b********g
发帖数: 404 | 1 不能说amazon netflix曾经down 过机，你MS就比别人强，对吧
这种比较有什么意义啊，跟牛屎版上一说国内腐败失职什么的马上就有人说美国也有，
单纯比较各种性能和可靠性的文章不少啊，
apache vs IIS，mysql vs sqlserver，
谈到desktop上的IE vs FF的时候很多MS fan就用市场占有率说事，
到了server上的时候怎么不说啦，又改说稳定性啦，
然后就说你gmail也down了，为啥我就不能down LSE，
要知道人家gmail现在还是beta，人家哪天关了你也没办法，
真正拥护open source的人是因为理念，
如果你愿意可以自己修正任何bug，
并且别人修复了bug你也可以享用，
拥护MS的人认为有了bug能知道找谁，
但是MS修复bug的速度和水平，
你要不是boeing，等上一两年简直是再平常不过的事了，
这就是区别 |
N********n
发帖数: 8363 | 2
Is that so? Firefox had that notorious memory leak problem
since early on. How long did it take for them to fix it? 1
year, 2, 3 or 4? Does it get fixed at all? What happened
to that OSS fixes bugs fast hype?
【在 b********g 的大作中提到】
: 不能说amazon netflix曾经down 过机，你MS就比别人强，对吧
: 这种比较有什么意义啊，跟牛屎版上一说国内腐败失职什么的马上就有人说美国也有，
: 单纯比较各种性能和可靠性的文章不少啊，
: apache vs IIS，mysql vs sqlserver，
: 谈到desktop上的IE vs FF的时候很多MS fan就用市场占有率说事，
: 到了server上的时候怎么不说啦，又改说稳定性啦，
: 然后就说你gmail也down了，为啥我就不能down LSE，
: 要知道人家gmail现在还是beta，人家哪天关了你也没办法，
: 真正拥护open source的人是因为理念，
: 如果你愿意可以自己修正任何bug，
|
g*****g
发帖数: 34805 | 3 Which bug are you talking about? All bugs are here, why not show me
which one.
https://bugzilla.mozilla.org/
【在 N********n 的大作中提到】
:
: Is that so? Firefox had that notorious memory leak problem
: since early on. How long did it take for them to fix it? 1
: year, 2, 3 or 4? Does it get fixed at all? What happened
: to that OSS fixes bugs fast hype?
|
r****y
发帖数: 26819 | 4 乍看标题，我还以为是说 MS' failure is not *NIX's success，觉得还挺有道理的。
【在 b********g 的大作中提到】
: 不能说amazon netflix曾经down 过机，你MS就比别人强，对吧
: 这种比较有什么意义啊，跟牛屎版上一说国内腐败失职什么的马上就有人说美国也有，
: 单纯比较各种性能和可靠性的文章不少啊，
: apache vs IIS，mysql vs sqlserver，
: 谈到desktop上的IE vs FF的时候很多MS fan就用市场占有率说事，
: 到了server上的时候怎么不说啦，又改说稳定性啦，
: 然后就说你gmail也down了，为啥我就不能down LSE，
: 要知道人家gmail现在还是beta，人家哪天关了你也没办法，
: 真正拥护open source的人是因为理念，
: 如果你愿意可以自己修正任何bug，
|
t********e
发帖数: 880 | 5 一个捕风捉影、毫无根据的谣言就能让M$黑兴奋成这个样子 |
x******g
发帖数: 3952 | 6 All bugs are here.
笑死人了。要是所有的bug都在这儿，mozilla就是天下最完美的software了。
【在 g*****g 的大作中提到】
: Which bug are you talking about? All bugs are here, why not show me
: which one.
: https://bugzilla.mozilla.org/
|
g*****g
发帖数: 34805 | 7 It's not a bug if nobody finds it. It's open source and everyone
can log bugs there, that's as complete as it can get.
M$ likes to compare the number of bugs found over a year, and by
their standard, if it's not a bug found by outsiders, it's not
a bug, and they can sneak in and fix it in windows update without
reporting to you. I am not surprised M$'s products always have
fewer bugs.
【在 x******g 的大作中提到】
: All bugs are here.
: 笑死人了。要是所有的bug都在这儿，mozilla就是天下最完美的software了。
|
x******g
发帖数: 3952 | 8 everyone can doesn't mean everyone will...
as complete as it can get doesn't mean "all bugs are here"...
【在 g*****g 的大作中提到】
: It's not a bug if nobody finds it. It's open source and everyone
: can log bugs there, that's as complete as it can get.
: M$ likes to compare the number of bugs found over a year, and by
: their standard, if it's not a bug found by outsiders, it's not
: a bug, and they can sneak in and fix it in windows update without
: reporting to you. I am not surprised M$'s products always have
: fewer bugs.
|
g*****g
发帖数: 34805 | 9 I guess if the bug's that notarious, it should have made the list,
don't you think?
【在 x******g 的大作中提到】
: everyone can doesn't mean everyone will...
: as complete as it can get doesn't mean "all bugs are here"...
|
N********n
发帖数: 8363 | 10
I still remember how that leading QA from Mozilla insisted
the memory leak was not a bug but a feature. HOHO. Forget
about his name tho.
【在 x******g 的大作中提到】
: All bugs are here.
: 笑死人了。要是所有的bug都在这儿，mozilla就是天下最完美的software了。
|
|
|
g*****g
发帖数: 34805 | 11 You know what, IE hasn't been able to be compatible with W3C
and they call it a feature too. Now why not show me the link,
or it's just a hearsay?
【在 N********n 的大作中提到】
:
: I still remember how that leading QA from Mozilla insisted
: the memory leak was not a bug but a feature. HOHO. Forget
: about his name tho.
|
x******g
发帖数: 3952 | 12 No need for a link. Just open a few windows and watch
firefox's memory footprint grows.
That's why I restart firefox at least once a day.
【在 g*****g 的大作中提到】
: You know what, IE hasn't been able to be compatible with W3C
: and they call it a feature too. Now why not show me the link,
: or it's just a hearsay?
|
z***e
发帖数: 5393 | 13 抛开MS不谈，我认为Open source的理念压根就是错误的。
什么叫有了bug你自己可以修复？人人都来修复，那真正的商家怎么去保证产品的稳定
性？我又怎么相信你的修复就是可靠的了？
windows vista大部分的不兼容是因为以前的第三方软件多少利用了一些比较tricky的
技巧，例如说啊我知道这个指针肯定是这个位置，不必非要去用MS的API，我直接call
这个地址就好了。
同样的行为在linux也存在，而Open Source缺乏这种强制性的统一标准，说白了，开发
者要太监你也没办法。
引用一段slashdot上面的一个评论：
Nah, like me, we just got older. When were were young Microsoft was pure
evil, Linux was the savior. All the MS did was wrong and all that Linux did
was good. Then we entered the real world. Finding the Microsoft actually
makes
【在 b********g 的大作中提到】
: 不能说amazon netflix曾经down 过机，你MS就比别人强，对吧
: 这种比较有什么意义啊，跟牛屎版上一说国内腐败失职什么的马上就有人说美国也有，
: 单纯比较各种性能和可靠性的文章不少啊，
: apache vs IIS，mysql vs sqlserver，
: 谈到desktop上的IE vs FF的时候很多MS fan就用市场占有率说事，
: 到了server上的时候怎么不说啦，又改说稳定性啦，
: 然后就说你gmail也down了，为啥我就不能down LSE，
: 要知道人家gmail现在还是beta，人家哪天关了你也没办法，
: 真正拥护open source的人是因为理念，
: 如果你愿意可以自己修正任何bug，
|
N********n
发帖数: 8363 | 14 How about searching Firefox memory leak online instead of acting
if there's no such problem here.
【在 g*****g 的大作中提到】
: You know what, IE hasn't been able to be compatible with W3C
: and they call it a feature too. Now why not show me the link,
: or it's just a hearsay?
|
C********n
发帖数: 6682 | 15 guess u r living in 2004
memory leak problem is already fixed in ff 2.x, if u still got that problem,
either ur addons have problems or u do
btw , not all memory increasements were memory leaking
【在 x******g 的大作中提到】
: No need for a link. Just open a few windows and watch
: firefox's memory footprint grows.
: That's why I restart firefox at least once a day.
|
C********n
发帖数: 6682 | 16
this means u really know nothing about open source software patching , no ,
u
don't really apply ur patch right to the source code and release it to the
public
call
this is kind funny, vista is good, just others dunno how to use it , ur
smart
ass
well, even idiots can comment on slashdot, but only smart ppl can get things
done
【在 z***e 的大作中提到】
: 抛开MS不谈，我认为Open source的理念压根就是错误的。
: 什么叫有了bug你自己可以修复？人人都来修复，那真正的商家怎么去保证产品的稳定
: 性？我又怎么相信你的修复就是可靠的了？
: windows vista大部分的不兼容是因为以前的第三方软件多少利用了一些比较tricky的
: 技巧，例如说啊我知道这个指针肯定是这个位置，不必非要去用MS的API，我直接call
: 这个地址就好了。
: 同样的行为在linux也存在，而Open Source缺乏这种强制性的统一标准，说白了，开发
: 者要太监你也没办法。
: 引用一段slashdot上面的一个评论：
: Nah, like me, we just got older. When were were young Microsoft was pure
|
x******g
发帖数: 3952 | 17 firefox memory problem never goes away completely.
it is better nowadays than 2004.
but you'll still notice it speedup a lot once you restart firefox.
all chornical memory footprint growth might not be called memory leak.
but they are equally bad whatever you call it.
problem,
【在 C********n 的大作中提到】
: guess u r living in 2004
: memory leak problem is already fixed in ff 2.x, if u still got that problem,
: either ur addons have problems or u do
: btw , not all memory increasements were memory leaking
|
C********n
发帖数: 6682 | 18
hmm, u better learn more, not all growth should be called memory leak, and
not
all memory leak was caused by ff, and I would say that most website ur
ff had problem with were those sites followed IE's stupid js standard (in
which it allows tons of mistakes while w3c standard does not).
guess u cant learn that
【在 x******g 的大作中提到】
: firefox memory problem never goes away completely.
: it is better nowadays than 2004.
: but you'll still notice it speedup a lot once you restart firefox.
: all chornical memory footprint growth might not be called memory leak.
: but they are equally bad whatever you call it.
:
: problem,
|
x******g
发帖数: 3952 | 19 There are a lot of smart ppl working at MSFT.
,
things
【在 C********n 的大作中提到】
:
: hmm, u better learn more, not all growth should be called memory leak, and
: not
: all memory leak was caused by ff, and I would say that most website ur
: ff had problem with were those sites followed IE's stupid js standard (in
: which it allows tons of mistakes while w3c standard does not).
: guess u cant learn that
|
N********n
发帖数: 8363 | 20
The only add-on I had is a Flash Blocker b4 FF3. Every time I
opened up to 50 tabs, Firefox began to suck. Sometimes hang for
40 seconds or more frequently. The memory usage was close to
500M. Closing tabs didn't help. The memory usage only grew. If
one tab crashed, the other tabs were all gone.
The problem was there throughout 1.x and 2.x. Only in 3 does
it get real improvement.
【在 C********n 的大作中提到】
: guess u r living in 2004
: memory leak problem is already fixed in ff 2.x, if u still got that problem,
: either ur addons have problems or u do
: btw , not all memory increasements were memory leaking
|
|
|
C********n
发帖数: 6682 | 21 Yeah
and those smart ppl, when they got into M$, they can only produce
shit like windows ME, windows vista
guess M$ is worse than u thought , hmm ?
【在 x******g 的大作中提到】
: There are a lot of smart ppl working at MSFT.
:
: ,
: things
|
N********n
发帖数: 8363 | 22
说人人都可以改是蒙外行的。如果整个体系设计有误，咋改啊。改了一
处影响其他模块咋办？如果甲改一块，乙改一块，怎么保证甲乙的改动
互相兼容，谁来协调? 说得轻巧，好像改几十万行代码多简单似的。
【在 z***e 的大作中提到】
: 抛开MS不谈，我认为Open source的理念压根就是错误的。
: 什么叫有了bug你自己可以修复？人人都来修复，那真正的商家怎么去保证产品的稳定
: 性？我又怎么相信你的修复就是可靠的了？
: windows vista大部分的不兼容是因为以前的第三方软件多少利用了一些比较tricky的
: 技巧，例如说啊我知道这个指针肯定是这个位置，不必非要去用MS的API，我直接call
: 这个地址就好了。
: 同样的行为在linux也存在，而Open Source缺乏这种强制性的统一标准，说白了，开发
: 者要太监你也没办法。
: 引用一段slashdot上面的一个评论：
: Nah, like me, we just got older. When were were young Microsoft was pure
|
C********n
发帖数: 6682 | 23 你自己不会去看一下linux内核的维护原则？
你当code是你想改久改的？ 你改了人家还不一定接受呢
【在 N********n 的大作中提到】
:
: 说人人都可以改是蒙外行的。如果整个体系设计有误，咋改啊。改了一
: 处影响其他模块咋办？如果甲改一块，乙改一块，怎么保证甲乙的改动
: 互相兼容，谁来协调? 说得轻巧，好像改几十万行代码多简单似的。
|
N********n
发帖数: 8363 | 24
Then just cut the crap that everyone can fix, and it's fast.
Bug fixes go through strict scrutiny which takes time and
it's not like everyone can do it.
【在 C********n 的大作中提到】
: 你自己不会去看一下linux内核的维护原则？
: 你当code是你想改久改的？ 你改了人家还不一定接受呢
|
C********n
发帖数: 6682 | 25 所有人都可以修改代码给自己使用
并不是所有人都有权利让自己的代码进入发行版
话说这里面的逻辑关系有那么复杂吗
你只要不公开发行，你随便patch,你要是想公开发行，就需要发送patch,请求审查
或者你干脆自己开个分支
没有限制你fix的自由吧
【在 N********n 的大作中提到】
:
: Then just cut the crap that everyone can fix, and it's fast.
: Bug fixes go through strict scrutiny which takes time and
: it's not like everyone can do it.
|
N********n
发帖数: 8363 | 26
If I have ver 1.0, and you go fork off one and add changes
then it becomes your own burden to be compatible with future
versions I produce. My 1.1 or whatever could bring in fixes
not working with your change. It becomes messier if you have
other code based on your own change. Fork itself could create
versioning problems, that's why it's not all that good.
【在 C********n 的大作中提到】
: 所有人都可以修改代码给自己使用
: 并不是所有人都有权利让自己的代码进入发行版
: 话说这里面的逻辑关系有那么复杂吗
: 你只要不公开发行，你随便patch,你要是想公开发行，就需要发送patch,请求审查
: 或者你干脆自己开个分支
: 没有限制你fix的自由吧
|
C********n
发帖数: 6682 | 27 if u want to fork it , u should expect the problems
and u can choose not to fork it , its ur choice
while with m$, u can hope in 3 years u can get a fix or a new function
so I guess u mean that more options is worse than no options ?
【在 N********n 的大作中提到】
:
: If I have ver 1.0, and you go fork off one and add changes
: then it becomes your own burden to be compatible with future
: versions I produce. My 1.1 or whatever could bring in fixes
: not working with your change. It becomes messier if you have
: other code based on your own change. Fork itself could create
: versioning problems, that's why it's not all that good.
|
z***e
发帖数: 5393 | 28 举个例而已，当然不可能说我今天修改了apache一点东西，然后明天就成了新patch给
你下载。
但是我的重点是：Open source这种模式是走不通的。它无非强调两点：第一，全世界
都能修改错误，第二，你可以根据自己需要进行定制。问题是这两点都带出了一个问题
：第三方软件商如何适应？
作为一个OS，Linux为系统开发提供了便利；但作为一个开发平台，Linux为第三方开发
提供的可以说是零。
【在 N********n 的大作中提到】
:
: If I have ver 1.0, and you go fork off one and add changes
: then it becomes your own burden to be compatible with future
: versions I produce. My 1.1 or whatever could bring in fixes
: not working with your change. It becomes messier if you have
: other code based on your own change. Fork itself could create
: versioning problems, that's why it's not all that good.
|
z***e
发帖数: 5393 | 29 There are WinMe/vista, but there are also Windows NT4/2000/XP.
即便是vista，作为client OS是比较失败，但基于vista的windows 2008 server是非常
出色。
【在 C********n 的大作中提到】
: Yeah
: and those smart ppl, when they got into M$, they can only produce
: shit like windows ME, windows vista
: guess M$ is worse than u thought , hmm ?
|
C********n
发帖数: 6682 | 30
第三方显然只需要跟kernel同步，当然你可以讨论桌面环境如kde gnome,但那是另一个
问题，而且说实话，我不觉得桌面将来会是个大问题，java swing 就是一条路，现在
matlab 和maple的界面完全不是问题
比起来 msdn,linux的确差点
【在 z***e 的大作中提到】
: 举个例而已，当然不可能说我今天修改了apache一点东西，然后明天就成了新patch给
: 你下载。
: 但是我的重点是：Open source这种模式是走不通的。它无非强调两点：第一，全世界
: 都能修改错误，第二，你可以根据自己需要进行定制。问题是这两点都带出了一个问题
: ：第三方软件商如何适应？
: 作为一个OS，Linux为系统开发提供了便利；但作为一个开发平台，Linux为第三方开发
: 提供的可以说是零。
|
|
|
C********n
发帖数: 6682 | 31 2008 server 不是基于vista的
而且说实话一个 server os 弄那么花的界面根本没用，不如老老实实的吧
该做的做好，我想这也是为什么 svr 2008会出现无GUI的版本的原因吧
【在 z***e 的大作中提到】
: There are WinMe/vista, but there are also Windows NT4/2000/XP.
: 即便是vista，作为client OS是比较失败，但基于vista的windows 2008 server是非常
: 出色。
|
z***e
发帖数: 5393 | 32 这不是问题所在，假设我是某个软件商，我知道vista有某某问题，我可以问微软了解
这个问题什么时候会解决或者根本不会解决，然后为新版本准备或者想办法work
around，这不会造成我project的delay，也不会推出后要重来。
但是对于linux，我知道有这个bug，这个bug可能明天就被修复，也可能永远不会，我
的project今天可能不得不设计复杂的work around，但一旦被修复我的work around完
全变成白痴甚至形成新的bug.
我认为软件开发平台就是需要统一的标准，即便Windows垮台，Linux也自然会有各大公
司联合起来制定新的标准，Open source永远就是一个拿来利用的工具罢了。说穿了，
你们工作的code，会全部拿出来open source掉吗？
【在 C********n 的大作中提到】
: if u want to fork it , u should expect the problems
: and u can choose not to fork it , its ur choice
: while with m$, u can hope in 3 years u can get a fix or a new function
: so I guess u mean that more options is worse than no options ?
|
N********n
发帖数: 8363 | 33
Having known problems/limits and then finding a recommended
work around is much better than creating your own and then
have to deal with the burden. And other than this "3-year"
fud, where's your proof OSS fixes things faster?
【在 C********n 的大作中提到】
: if u want to fork it , u should expect the problems
: and u can choose not to fork it , its ur choice
: while with m$, u can hope in 3 years u can get a fix or a new function
: so I guess u mean that more options is worse than no options ?
|
z***e
发帖数: 5393 | 34 2008 server怎么不是基于vista内核的？当然有些设定不一样。
【在 C********n 的大作中提到】
: 2008 server 不是基于vista的
: 而且说实话一个 server os 弄那么花的界面根本没用，不如老老实实的吧
: 该做的做好，我想这也是为什么 svr 2008会出现无GUI的版本的原因吧
|
C********n
发帖数: 6682 | 35
based on the bugfix record of IE and FF, I would say OSS fixes things
much faster
【在 N********n 的大作中提到】
:
: Having known problems/limits and then finding a recommended
: work around is much better than creating your own and then
: have to deal with the burden. And other than this "3-year"
: fud, where's your proof OSS fixes things faster?
|
N********n
发帖数: 8363 | 36
Exactly. Those big corps sing praise for FOSS only because they
like the cheap labor. They have no problem taking away your hard
work but you can forget about them opening up source code of
their money makers to you.
【在 z***e 的大作中提到】
: 2008 server怎么不是基于vista内核的？当然有些设定不一样。
|
C********n
发帖数: 6682 | 37 IBM 倒真是跟你说的一样
【在 N********n 的大作中提到】
:
: Exactly. Those big corps sing praise for FOSS only because they
: like the cheap labor. They have no problem taking away your hard
: work but you can forget about them opening up source code of
: their money makers to you.
|
z***e
发帖数: 5393 | 38 我觉得分歧在于：
1. Open source的人相信全世界志愿者一定能贡献比商业公司更大的力量，而类似我的
人就绝不信不拿钱的人能持续稳定真正拿出好东西来。不拿MS说，拿Google，今天
Google要开发Chrome，肯定会挖FF的人，如果把IE打垮，无非是把IE的名字换成Chrome
而已。
2. 我以前发过帖子问，成功地Open source软件到底有几个？出名的Linux/Apache/FF
，MySQL被谁bs了。然后呢？实际上所有稍微成功点的Opensourc也都是围绕Linux+
Apache这个web server系统在做，除此之外再没有了。哦，还有GCC是不错的。
【在 C********n 的大作中提到】
: IBM 倒真是跟你说的一样
|
C********n
发帖数: 6682 | 39
Chrome
FF
MYSQL is still open source , java is open source now , perl is open source
PHP is open source , EMACS is OPEN SOURCE ECLIPSE/NETBEANS GIMP
tons there
【在 z***e 的大作中提到】
: 我觉得分歧在于：
: 1. Open source的人相信全世界志愿者一定能贡献比商业公司更大的力量，而类似我的
: 人就绝不信不拿钱的人能持续稳定真正拿出好东西来。不拿MS说，拿Google，今天
: Google要开发Chrome，肯定会挖FF的人，如果把IE打垮，无非是把IE的名字换成Chrome
: 而已。
: 2. 我以前发过帖子问，成功地Open source软件到底有几个？出名的Linux/Apache/FF
: ，MySQL被谁bs了。然后呢？实际上所有稍微成功点的Opensourc也都是围绕Linux+
: Apache这个web server系统在做，除此之外再没有了。哦，还有GCC是不错的。
|
b********g
发帖数: 404 | 40 it is not ff's fault, blame the add-ons.
anyway, ff still costs less mem then IE. LOL
【在 N********n 的大作中提到】
:
: Exactly. Those big corps sing praise for FOSS only because they
: like the cheap labor. They have no problem taking away your hard
: work but you can forget about them opening up source code of
: their money makers to you.
|
|
|
b********g
发帖数: 404 | 41 你还是不懂这个东西的工作原则，
MS绝对不会为你一个小软件商提供底层bug修复的，
你去看看kernel版本控制是怎么实现的吧，
说个简单的例子，ubuntu新版本设计的时候难道是等从kernel到上层逐渐固定才行吗？
kernel没有固定人家就不能编译kde，gnome，ff了吗？
所谓open source不是让你全部拿出来open，
是你重新发行的时候必须open，你自己用，怎么改都行，
没人管你
【在 z***e 的大作中提到】
: 这不是问题所在，假设我是某个软件商，我知道vista有某某问题，我可以问微软了解
: 这个问题什么时候会解决或者根本不会解决，然后为新版本准备或者想办法work
: around，这不会造成我project的delay，也不会推出后要重来。
: 但是对于linux，我知道有这个bug，这个bug可能明天就被修复，也可能永远不会，我
: 的project今天可能不得不设计复杂的work around，但一旦被修复我的work around完
: 全变成白痴甚至形成新的bug.
: 我认为软件开发平台就是需要统一的标准，即便Windows垮台，Linux也自然会有各大公
: 司联合起来制定新的标准，Open source永远就是一个拿来利用的工具罢了。说穿了，
: 你们工作的code，会全部拿出来open source掉吗？
|
b********g
发帖数: 404 | 42
Chrome
这个是长流水，你不做有别人做就ok了，
linux kernel里面的代码有很多人只贡献了一次就消失了，
真正愿意做的都是凭借兴趣，
你应该知道，兴趣驱动要比金钱驱动更有成效
FF
这里还不如反过来说，你说哪里open source没有涉足的？
当然，论UI，论用户体验，我承认open source做得很差，
但是长进是有目共睹的，看看open source软件更新的速度，
再对比一下MS软件更新的速度，不需要我找数据了吧
【在 z***e 的大作中提到】
: 我觉得分歧在于：
: 1. Open source的人相信全世界志愿者一定能贡献比商业公司更大的力量，而类似我的
: 人就绝不信不拿钱的人能持续稳定真正拿出好东西来。不拿MS说，拿Google，今天
: Google要开发Chrome，肯定会挖FF的人，如果把IE打垮，无非是把IE的名字换成Chrome
: 而已。
: 2. 我以前发过帖子问，成功地Open source软件到底有几个？出名的Linux/Apache/FF
: ，MySQL被谁bs了。然后呢？实际上所有稍微成功点的Opensourc也都是围绕Linux+
: Apache这个web server系统在做，除此之外再没有了。哦，还有GCC是不错的。
|
c****e
发帖数: 1453 | 43 发信人: ayanami (螃蟹@FROSTSHOCK PWNS YOU!!11!), 信区: Linux
标 题: A 7 years bug!
发信站: BBS 未名空间站 (Fri Sep 5 19:53:04 2008)
https://bugzilla.mozilla.org/show_bug.cgi?id=86607
Geez fucking christ.
I am full of despise for Mozilla developers now!
【在 g*****g 的大作中提到】
: Which bug are you talking about? All bugs are here, why not show me
: which one.
: https://bugzilla.mozilla.org/
|
C********n
发帖数: 6682 | 44 no its not a bug
he is just asking for a new option
【在 c****e 的大作中提到】
: 发信人: ayanami (螃蟹@FROSTSHOCK PWNS YOU!!11!), 信区: Linux
: 标 题: A 7 years bug!
: 发信站: BBS 未名空间站 (Fri Sep 5 19:53:04 2008)
: https://bugzilla.mozilla.org/show_bug.cgi?id=86607
: Geez fucking christ.
: I am full of despise for Mozilla developers now!
|
N********n
发帖数: 8363 | 45
How is that not FF's fault when 3.0 with the same add-ons has
much less leakage. The one tab crashing the whole browser has
been there for years and still not corrected. Is that Add-on's
fault as well? LOL. IE8 runs separate processes, when will FF?
【在 b********g 的大作中提到】
: it is not ff's fault, blame the add-ons.
: anyway, ff still costs less mem then IE. LOL
|
b********g
发帖数: 404 | 46 do you think MS will fix a 7y old bug?
They even stopped support for XP.
【在 c****e 的大作中提到】
: 发信人: ayanami (螃蟹@FROSTSHOCK PWNS YOU!!11!), 信区: Linux
: 标 题: A 7 years bug!
: 发信站: BBS 未名空间站 (Fri Sep 5 19:53:04 2008)
: https://bugzilla.mozilla.org/show_bug.cgi?id=86607
: Geez fucking christ.
: I am full of despise for Mozilla developers now!
|
b********g
发帖数: 404 | 47 ok, you do agree FF3.0 is improving a lot.
If you want to compare mem usage, do the math yourself for IE and FF.
If you want to say multi-process, chrome has it also.
and chrome is true separate process. IE8 has multi-process,
but it is not one process for each tab. check it out yourself.
【在 N********n 的大作中提到】
:
: How is that not FF's fault when 3.0 with the same add-ons has
: much less leakage. The one tab crashing the whole browser has
: been there for years and still not corrected. Is that Add-on's
: fault as well? LOL. IE8 runs separate processes, when will FF?
|
z***e
发帖数: 5393 | 48 1. 这就是分歧，我的确不认为兴趣驱动会胜过金钱驱动----更何况，这世界上有拿着
钱做感兴趣的
2. 更新速度对商业软件毫无意义；即便对个人，我也就10年前会玩玩所谓“最新版”
。所谓的更新速度仅仅对免费玩票的“软件爱好者”有点吸引力罢了。
【在 b********g 的大作中提到】
: ok, you do agree FF3.0 is improving a lot.
: If you want to compare mem usage, do the math yourself for IE and FF.
: If you want to say multi-process, chrome has it also.
: and chrome is true separate process. IE8 has multi-process,
: but it is not one process for each tab. check it out yourself.
|
b********g
发帖数: 404 | 49
it is important, that's why google pays you to do whatever you want for 20%
of your working time.
i am not talking about minu update.
if you have a critical bug, open source software can fix it in an hour.
good example is the debian SSH week rng.
but it is possible for MS to fix the same kinda bug in weeks.
good example is windows messanger remote overflow execution.
【在 z***e 的大作中提到】
: 1. 这就是分歧，我的确不认为兴趣驱动会胜过金钱驱动----更何况，这世界上有拿着
: 钱做感兴趣的
: 2. 更新速度对商业软件毫无意义；即便对个人，我也就10年前会玩玩所谓“最新版”
: 。所谓的更新速度仅仅对免费玩票的“软件爱好者”有点吸引力罢了。
|
j*****h
发帖数: 2577 | 50 期待你去apple踢场子
【在 z***e 的大作中提到】
: 1. 这就是分歧，我的确不认为兴趣驱动会胜过金钱驱动----更何况，这世界上有拿着
: 钱做感兴趣的
: 2. 更新速度对商业软件毫无意义；即便对个人，我也就10年前会玩玩所谓“最新版”
: 。所谓的更新速度仅仅对免费玩票的“软件爱好者”有点吸引力罢了。
|
|
|
r****y
发帖数: 26819 | 51 ft。。。
MS哪里stopped support for XP了，依据？
【在 b********g 的大作中提到】
: do you think MS will fix a 7y old bug?
: They even stopped support for XP.
|
g*****g
发帖数: 34805 | 52 你是真不懂open source假不懂，外行不要来充内行行不行？
绝大多数的项目只有一个小圈子有commit权限，其他人可以
提交修改，这些人认可之后会提交进入正式build。
但是open source可以
1. 帮助你理解代码，有时候出莫名其妙的错不看源代码是
很难弄明白的。
2. 在没有workaround的时候自己fix。如果你不能等微软
6个月，那么有办法总比没办法好。这种事情我基本上3个
月干一次。而我们的C++小组，做outlook plugin的，每次
都会说这个干不了，那个干不了，因为M$ limiation。花了
大价钱定了M$的support，成天打电话，问题一样解决不了。
【在 N********n 的大作中提到】
:
: How is that not FF's fault when 3.0 with the same add-ons has
: much less leakage. The one tab crashing the whole browser has
: been there for years and still not corrected. Is that Add-on's
: fault as well? LOL. IE8 runs separate processes, when will FF?
|
g*****g
发帖数: 34805 | 53 不懂就别瞎鸡巴扯蛋。JEE里面这些年的主流框架，struts，spring, hibernate,
seam等等，哪个不是作为中间件的，那个不是open source的。不能用于第三方开发
我们都吃啥。
【在 z***e 的大作中提到】
: 举个例而已，当然不可能说我今天修改了apache一点东西，然后明天就成了新patch给
: 你下载。
: 但是我的重点是：Open source这种模式是走不通的。它无非强调两点：第一，全世界
: 都能修改错误，第二，你可以根据自己需要进行定制。问题是这两点都带出了一个问题
: ：第三方软件商如何适应？
: 作为一个OS，Linux为系统开发提供了便利；但作为一个开发平台，Linux为第三方开发
: 提供的可以说是零。
|
g*****g
发帖数: 34805 | 54 所有JEE的流行框架都是OSS的，你知道这些框架之上产生了多少系统吗？
肯定比基于.net的所有系统每年的revenue高就是。没常识也要会掩饰。
Chrome
FF
【在 z***e 的大作中提到】
: 我觉得分歧在于：
: 1. Open source的人相信全世界志愿者一定能贡献比商业公司更大的力量，而类似我的
: 人就绝不信不拿钱的人能持续稳定真正拿出好东西来。不拿MS说，拿Google，今天
: Google要开发Chrome，肯定会挖FF的人，如果把IE打垮，无非是把IE的名字换成Chrome
: 而已。
: 2. 我以前发过帖子问，成功地Open source软件到底有几个？出名的Linux/Apache/FF
: ，MySQL被谁bs了。然后呢？实际上所有稍微成功点的Opensourc也都是围绕Linux+
: Apache这个web server系统在做，除此之外再没有了。哦，还有GCC是不错的。
|
g*****g
发帖数: 34805 | 55 This is not true either, I used a lot of Apache license, LGPL license,
I don't have to open my codes.
【在 b********g 的大作中提到】
: 你还是不懂这个东西的工作原则，
: MS绝对不会为你一个小软件商提供底层bug修复的，
: 你去看看kernel版本控制是怎么实现的吧，
: 说个简单的例子，ubuntu新版本设计的时候难道是等从kernel到上层逐渐固定才行吗？
: kernel没有固定人家就不能编译kde，gnome，ff了吗？
: 所谓open source不是让你全部拿出来open，
: 是你重新发行的时候必须open，你自己用，怎么改都行，
: 没人管你
|
g*****g
发帖数: 34805 | 56 That's an enhancement request, and nothing to do with
memory leak. Developers may think this is not critical
and focus on more important ones, this happens in every
project, get a life.
【在 c****e 的大作中提到】
: 发信人: ayanami (螃蟹@FROSTSHOCK PWNS YOU!!11!), 信区: Linux
: 标 题: A 7 years bug!
: 发信站: BBS 未名空间站 (Fri Sep 5 19:53:04 2008)
: https://bugzilla.mozilla.org/show_bug.cgi?id=86607
: Geez fucking christ.
: I am full of despise for Mozilla developers now!
|
g*****g
发帖数: 34805 | 57 You need to learn better, open source doesn't mean you can't get money
out of it. Redhat, MySQL are just some examples. There are ways like
publishing books, training courses, developer certificates, techical
support etc. It's just different style. Only best open source products
thrive, so are close source ones.
The benefit of open source, as I said.
1. It helps you understand the api much better
2. You always have an option when the underlying api has a bug.
And more importantly, no close sourc
【在 z***e 的大作中提到】
: 1. 这就是分歧，我的确不认为兴趣驱动会胜过金钱驱动----更何况，这世界上有拿着
: 钱做感兴趣的
: 2. 更新速度对商业软件毫无意义；即便对个人，我也就10年前会玩玩所谓“最新版”
: 。所谓的更新速度仅仅对免费玩票的“软件爱好者”有点吸引力罢了。
|
x******g
发帖数: 3952 | 58 You can just keep on blaming other things for FF problem.
【在 C********n 的大作中提到】
: no its not a bug
: he is just asking for a new option
|
x******g
发帖数: 3952 | 59 MSFT is way better than you think.
【在 C********n 的大作中提到】
: Yeah
: and those smart ppl, when they got into M$, they can only produce
: shit like windows ME, windows vista
: guess M$ is worse than u thought , hmm ?
|
N********n
发帖数: 8363 | 60
Now you are spinning. I'm asking you how long it takes for FF
to abandon the wrong multi-thread architecture, what do you
talk about Chrome for?
Just face it: When you have a far-fetched, architecture level
problem, you have to spend a lot of time to fix it. Saying
that it'll be fast simply b/c it's open is nonsense.
And mem usage and leak are two different things. You can use
a lot of mem so long you release it when you don't need it.
FF 1.X, 2.X ate mem like a freak and refused to release it
【在 b********g 的大作中提到】
: ok, you do agree FF3.0 is improving a lot.
: If you want to compare mem usage, do the math yourself for IE and FF.
: If you want to say multi-process, chrome has it also.
: and chrome is true separate process. IE8 has multi-process,
: but it is not one process for each tab. check it out yourself.
|
|
|
g*****g
发帖数: 34805 | 61 Come on, since IE8 and Chrome have this mult-process architecture,
now all of sudden FF is wrong in using multi-thread? Can you distinguish
the difference between a feature and a bug?
Why not tell me when will IE8 use an easy to extend architecture
like FF so we'll have at least a nice adblock plugin? It's been
more than 10 years and M$ has no answer. I guess that architecture
change is much more far-fetched than changing thread to process.
Now let's get to something more fundamental. It takes 8
【在 N********n 的大作中提到】
:
: Now you are spinning. I'm asking you how long it takes for FF
: to abandon the wrong multi-thread architecture, what do you
: talk about Chrome for?
: Just face it: When you have a far-fetched, architecture level
: problem, you have to spend a lot of time to fix it. Saying
: that it'll be fast simply b/c it's open is nonsense.
: And mem usage and leak are two different things. You can use
: a lot of mem so long you release it when you don't need it.
: FF 1.X, 2.X ate mem like a freak and refused to release it
|
c****n
发帖数: 21367 | 62 well... from a security researchers' point of view
averagely, 1 out of 158 system vul. is found by report,
rests are found by incidents.
come to this case, bugzilla list might count for 10% of FF's
KNOWN bugs with very optimistic estimation
for the FF's development group, I had contacted to them right
after 1.0 released about some code flaws. they are generally
arrogant and aggressive RTFM type people back to that time.
maybe they are tech elites, but they lack of any sense what
is the value of
【在 g*****g 的大作中提到】
: It's not a bug if nobody finds it. It's open source and everyone
: can log bugs there, that's as complete as it can get.
: M$ likes to compare the number of bugs found over a year, and by
: their standard, if it's not a bug found by outsiders, it's not
: a bug, and they can sneak in and fix it in windows update without
: reporting to you. I am not surprised M$'s products always have
: fewer bugs.
|
b********g
发帖数: 404 | 63 You wanna try open 50 tabs in IE7, see what's going on.
【在 N********n 的大作中提到】
:
: Now you are spinning. I'm asking you how long it takes for FF
: to abandon the wrong multi-thread architecture, what do you
: talk about Chrome for?
: Just face it: When you have a far-fetched, architecture level
: problem, you have to spend a lot of time to fix it. Saying
: that it'll be fast simply b/c it's open is nonsense.
: And mem usage and leak are two different things. You can use
: a lot of mem so long you release it when you don't need it.
: FF 1.X, 2.X ate mem like a freak and refused to release it
|
g*****g
发帖数: 34805 | 64 IE doesn't have an adblock plugin. No matter how one optimizes it,
it won't match the footprint of Firefox. Lots of memory is wasted
for ads. and you can't do nothing about it. I often open 50+ tags
for firefox, and it runs for weeks without a crash.
【在 b********g 的大作中提到】
: You wanna try open 50 tabs in IE7, see what's going on.
|
H***3
发帖数: 821 | 65 Ahh, any FF's problem is actually not a problem... hmm... ever if it really
is, it must be MS's fault....
hmm...interesting...
【在 C********n 的大作中提到】
: no its not a bug
: he is just asking for a new option
|
x******g
发帖数: 3952 | 66 you forgot it could also come from add-ons.
really
【在 H***3 的大作中提到】
: Ahh, any FF's problem is actually not a problem... hmm... ever if it really
: is, it must be MS's fault....
: hmm...interesting...
|
x******g
发帖数: 3952 | 67 or user could learn more...
really
【在 H***3 的大作中提到】
: Ahh, any FF's problem is actually not a problem... hmm... ever if it really
: is, it must be MS's fault....
: hmm...interesting...
|
H***3
发帖数: 821 | 68 Just curious, how much time do you spend on patching open source softwares?
If you don't spend much, then why do I need that "freedom" to modify the
source code?
If you spend a lot of time, this fact alone already scares the shit off
average Joes like me.
【在 C********n 的大作中提到】
: 所有人都可以修改代码给自己使用
: 并不是所有人都有权利让自己的代码进入发行版
: 话说这里面的逻辑关系有那么复杂吗
: 你只要不公开发行，你随便patch,你要是想公开发行，就需要发送patch,请求审查
: 或者你干脆自己开个分支
: 没有限制你fix的自由吧
|
g*****g
发帖数: 34805 | 69 Bugs of OSS get fixed much faster than closed source in average, that's
simply a fact, so average joe does benefit from it.
http://en.wikipedia.org/wiki/Open-source_software
most studies show that open-source software does have a higher flaw
discovery, quicker flaw discovery, and quicker turn around on patches.
I am not saying all software should be open, but OSS exist there
for a good reason.
【在 H***3 的大作中提到】
: Just curious, how much time do you spend on patching open source softwares?
: If you don't spend much, then why do I need that "freedom" to modify the
: source code?
: If you spend a lot of time, this fact alone already scares the shit off
: average Joes like me.
|
r****t
发帖数: 10904 | 70 You need to understand the definition of "bug" for developers. Many
glitches cannot be called a bug if they are not reported and confirmed.
【在 x******g 的大作中提到】
: All bugs are here.
: 笑死人了。要是所有的bug都在这儿，mozilla就是天下最完美的software了。
|
|
|
x******g
发帖数: 3952 | 71 So what's not there is not bug. I get it.
【在 r****t 的大作中提到】
: You need to understand the definition of "bug" for developers. Many
: glitches cannot be called a bug if they are not reported and confirmed.
|
g*****g
发帖数: 34805 | 72 I don't know what point you are trying to make here.
Neverlearn pointed out there's a critical memory leak
not fixed for years in FF. I have hard time to imagine
that's not logged somewhere in the bugzilla so I want
to see which one and its status. And he immeditately
spin around.
Too bad IE doesn't have a public bugzilla, otherwise
there'd be tons of W3C incompatibility bugs which are still there
today. Or should we call it a feature, strategy?
【在 x******g 的大作中提到】
: So what's not there is not bug. I get it.
|
r****t
发帖数: 10904 | 73 That you do not need the 'freedom' doesn't mean others do not. You should
not complain because others do not complain. Your shit's off might not be a
bad sign either.
【在 H***3 的大作中提到】
: Just curious, how much time do you spend on patching open source softwares?
: If you don't spend much, then why do I need that "freedom" to modify the
: source code?
: If you spend a lot of time, this fact alone already scares the shit off
: average Joes like me.
|
c****e
发帖数: 1453 | 74 If something is hard to repro does not necessarily mean it's not a bug. Most
of the race conditions cannot be reproced deterministically. In a larget
distributed system, it takes a week to captures the stack trace of certain
conditions. But bug is bug, you call it glitche because you don't understand
it.
【在 r****t 的大作中提到】
: You need to understand the definition of "bug" for developers. Many
: glitches cannot be called a bug if they are not reported and confirmed.
|
g*****g
发帖数: 34805 | 75 I don't understand why we are arguing on this. Does
IE avoid this problem and FF can't? If you want to
argue not all bugs can be captured, be my guest,
but what's the point to argue a fact like that?
Most
understand
【在 c****e 的大作中提到】
: If something is hard to repro does not necessarily mean it's not a bug. Most
: of the race conditions cannot be reproced deterministically. In a larget
: distributed system, it takes a week to captures the stack trace of certain
: conditions. But bug is bug, you call it glitche because you don't understand
: it.
|
r****t
发帖数: 10904 | |
r****t
发帖数: 10904 | 77 it is a request to disable a function. But people vote for not disable that
function.
【在 C********n 的大作中提到】
: no its not a bug
: he is just asking for a new option
|
r****t
发帖数: 10904 | 78 Use your eyes and read that link. Please do not mislead.
【在 c****e 的大作中提到】
: 发信人: ayanami (螃蟹@FROSTSHOCK PWNS YOU!!11!), 信区: Linux
: 标 题: A 7 years bug!
: 发信站: BBS 未名空间站 (Fri Sep 5 19:53:04 2008)
: https://bugzilla.mozilla.org/show_bug.cgi?id=86607
: Geez fucking christ.
: I am full of despise for Mozilla developers now!
|
r****t
发帖数: 10904 | 79 They will. They attempted to do that already. They'll do that again.
【在 r****y 的大作中提到】
: ft。。。
: MS哪里stopped support for XP了，依据？
|
r****t
发帖数: 10904 | 80 Resorting back to base-less claim is not a polite practice.
【在 x******g 的大作中提到】
: MSFT is way better than you think.
|
|
|
x******g
发帖数: 3952 | 81 FF 1.0 support disappeared long time ago I think.
【在 r****t 的大作中提到】
: They will. They attempted to do that already. They'll do that again.
|
r****t
发帖数: 10904 | 82 Isn't this the whole point of open source in preventing excessive forking, right? If you
do not cooperate then you bear more burden.
【在 N********n 的大作中提到】
:
: Now you are spinning. I'm asking you how long it takes for FF
: to abandon the wrong multi-thread architecture, what do you
: talk about Chrome for?
: Just face it: When you have a far-fetched, architecture level
: problem, you have to spend a lot of time to fix it. Saying
: that it'll be fast simply b/c it's open is nonsense.
: And mem usage and leak are two different things. You can use
: a lot of mem so long you release it when you don't need it.
: FF 1.X, 2.X ate mem like a freak and refused to release it
|
r****t
发帖数: 10904 | 83 Does that bother you? If it bothers the users, it will be supported.
Debian supports hardwares that you might never heard of, because that
support matters to people who are still using it.
【在 x******g 的大作中提到】
: FF 1.0 support disappeared long time ago I think.
|
r****t
发帖数: 10904 | 84 Exactly, what's not found is not a bug. What's not repeatable is not a bug (
it might be several intervening bugs after confirmed). Developers cannot begin
working on that problem if it is not identified and confirmed.
【在 x******g 的大作中提到】
: So what's not there is not bug. I get it.
|
x******g
发帖数: 3952 | 85 Glad you said it. Go tell whoever I was replying with that post.
【在 r****t 的大作中提到】
: Resorting back to base-less claim is not a polite practice.
|
c****e
发帖数: 1453 | 86 I don't understand either. Whatever people are talking about , you bash IE
or anything from MS. Where did I imply IE is immune to these problems. It's
you who are trying to be a zealot.
Of course, not all the bugs can be captured without appropriate tools and setup. Repro is even harder.Memory leak in FF has been around at least 2-3 years before it's getting better, that's not a quick fix. Fixing a bug could be hard or easy. For easy bugs, OSS may be more responsive. But for hard ones, it's the |
g*****g
发帖数: 34805 | 87 It's funny you argue all day and cannot find a link in bugzilla to
prove your point: FF can't get a critical memory leak bug fixed in
years. All your argument is I hear this, I think that. I don't give
a damn. You have the easy proof there and you take the detour,
so much about zealot. Unless you find the link, this conversation
is over.
s
setup. Repro is even harder.Memory leak in FF has been around at least 2-3
years before it's getting better, that's not a quick fix. Fixing a bug could
be ha
【在 c****e 的大作中提到】
: I don't understand either. Whatever people are talking about , you bash IE
: or anything from MS. Where did I imply IE is immune to these problems. It's
: you who are trying to be a zealot.
: Of course, not all the bugs can be captured without appropriate tools and setup. Repro is even harder.Memory leak in FF has been around at least 2-3 years before it's getting better, that's not a quick fix. Fixing a bug could be hard or easy. For easy bugs, OSS may be more responsive. But for hard ones, it's the
|
x******g
发帖数: 3952 | 88 你为啥总一口咬定bugzilla里没有的就不是bug呢？
人民群众不是一整天都在给你讲这个不对么？就是不理解？
could
【在 g*****g 的大作中提到】
: It's funny you argue all day and cannot find a link in bugzilla to
: prove your point: FF can't get a critical memory leak bug fixed in
: years. All your argument is I hear this, I think that. I don't give
: a damn. You have the easy proof there and you take the detour,
: so much about zealot. Unless you find the link, this conversation
: is over.
:
: s
: setup. Repro is even harder.Memory leak in FF has been around at least 2-3
: years before it's getting better, that's not a quick fix. Fixing a bug could
|
Q**g
发帖数: 183 | 89 you don't bugzilla if you have no clue why it is happening. But please don't
pretend that the problem doesn't exist.
http://kb.mozillazine.org/Memory_Leak#Memory_leaks
could
【在 g*****g 的大作中提到】
: It's funny you argue all day and cannot find a link in bugzilla to
: prove your point: FF can't get a critical memory leak bug fixed in
: years. All your argument is I hear this, I think that. I don't give
: a damn. You have the easy proof there and you take the detour,
: so much about zealot. Unless you find the link, this conversation
: is over.
:
: s
: setup. Repro is even harder.Memory leak in FF has been around at least 2-3
: years before it's getting better, that's not a quick fix. Fixing a bug could
|
g*****g
发帖数: 34805 | 90 A critical memory leak that's not fixed in years is not logged
in bugzilla? Are you retarded? There are tons of bugs logged
even for 1K memory leak, why not this "CRITICAL" one?
【在 x******g 的大作中提到】
: 你为啥总一口咬定bugzilla里没有的就不是bug呢？
: 人民群众不是一整天都在给你讲这个不对么？就是不理解？
:
: could
|
|
|
N********n
发帖数: 8363 | 91 WTF? A one-tab-crash-all multi-thread architecture is not a
bad design needs fixing? You blew that Weapon of Mass Denial
a little hard. Get a grip!
Oh wait, I get it: Since you cannot log it as a bug, you don't
need to fix it no matter how friggin' stupid it is, right?
That's smart. Sure, it doesn't show up in the bug list, how
convenient, way to prove OSS fixes problems fast.
【在 g*****g 的大作中提到】
: Come on, since IE8 and Chrome have this mult-process architecture,
: now all of sudden FF is wrong in using multi-thread? Can you distinguish
: the difference between a feature and a bug?
: Why not tell me when will IE8 use an easy to extend architecture
: like FF so we'll have at least a nice adblock plugin? It's been
: more than 10 years and M$ has no answer. I guess that architecture
: change is much more far-fetched than changing thread to process.
: Now let's get to something more fundamental. It takes 8
|
g*****g
发帖数: 34805 | 92 WTF are you talking about? You can't commit code to FF
without establishing yourself, that's true. But you can always
log bug in their bugzilla. If a bug is critical and has existed
for years like you said, you must be blind to not find it in
their bugzilla.
Most people are still using IE7 and IE6, do they treat their
multi-thread design as a bug? You need to get the basic knowledge
between a bug and a feature. FF can improve on stability and
prevent crash at all, and multi-process is not necces
【在 N********n 的大作中提到】
: WTF? A one-tab-crash-all multi-thread architecture is not a
: bad design needs fixing? You blew that Weapon of Mass Denial
: a little hard. Get a grip!
: Oh wait, I get it: Since you cannot log it as a bug, you don't
: need to fix it no matter how friggin' stupid it is, right?
: That's smart. Sure, it doesn't show up in the bug list, how
: convenient, way to prove OSS fixes problems fast.
|
H***3
发帖数: 821 | 93 I remember many times I opened some website and IE just died on me. So I
killed IE in task manager and reopened the same website -- and magically it
just worked fine again.
I am so sorry for cursing your ansisters so badly, microsoft. I didn't know
it's not a bug.
(
begin
【在 r****t 的大作中提到】
: Exactly, what's not found is not a bug. What's not repeatable is not a bug (
: it might be several intervening bugs after confirmed). Developers cannot begin
: working on that problem if it is not identified and confirmed.
|
g*****g
发帖数: 34805 | 94 If IE sets up a public bugzilla, what you found probably would be reproduced
in millions of others. If you don't bark, you bet others will.
Now if IE is also open source, 1 out of 1000 users may try to trace down the
problem. 1 out of those 10 may eventually find the problem and even attempt
to fix it.
This is how this system works. A critical bug may not be easy to fix, but
there's no way it won't be reported when it has millions of users and has
been released for several months.
it
know
【在 H***3 的大作中提到】
: I remember many times I opened some website and IE just died on me. So I
: killed IE in task manager and reopened the same website -- and magically it
: just worked fine again.
: I am so sorry for cursing your ansisters so badly, microsoft. I didn't know
: it's not a bug.
:
: (
: begin
|
H***3
发帖数: 821 | 95 Wait a minute, I am confused. "open-source software does have a higher flaw
discovery, quicker flaw discovery, and quicker turn around on patches", is
that because OSSs get bugs fixed more efficiently, or it's simply because
there are too many bugs from the beginning? For an average USER, what's the
most important? A bug-full software with the "freedom" to fix it by your own
, or a relatively stable one with closed source?
I admit there are a lot of wonderful OSS softwares out there, and in fact
【在 g*****g 的大作中提到】
: Bugs of OSS get fixed much faster than closed source in average, that's
: simply a fact, so average joe does benefit from it.
: http://en.wikipedia.org/wiki/Open-source_software
: most studies show that open-source software does have a higher flaw
: discovery, quicker flaw discovery, and quicker turn around on patches.
: I am not saying all software should be open, but OSS exist there
: for a good reason.
|
H***3
发帖数: 821 | 96 The point by goodbug is: the "memory leak" talked here is actually not bug.
How about we call it a design flaw? I remember the main reason I stopped
using FF2.0 is simply it eats up my memory very easily -- it didnot crash on
me so I think it is not a bug too, but I uninstalled FF2.0 nevertheless.
【在 g*****g 的大作中提到】
: A critical memory leak that's not fixed in years is not logged
: in bugzilla? Are you retarded? There are tons of bugs logged
: even for 1K memory leak, why not this "CRITICAL" one?
|
g*****g
发帖数: 34805 | 97 They cached a page even when it's closed, if that's what you are talking
about. This is to re-open a closed page quickly. I can understand some
don't like it, but it's not a memory leak. I believe IE7 did something
similar.
.
on
【在 H***3 的大作中提到】
: The point by goodbug is: the "memory leak" talked here is actually not bug.
: How about we call it a design flaw? I remember the main reason I stopped
: using FF2.0 is simply it eats up my memory very easily -- it didnot crash on
: me so I think it is not a bug too, but I uninstalled FF2.0 nevertheless.
|
b********g
发帖数: 404 | 98 I am talking about GPLv3, the most recent version.
【在 g*****g 的大作中提到】
: This is not true either, I used a lot of Apache license, LGPL license,
: I don't have to open my codes.
|
b********g
发帖数: 404 | 99 Ok now, let me ask you how long it takes MSFT upgrade IE6 to IE7?
Let me ask you another question.
Why IE consumes the same, or even more mem is not called mem leak?
You can say FF2 has memory leak problem. Most people will agree in some kind.
But FF3 fixed it pretty well.
Now, if you return to the same problem on IE.
Nobody is talking about that. Why? Cos you can't do anything about it.
It is there for a long time.
No matter how you complain before FF2 came out,
MSFT did not bother to upgrade I
【在 N********n 的大作中提到】
: WTF? A one-tab-crash-all multi-thread architecture is not a
: bad design needs fixing? You blew that Weapon of Mass Denial
: a little hard. Get a grip!
: Oh wait, I get it: Since you cannot log it as a bug, you don't
: need to fix it no matter how friggin' stupid it is, right?
: That's smart. Sure, it doesn't show up in the bug list, how
: convenient, way to prove OSS fixes problems fast.
|
d****n
发帖数: 12461 | 100 mozilla的开源产品，例如firefox，是一个project。一个project只有一条主线，叫做
trunk。tinderbox每小时编译最新的源代码，每天有nightly，这样保证trunk是不断生
长的。假如某个东西要单独改进，通常就从某天的trunk搞一个branch出来。从这天开
始，以后branch的东西和trunk的可能不一样。改进完了，再merge到trunk里面。发行
日期是事先计划好的，例如每6个月或者12个月发布一个新版本。每次到了发行日期，
就从一个比较稳定的trunk或者branch出发，把大致功能freeze了，然后不断修正代码
，出alpha,beta,rc，最后到正式发行版。
这样对比许多闭源开发，好处是显而易见的。首先，bug能够在第一时间发现，然后按
照轻重程度，分配给不同的程序员修正。其次，功能可以逐步加入，不像闭源开发，版
本和版本之间只修正bug，功能上没什么改进。最后，万一方向错误，大家也可以及时
修正，不像闭源发展几年，结果大家都很失望，例如Win Me, Vista之类的。
call
【在 z***e 的大作中提到】
: 抛开MS不谈，我认为Open source的理念压根就是错误的。
: 什么叫有了bug你自己可以修复？人人都来修复，那真正的商家怎么去保证产品的稳定
: 性？我又怎么相信你的修复就是可靠的了？
: windows vista大部分的不兼容是因为以前的第三方软件多少利用了一些比较tricky的
: 技巧，例如说啊我知道这个指针肯定是这个位置，不必非要去用MS的API，我直接call
: 这个地址就好了。
: 同样的行为在linux也存在，而Open Source缺乏这种强制性的统一标准，说白了，开发
: 者要太监你也没办法。
: 引用一段slashdot上面的一个评论：
: Nah, like me, we just got older. When were were young Microsoft was pure
|
|
|
d****n
发帖数: 12461 | 101 当然不是这样的。
假如ms说明天不开发office了，不符合公司发展策略，不符合股东利益，怎么办？那大
家就只能继续用老版本了。但是如果是open source，假设原先的负责机构说不开发了
，后面还是有人接手，还有社区，最终决定是否继续开发的是用户和市场。这就是
chrome和ie的本质区别呀。
其次，open source不代表要做成最好的。如果都做成最好的，那还要什么商业软件呀
。open source也不代表免费的。开发者一样可以拿工资。
所以open source的理念是可持续发展。不像闭源项目，要受到公司业绩，股东利益这
样无关因素的影响。
Chrome
FF
【在 z***e 的大作中提到】
: 我觉得分歧在于：
: 1. Open source的人相信全世界志愿者一定能贡献比商业公司更大的力量，而类似我的
: 人就绝不信不拿钱的人能持续稳定真正拿出好东西来。不拿MS说，拿Google，今天
: Google要开发Chrome，肯定会挖FF的人，如果把IE打垮，无非是把IE的名字换成Chrome
: 而已。
: 2. 我以前发过帖子问，成功地Open source软件到底有几个？出名的Linux/Apache/FF
: ，MySQL被谁bs了。然后呢？实际上所有稍微成功点的Opensourc也都是围绕Linux+
: Apache这个web server系统在做，除此之外再没有了。哦，还有GCC是不错的。
|
g*****g
发帖数: 34805 | 102 开源软件只是不卖代码，而不是不卖服务。开源闭源跟软件质量没有必然关系。
同样的用户数目，开源的软件通常release和patch的周期都短，对终端用户有
好处。
【在 d****n 的大作中提到】
: 当然不是这样的。
: 假如ms说明天不开发office了，不符合公司发展策略，不符合股东利益，怎么办？那大
: 家就只能继续用老版本了。但是如果是open source，假设原先的负责机构说不开发了
: ，后面还是有人接手，还有社区，最终决定是否继续开发的是用户和市场。这就是
: chrome和ie的本质区别呀。
: 其次，open source不代表要做成最好的。如果都做成最好的，那还要什么商业软件呀
: 。open source也不代表免费的。开发者一样可以拿工资。
: 所以open source的理念是可持续发展。不像闭源项目，要受到公司业绩，股东利益这
: 样无关因素的影响。
:
|
