 05/26/2008, -4/+27you're*
- arjie, on 05/26/2008, -0/+22More like ' Oh fsync!'. But yeah, whatever.
- drakenlot, on 05/26/2008, -6/+28Honestly, if I wasn't currently on Windows, I'd want to spread the word that I wasn't on Windows.
- megamod, on 05/26/2008, -6/+26Dugg for "We don’t want to lose the user’s data, because that makes users sad, and we like to make users happy." I would've appreciated it even better if they had used smiley faces instead of "sad"/"happy"
- shakin, on 05/26/2008, -3/+22"Its nice that these guys found a bug, but they're making way too big of a deal out of it. Its a RELEASE CANDIDATE, not an official release."
Mozilla uses the proper definition for release candidates. That is, they are candidates to be released and if no major bugs are found they are renamed final and released as-is. Every other Firefox release does it the same way: the last release candidate is identical to the final release. That's why they say the bug may be fixed for 3.0 if they have an RC2 release. If they don't have an RC2 then RC1 is the final release.
Personally, I've already stopped using Firefox for most things because performance is so bad now. Even when I quit it the process sticks around and if I don't kill the process it sometimes eats up huge amounts of CPU time. - SniperZero, on 05/26/2008, -2/+20Digg and Google run pretty ***** too haven't you noticed?
- ileftfark, on 05/26/2008, -12/+29Maybe this a better question for Slashdot, but was SqLite really the best choice for db? I use it (very basically) to manage my music libary, and I find it slow and glitchy. Again, I'm no SQL expert, but was this the best choice?
- MacSuxWindozSux, on 05/26/2008, -0/+16Small, Lightweight, and Easy to use.
You could code your own management of the data, but SQLite is easier to use. - sassafras1232, on 05/26/2008, -0/+16If the RC is too late to find bugs then what is the point? If they can't fix any bugs they find in it then isn't it just a release? Alpha means "there's so many bugs in it that we'll need to tell our testers what to ignore while we fix it." Beta means "There's definitely still bugs but a regular guy off the street can mostly use it successfully." RC means "we don't think there's any bugs, but we're not making any guarantees."
They found a bug in a release candidate. This is unfortunate, but should not be terribly surprising. Should they have done better beta testing? Yes, probably. But they covered their backs by calling this a "release candidate" not a "release". Now they should fix it and do another release candidate. - drakenlot, on 05/26/2008, -3/+18That was a major change for 3.0, it doesn't do that anymore.
Plus, it's leaps and bounds faster than other browsers.
oh, what browser did you switch to, if you say IE, I'll be forced to end you. - supermanred, on 05/26/2008, -2/+17The banking system seems to be working fine.
- buckrogers1965, on 05/26/2008, -7/+21Funny. As far as I can see on Ubuntu is that firefox 3 works a lot better and faster and has a lot less 30 seconds lags than firefox 2 does.
- Ksg89, on 05/26/2008, -19/+33Well ain't you special
- midtown, on 05/26/2008, -0/+14That's because Ubuntu fixed it weeks ago. If you got it from the Ubuntu repos you are fine.
- Avian00, on 05/26/2008, -2/+16Google Cache:
http://209.85.129.104/search?hs=sUq&hl=en&lr=&c2co ... - Bith8654, on 05/26/2008, -3/+15I'm pretty sure it's 3
- Timmmm, on 05/26/2008, -1/+13I just upgraded ubuntu and now have firefox beta 5. It's so much faster! Digg is now tolerable..
- PatrickBrown, on 05/26/2008, -3/+15I am more concerned about the rendering problem. Firefox is almost unusable on web pages that have large fixed backgrounds.
I can't get the following pages to scroll in any standard manner (smooth scrolling off/on):
http://www.gameswelt.de/
http://www.cssmagazine.it/
The first is somewhat slower. The second is really slow. Also, any expanding / collapsing elements on a page via javascript are really slow. For those on Linux that cannot tell the difference, try using Opera at the same time for those pages and you will see precisely what I am talking about. - macbookhair, on 05/26/2008, -6/+18no your not right.
- maninalift, on 05/26/2008, -0/+11nonsense!
- phaedrusiszen, on 05/26/2008, -1/+12Fail. Since I don't use ext3, I'm not experiencing any of these issues.
If ext3 doesn't work for you, use another file system. Try that with Windows. - inactive, on 05/26/2008, -2/+13This is a release candidate, not beta. I'd expect it to be a nearly exactly what you would see when the official release comes out. Typically, that's how it works, a release candidate is made, and if everything looks good enough they just remove the candidate name and you have a finished version.
We'll have to see if these issues are looked into before the official release. - FireSlash, on 05/26/2008, -0/+11On which version of FF? What OS?
I've got three FF3 (RC1) windows open (with a bunch of youtube videos and other junk in my tabs)
2.1% CPU usage (Out of 800% total), and 154mb memory usage.
But lets face it, open up enough windows/tabs, you can do anything. Also, a single webpage with some bad javascript could do this on any browser.
Troll less imo. - luchid, on 05/26/2008, -1/+11Not a problem. We never liked that cat anyway.
- Zippo, on 05/26/2008, -1/+11Yeah, isn't that the purpose of a release CANDIDATE? The product is almost ready... just needs more bug testing...
Besides, Firefox is open source... it's not like the bug isn't going to be fixed before 3.0 final. - scratched, on 05/26/2008, -0/+10There was actually a good slashdot article on lightweight database just a few days ago.
http://developers.slashdot.org/article.pl?sid=08/0 ...
Basically, SQL lite is the best choice for a high-speed, light-weight DB in this situation. I'm not exactly sure what kind of setup you have, but SQL lite is usually very fast in most cases. There are lighter, faster DBs, but they are not written in C++, which is what Firefox needs. - drakenlot, on 05/26/2008, -1/+10You good sir, fail.
FF2 has that problem, but FF3 does not.
To fix your comparisons,
FF3>IE7>FFx>IE6 - rexamafex, on 05/26/2008, -1/+10Now if only Microsoft can take on this same train of thought.....
- shvmoz, on 05/26/2008, -0/+8Cracking the case on FF3 by doing an RC2 adds a good chunk of time before release. Releasing RC1 as 3.0 and doing 3.0.1 after the scheduled 4-6 weeks might well get builds with the mitigation patch into the hands of users only a week or three later than restarting the RC cycle. Given that the vast majority of Linux users -- the only ones who are on filesystems with this fsync bug -- will get their builds with the mitigation patch from distributors, that seems like a reasonable path to consider.
This "major bug" only affects users on Linux with certain (albeit common) filesystem configurations who are doing a lot of I/O on the same filesystem as Firefox's profile, and it happens by default with many editors and mail clients as well. It would have been "downright bizarre" to sound the klaxons and restart the FF3 release cycle *just* for this mitigation patch, when Linux users would be getting the patch from their distributors anyway. What triggered Mr Clinton's outraged and IMO misplaced accusations of "abandoning Linux" was the decision in the bug that this ext3 issue alone would not be sufficient to respin. - DarknessGP, on 05/26/2008, -1/+9It's not just the bookmarking. It's all the stored password, users settings, history, etc. It has to ensure data integrity, because do you really want to lose all your firefox settings (and any other program that was running a SQLLite database) because your system crashed? no, you don't. They don't use a "full fledged database" because they are not as fast or lightweight, and would still suffer the need for fsyncing. The truly ridiculous thing is that the fsync call on linux is writing all buffered data to disk, instead of just what the program that called it is using. That's just bad planning on the OS writer's part.
- inactive, on 05/26/2008, -1/+9You lost the right to be a valid source of information when you picked such a ***** name.
Plus, No, apparently you can't read: It's not a release version, It can be fixed and has in Ubuntu, and it can in theory happen on other OS's that may have this problem. - felderado, on 05/26/2008, -1/+9If things get unmanageably huge then you've made the wrong choice and you might want to try PostgreSQL.
There, fixed it for you. - dood, on 05/26/2008, -0/+8Right. I think that was the big deal when this was first announced. It wasn't surprising that there was a major bug in a release candidate. ***** happens. What was surprising was the announcement that they would go ahead and release the code with a major bug in it. That's downright bizarre. They're acting very much like a closed source game vendor, having to hit certain release dates for marketing purposes.
- midtown, on 05/26/2008, -1/+9Yup, Ubuntu fixed this weeks ago in their repo: https://bugs.launchpad.net/ubuntu/+bug/215728.
- inactive, on 05/26/2008, -1/+81. Opera beats the living HELL out of any Internet Explorer.
If you are a web developer, Internet Explorer is a joke that people have to work around because ***** like you can't get over it.
2. Firefox 3 does not have as many memory leaks, better performance, better standards compliance as FF2
So stop being an *****. It's now well enough to run on embedded devices.
3. IE sucks.
Is there anymore to say? Internet Explorer 7/8 is still crashing on me! On both PCs! they're freshly installed XP's with Code::Blocks and MinGW, and Firefox 3 and IE 8 (Application dev, and web dev.) So it's obvious: Internet Explorer is still very buggy compared to other browsers.
4. Webkit is the *****.
For right now, It's pwning almost every other rendering engine out there. I'm going to try it for sure. - shakin, on 05/26/2008, -0/+7If I'm experiencing the Firefox performance bug then I can't be using Windows.
I am using Opera right now. Opera has usability problems, but it performs really well.
I did really like the Firefox 3 betas until this bug reared its ugly head. FF3 has a lot of improvements over 2 and I was really excited about it.
The problem I'm having with the process not dying is unique to RC 1 (beta 5 worked fine) on OpenSuse. The Firefox 3 beta that comes with Ubuntu 8.04 works fine, possibly because it is patched already. - fittysix, on 05/26/2008, -0/+7I think you're overestimating the severity of this bug. Not a single Linux distro ships Firefox that Mozilla built, they all make their own builds for their repositories. They can all take the patch that was made to fix this issue, and they probably all will.
The decision to make RC2 will happen tomorrow most likely, and I'm sure this one issue, despite the fact that it's a completely non-issue, will probably be weighed heavily during that meeting, all because of some stupid press seriously overstating the issue. The Mozilla guys just want to release this browser, they've made it so much better than Firefox 2, and they want to get it in to the hands of the majority of users. At this point they've seriously over-extended the development time, Firefox 2 is nearly 2 years old at this point, a little far beyond their yearly release goals. I do realise Mozilla has never hit a release time frame, they always miss it by about 2 months, and this has already happened with 3 (Beta 5 wasn't in the original plan)
They can develop and develop until they've fixed every single last bug in the bug system (there's some 100 being reported a day, and usually at least half of those are valid) but if they do that there will never be a release. At some point you just have to say "OK, there's nothing seriously wrong with it, so lets release it" because without doing that there's just no way to get large-scale user feedback, and if you've ever programmed anything, you know that what doesn't go wrong on your internal tests can and will go wrong, in large numbers, once the user has the end product. Users systems are varying things, it's almost impossible to say what's going to happen to the browser once it hits 100mil computers until it actually hits them. - funkyloki, on 05/26/2008, -6/+13FTA: "Other filesystems (like REISER4, I have heard) have similar problems, and I presume that their developers are also working on resolving them."
I don't think the guy who developed reiser4 is going to be working on any fixes to his file system for quite a while. http://en.wikipedia.org/wiki/Hans_Reiser - secrity, on 05/26/2008, -3/+9Not so much. This is a release candidate, the released version should be OK. Also, if this were IE, nobody would know what the problem really is.
- shakin, on 05/26/2008, -0/+6That's like saying Obama is a candidate to represent the Democrats, but he'll get better before he's actually allowed to do so.
The word candidate means that Mozilla thinks this build is good enough to be released and they intend to do so if a showstopper bug isn't found. Apparently this isn't a showstopper bug for them. That's why they said they aren't sure if there will be a RC2. - lord2800, on 05/26/2008, -0/+6I've been playing with SQLite for a new website I'm building, and it's far more scalable than I ever imagined. The site itself is relatively low traffic, but no queries take any exorbitantly long amount of time, even under peak usage. I'm thoroughly impressed with the speed of SQLite.
- inactive, on 05/26/2008, -1/+7am i rite = amirite. Come on, if you're going to use a phrase/meme you could try to get it right.
- inactive, on 05/26/2008, -0/+6Actually, No need to be an ass. I don't disagree that's a better idea.
- Krissam, on 05/26/2008, -0/+6it is...
http://www.sqlite.org/download.html - lord2800, on 05/26/2008, -0/+6The problem is specifically the FS drivers not implementing fsync correctly. On some FSes, it's a NOOP. On others, it's a full buffer sync. And on still yet others, it's a proper sync of _that file_ to the disk. And as pointed out above, it's not just the bookmarking subsystem, it's a good portion of all of the settings. Moving to a properly ordered model of data is always a good thing, and I welcome the move from an arbitrary file format to a real database.
-
Show 51 - 100 of 163 discussions
Browsing Digg on your phone just got easier with our enhancements to the 
© Digg Inc. 2009
— Content created and posted by Digg users is