S**I 发帖数: 15689 | 1 you're right, H-1B and H-4 are decoupled now. |
|
n****i 发帖数: 1772 | 2 BSO的亮点是asset 增长和工资decoupled吧. 楼主隐形BSO自己是投资高手。
不如加一项: 可用来投资的资产。 |
|
m********8 发帖数: 7463 | 3 一般傻子确实理解不了decouple from the system 的含义。 |
|
f***y 发帖数: 30 | 4 【 以下文字转载自 Stock 讨论区 】
发信人: guvest (我爱你老婆Anna), 信区: Stock
标 题: [bssd] 结局已经很明确了
发信站: BBS 未名空间站 (Mon Nov 19 02:47:48 2018, 美东)
本版不少老站友。那么这些老站友的共同点是什么呢?说了你们可能不信。也完全感觉
不到。大家的共同点其实是对语言-世界的关系的信仰。没有这种信仰,一个人不可能
长期呆在老刑这个破站。这个破站可能是历史最长的没有敏感词自我审查的站了。
语言是人思维的基本工具,也是一个人对世界的基本模型。是一个社区,城市,国家也
是如此。但是就跟其他工具一样,语言这个工具也会造成人的异化,引起社会的前进或
者后退。
那么我国现在的基本语言是什么呢?我们有两套官方语言在两端,一套是习语近人,厉
害国,梁家河大学问。另一套是什么财务自由,创业,做大上市等等。
在这种语言割裂的情况下,社会现实的割裂情况只可能是更严重。显然,这种水平的一
个国家和社会是不可能担任世界领袖的。即便引领区域国家共同发展也不可能。我讲的
,不是一个“得道多助”的意思。而... 阅读全帖 |
|
|
h*******g 发帖数: 2201 | 6 Thanks for all the helpful posts that educated me.
My plan is to build a dream rifle AR for myself, shooting in the range and
try to improve my skill and defense in the 1/1000,000 chance of the so
called SHTF scenario. I would like to spend somewhere around $2000.00 to get
the best out of it. And I would like to order individual parts instead of
assembled upper to build to my liking, as Zman941's post said. I would not
care much about lower receiver, just want it to have a tight fit with the
upp... 阅读全帖 |
|
b*******t 发帖数: 33714 | 7 你这未必讨厌做饭,是讨厌目前自己做不好饭的状态,然后恶性循环
跟有些人减肥其实是一个性质。
不喜欢炒菜做别的就是了
我有时候默默觉得要自己能做好了再去说不喜欢才能理直气壮
不然没法把喜好和受挫情绪decouple |
|
h*h 发帖数: 27852 | 8 summary:
从小培养,以培养基本足球技术为主,少打比赛也不在乎比赛成绩,通过出卖成品青年
球星赚钱
How a Soccer Star Is Made
The youth academy of the famed dutch soccer club Ajax is grandiosely called
De Toekomst — The Future. Set down beside a highway in an unprepossessing
district of Amsterdam, it consists of eight well-kept playing fields and a
two-story building that houses locker rooms, classrooms, workout facilities
and offices for coaches and sports scientists. In an airy cafe and bar,
players are served meals and visitors can have a... 阅读全帖 |
|
d**y 发帖数: 18174 | 9 太牛了。这是干什么用的?两个多余的decoupling capacitors短路了? |
|
d**y 发帖数: 18174 | 10 果然是电容短路。我有点怀疑它是滤波电容。滤波电容通常比较大,你要是把它取下来
相当于断路,L-C滤波还怎么work?Decouple的电容去掉倒是通常无所谓,只是增加一
点波动,只要设计裕度足够就还好。
“接上2a的充电器紧急充个几分钟应付”有点奇怪。一般充电器的电流读数是Imax,即
支持的最大电流。充电过程中,通常regulate电压,不会regulate电流。根据症状判断
,或者短路在先,充电在后;或者充电器电压超过范围,比如正负5%合格,也许你的充
电器是正负10%。 |
|
|
h******g 发帖数: 11250 | 12 应该可以,这个问题全在机身顶着镜头的光圈拔杆
如果decouple了,应该就一样了
所以理论上所有转接镜头不能全开测光的都能用A档? |
|
|
S**C 发帖数: 2964 | 14 No. In such a way IC can run at most efficient point, decoupled with vehicle
speed and acceleration etc. |
|
g****t 发帖数: 31659 | 15 你不和对象有物理接触,如何观察?所以你设想的观察是纯数学的映射。不是物理观察
。只要是物理观察,那观察的作用在某些物理过程中就不能忽略。
观察的作用能不能彻底从物理世界中decouple或者分解掉。据我所知现在还不能。
再说了,你假设一个独立于所有观察和描述之外之外的存在有意义么?这可是个大问题。
一朵花,一棵树,一张桌子既然独立存在于所有的其他action,event....之外。那么,
你假设它的存在干嘛呢?这不是多余的假设么!这是博尔赫斯一篇小说对康德朴素的“
物自体”概念的评论。
按照我理解的海德格尔的说法。任何一个自然学科所处理的对象,也就是“存在”。其
本身就嵌入了做这个职业的人的common agreement。就已经不是独立于所有观察和描述
之外的客观存在了。只能是有限定的“物质存在”才能开始一个学科或者理论。要真正
理解存在,先要搞一个所谓的现象学还原过程,理解对存在有所思考的“此在”才行。
以你下面的话为例,你所说的“物质客观存在”是由你下面这段话来例证或者定义的。
但这并不排斥别的定义“物质客观存在”的可能。这就是说,不同对“存在”的看法的
学科,不同的理论都有产... 阅读全帖 |
|
T*******y 发帖数: 6523 | 16 Hi, freeman08:
Thanks for your clarification.
This is a very good clarification, and let's talk about the following
opinions.
时间上不可变的那种关系,一个事情必然引起另一个事情的关系。
When I used the word "order", I meant their relative sequence, or perhaps
more in time (of course the term "time" can be defined and talked more later
), but not a definite causality.
Theories in this physical world can be thought of as simplified models,
which tried to decouple the effects from many other things. However, this
world is inter... 阅读全帖 |
|
|
q**w 发帖数: 782 | 18 http://www.nature.com/nature/journal/v461/n7268/abs/nature08470.html
Letter
Nature 461, 1265-1268 (29 October 2009) | doi:10.1038/nature08470; Received
17 June 2009; Accepted 27 August 2009
Preserving electron spin coherence in solids by optimal dynamical decoupling
Jiangfeng Du1, Xing Rong1, Nan Zhao2, Ya Wang1, Jiahui Yang1 & R. B. Liu2
1 Hefei National Laboratory for Physical Sciences at Microscale and
Department of Modern Physics, University of Science and Technology of China,
Hefei, Anhui 2 |
|
s******y 发帖数: 28562 | 19 哇,听起来很牛的样子。
赞一个
Received
decoupling
China, |
|
s******y 发帖数: 28562 | 20 哇,听起来很牛的样子。
赞一个
Received
decoupling
China, |
|
T*********r 发帖数: 11175 | 21 哦,呵呵
不过我们这专业,澳洲有点散兵游勇了
完全跟世界decoupled |
|
v***v 发帖数: 5504 | 22 你这个回帖以及楼上说所有110变压器都有致命威胁贴都属于无营养强辩帖。并不能让
消费者感到更安全。
作为消费者,有必要把这个问题弄清楚。如果真的是苹果的设计问题,估计被电死一方
一定不会放过机会通过法院讨个公道,当然山寨充电器和正品充电器又有区别,但楼上
又有人说一般来讲,山寨充电器还更安全,如果两级绝缘做得好些的话。拭目以待后续
官司的说法。
另一方面来讲,如果真的是苹果设计有问题,俺很奇怪咋水版里还没有10大的帖子,呵
呵,俺老还是比较看好灌水人士科研水平的。交流电压是正弦曲线这个没错,好像依稀
记得3相的还能到300多伏。
有以下重点疑问:
1. 变压器output端地线和iphone金属壳相连?答案好像是Yes。但没发现这个有啥问题
,继续后边的问题
2. 变压器input端线圈和电路和output端线圈及电路应该是decouple的吧?如果是yes
,就没啥问题了。前端烧穿开路也好,短路也好,后端也没影响啊?
3. apple到底是用的陶瓷电容还是金属电容? 真正击穿的时候是开路还是短路?有人
说金属电容击穿瞬间也是短路,这个瞬间短路有无可能传递到output端(取决于问题2... 阅读全帖 |
|
l*s 发帖数: 783 | 23 I understand the main purpose of PRISM is decouple components. And
maintainability of your own code should be better. But that comes a price
that you are totally depends on a open source library(or third party
component) to do a lot of things for you and the question is that whether
this library is too big to fail? The worst case is that if this library's
development ceases, can your team maintain it(fixing bugs) by yourself? or
can you easily use another similar library to replace it? That's wh |
|
N********n 发帖数: 8363 | 24
只要使用AZURE上的SERVICE就要调用其REST API,只要调用REST API就要
走ASP.NET的SERVICE。调用方和AZURE SERVICE在结构上是DECOUPLED的,
根本就不存在啥“.Net上跑JDK”。你根本狗屁不通还胡扯。 |
|
N********n 发帖数: 8363 | 25
只要使用AZURE上的SERVICE就要调用其REST API,只要调用REST API就要
走ASP.NET的SERVICE。调用方和AZURE SERVICE在结构上是DECOUPLED的,
根本就不存在啥“.Net上跑JDK”。你根本狗屁不通还胡扯。 |
|
l*s 发帖数: 783 | 26 ☆─────────────────────────────────────☆
runPython (凸-.-) 于 (Mon Oct 15 00:29:15 2012, 美东) 提到:
还在犹豫:
语言上C#强于Java;
框架上MVC的ASP更易用,
但是JAVA的基于开源的生态系统已经很强大,
大公司都在用,找工作前景很不错。
感觉上java和j2EE的是中大公司搞,
C#和ASP是中小公司搞或者非IT的大公司。
可以看出来薪水还是有点差别的,平均来说。
JAVA略高
☆─────────────────────────────────────☆
a9 (嗯) 于 (Mon Oct 15 08:33:29 2012, 美东) 提到:
搞电子商务的很多大公司都在用.net
☆─────────────────────────────────────☆
NeverLearn (24K golden bear) 于 (Mon Oct 15 11:06:35 2012, 美东) 提到:
Java is paid high simply b/c it's c... 阅读全帖 |
|
N********n 发帖数: 8363 | 27
无脑MVC。易TEST, 易SCALE, 易DECOUPLE。 |
|
B*****i 发帖数: 80 | 28 前辈们,肿么理解下面的话,
"The essential difference is that EF allows for stronger decoupling between
the database schema and the classes that you query. Instead of querying
classes that closely represent the database schema, you query a higher-level
abstraction described by an Entity Data Model. This offers extra
flexibility, but incurs a cost in both performance and simplicity.”
能给个example吗? |
|
l*s 发帖数: 783 | 29 也不全是。这两天我仔细想了一下pros and cons.觉得从maintenance的角度讲这种在
controler中直接用ORM的做法还是不可取。按照他的逻辑IOC,Logger等所有framework
都不需要abstract和decouple了。结果回到了classic asp年代。 |
|
f***o 发帖数: 31 | 30 expression language, so that you can write scripts like ${foo.bar}.
those are supposed not java, but 'simpler' language for page authors.
SimpleTag or 'tag file'. the taglib can be implemented in JSP syntax,
not have to be in a java class. this is a lot easier.
SimpleTag is decoupled from servlet context, so that it 'COULD' be
used as template page independently in 'FUTURE'. I want this badly,
but still it's not available. |
|
f***o 发帖数: 31 | 31 apache group has quite some influence, but i don't think velocity is big deal.
(1) is driven by persistent request to cleanly seperate developers and
authors. if you ask me, screw page authors then fire them.
(2) is a correction to the previous design mistake. i don't think they were
too stupid to forsee the problem, probably no enough resource at the time.
(3) is not quite clear - seems they intend to decouple template from servelt,
but it's still not done yet, just a gesture. ahh... |
|
g*****g 发帖数: 34805 | 32 讨论一下这个问题,用一个singleton和用一个所有成员都是静态的类,
有哪些pros/cons。感觉上singleton可以实现interface,而静态类不行,
所以singleton在decouple和dyna binding上可能灵活一些。但如果我
只有一个实现呢?
比如我要做一个web service, 这个service接口很简单。类似于
Object process(int ID, Object input)
给个ID,去数据库里去取点相应数据,对input做处理然后输出。
我可以写一个bean实现这个方法, 用spring让它成为singleton来干这个活。
我也可以写一个bean,套接调用一个静态方法来干这个活。
有什么区别吗? |
|
g*****g 发帖数: 34805 | 33 Remote management is not new, every Java enterprise application has some
management components. IMHO, JMX is new in providing a standard way
to do management, therefore decouple management code from your applications.
Plus, it's sort of self descriptive and dynamic (similar to UDDI in
webservice)
so you may use some standard JMX agent to do the management, without knowing
the business logic of your management, that may save quite some code.
Plus, change some properties and make them effective wi |
|
b******y 发帖数: 1684 | 34 in this case, what's the difference between using jsp and servlet?
using jsp with frameworks like struts2, you can also decouple
business logic from presentation. |
|
j*****I 发帖数: 2626 | 35 理论性的探讨一下。觉得是,狗了一下,没有发现有人提到过这个优点啊?! |
|
g*****g 发帖数: 34805 | 36 I never try MongoDB, we do this for Cassandra. And I
can replace the DB by simply rewriting DAOs.
I think a flexibile architecture can save you
lots of trouble in the long run. 3 layer architecture
is for decoupling, not for simplicity. |
|
g*****g 发帖数: 34805 | 37 I can't agree with your approach. The relevant characteristics of OOP here
is encapsulation. Encapsulation means the class maintains some states, and
define some behaviors to operate on these states, it doesn't have
intelligence, that's not part of OOP.
J2EE deals with complicate problem. And it typically has layered structure.
Most complicate systems in the world are using a service-oriented
architecture. But that still doesn't deviate from what OO is about. Every
single service maintains state... 阅读全帖 |
|
|
g*****g 发帖数: 34805 | 39 I don't know what you are smoking. But I don't think you can convince anyone
Class.forName().newInstance() is not a reflection call.
Factory Pattern doesn't have to use reflection. True, but only reflection
can achieve the decoupling that separates the factory from knowing what it's
gonna create. This is a huge advantage over those languages that do not
have reflection.
It's not about statistics. This one reflection call is critical in framework
design.
from |
|
g*****g 发帖数: 34805 | 40 Of course it's still MVC, MVC is just a pattern to decouple view from model.
The difference is where C is running, in front end (JS) or in backend. |
|
d****i 发帖数: 4809 | 41 这个也是为什么我还是prefer有web.xml的原因,conf起来比较灵活方便,而且跟code
完全decouple。 |
|
z****e 发帖数: 54598 | 42 decoupling啊
这个是造成memory leak的主因
不过spring要用你两天吗?
我当初搞spring,就用了半个小时就搞定了 |
|
z****e 发帖数: 54598 | 43 以前c时代,你要负责管理所有内存对象的生命周期
一旦忘记释放内存,很快就leak了,就挂了
java提出之后,java说,这个交给jvm来干
然后优化jvm,从最早的reference count到现在的内存管理
其实挺复杂的,而且还在演化中
但是基本上说是,只要内存释放得当,绝大部分的内存都可以很好地被管理起来
这是一大进步,大大减少了内存泄漏的可能性
spring则提出了更高级的一步,spring说
你现在不仅内存释放不用管了,内存分配你也别管了
我来管,我来负责new这些对象,你只需要把逻辑部分代码写清楚
剩下的我来搞定,这样的话,内存泄漏几乎不可能存在
除非你蠢到家了,其实有了spring,再搞fp会很容易
java通过反射来变相搞fp,但就是fp本身不太容易理解
思路要转变,这个我不太赞同,oop更接近人的思维,而不是数学家的思维
等真搞数学,比如统计的时候,你再考虑用其他的语言也不迟
不管怎样,spring基本上把所有的beans生命周期都搞定了
spring理论上把你内存的利用tune到最优了
解决了不仅仅是decoupling还同时优化了内存的使用
不过这里面有一点点风... 阅读全帖 |
|
b****u 发帖数: 1130 | 44 我用python 而不是JAVA,但是这两个语言也差不多:
类似于multithread,由主线程控制子线程,主线程可以打开并运行多个子线程,控制子
线程。
那么在主程序中,运行到某语句,该语句是一个子程序,子程序是打开一个terminal,
并运行iperf -su (就是服务器端听)。然后该程序不会在这里hang住,而是接着执行
下一条语句,这条语句又是一个子程序,子程序打开一个terminal,并运行iperf -c (
就是客户端连接等等)。程序也不会在这里hang住,而是接着执行下一条语句。
大家明白我的意思了吗?如果用multithread实现很容易,但decouple不好,我不会采用
那么如果使用.batch文件,每个terminal的执行放在一个.py文件,然后
a.py &
b.py &
可能也能实现,但我不希望用这种方式。
我希望使用process,就是上述方法来实现,求帮忙 |
|