s******e
发帖数: 493 | 1 Used to write monte carlo simulation running tens millions of loop times.
Here was what I did. it might not be the best approach.
1.carefully check the logics for culprit. if failed,
2.then run the loop a couple of times and check all the major lines for
possible offenders. if still failed,
3.use bisect method to cut the number of loop times for each check. very
tedious... if still failed.
4.leave it alone and pray... |
|
l*********s
发帖数: 5409 | 2 I beg to differ. For multitasking, common users actually don't want be
bothered to memorize file names at all, and that is why GUI was invented.
Key-stroke wise, tile switching is more straight-forward and efficient as
well.Typing title name,even in incremental mode, is tedious at best: Lots of
people are decent at game controllers, but very poor at typing.
Searching winow/tab titles sounds amazing, but I think there are lots of
cases it does not work well.What about file name in Chinese |
|
u***t
发帖数: 3986 | 3 我目前在某Forture 100公司主持data migration的接收测试, 坑爹的offshore把简单
的问题搞大搞复杂来多要钱, 弄得新的数据库妖精一样...
问题: 我的approach还能再好点吗?
Source: SQL Server 2005
Target: DB2
如何测试Data Migration? 两种方法:
1. Sequential comparison
从Source DB得出个recordset A;
从Target DB的出个recordset B;
从A取第一个record, 到B用loop来找, 找到看B中的相关data是否正确.
(* 我枪弊了这个方案! Millions of rows, dozens of millions of fields. 一
个table 都要用无穷无尽的时间来execute. 而且很可能就out of memory. 除非是spot
check, 否则不可行)
2. Parallel comparison (*我的方案)
从Source DB得出个recordset A;
... 阅读全帖 |
|
r*******n
发帖数: 3020 | 4 If you use Ubuntu, I highly recommand the following:
https://github.com/fisadev/fisa-vim-config
It is pretty much easy to install which should save you from a lot of
tedious work of configurations that I don't even want to touch.
But if don't want to go with the above, you probably need to set
"filetype indent on" in your vimrc. |
|
t***t
发帖数: 6066 | 5 in the book, it listed 3 shortcomings for Scala:
1. strict static type
2. kind of tedious syntax
3. not pure immutable
but as I see, these are not real shortcomings. |
|
s********i
发帖数: 145 | 6 Some Important Theological Questions are Answered if we think of God as a
Computer Programmer.
Q: Does God control everything that happens in my life?
A: He could, if he used the debugger, but it's tedious to step through
all those variables.
Q: Why does God allow evil to happen?
A: God thought he eliminated evil in one of the earlier revs.
Q: Does God know everything?
A: He likes to think so, but he is often amazed to find out what goes
on in the overnite job.
Q: What causes God to interv... 阅读全帖 |
|
n*w
发帖数: 3393 | 7 下班浏览的一下goodbug给的那篇硕士thesis。先不说论文质量。通篇也没说linq是
syntactic sugar。
其中涉及linq和syntactic sugar有关系有两处。
1. A number of new features were added to C# 3.0 in order to provide LINQ.
Individually most of these features can be classified as syntactic sugar
that, while helpful in cutting down on the tediousness of repetitive code,
have viable – albeit more verbose – workarounds in the language。
这里讲到linq用了c# 3.0里syntactic sugar的语言features。最明显的应该是type
inference。很多情况linq返回的type非常复杂。type inference在这里是极方便。
2. The q... 阅读全帖 |
|
c****e
发帖数: 1453 | 8 It's always good you get a back and can rely on lower level infastructure.
Unfortunately, I got much experience working in the layer that people rely
on, it's very stressful and tedious ops kill dev's engergy in a big way.
Also the deployment model in your team assuems a good backend store? Amazon
offers cloud formation but still it's more challenging to upgrade data store
compared with app server.
built
of
that |
|
N********n
发帖数: 8363 | 9
Like someone already points out, asynchronous callback is basically
a 'goto' statement. If you can avoid goto in your code you improve
its quality a lot.
c# marks code w/ 'async' and 'await' keywords to tell compiler to
generate the tedious control code behind the scene for you. No need
of explicit asynchronous callbacks.
That way when you have to execute several asynchronous methods to
finish one job. You don't have to break it down into several "goto"
callbacks. You could wrap them up logical... 阅读全帖 |
|
t***t
发帖数: 6066 | 10 it seems google gradually abandon ship on javascript and move to dart,
another java-like static typed language for its UI. google almost abandon
its gwt totally.
this is the trend. dynamic language though is esier to start and small
projects, it cost much more to maintain and for large project.
UI, frontend, backend are all battlefields for different languages and
frameworks.
js is trying to invade into backend, and java is trying to invade into UI.
but java is tedious that's why java-like lang ... 阅读全帖 |
|
c*****e
发帖数: 3226 | 11 你们这些人还在这里纠结 docker 与 puppet 的区别，看大牛说的：
When small changes are needed to your servers, you can just use your CM tool
to manage those changes. Over time the images will diverge from your
current server configurations, so periodically you would create new server
images to keep them closer aligned.
This is a variant of the Golden Image pattern that allows you to have the
speed of using images, but helps you avoid the tedious image re-creation
problem for small changes. |
|
c********1
发帖数: 421 | 12 If your purpose is the isolation, I think docker is what you want.
Vagrant is a virtual machine manager, it allows you to script the virtual
machine configuration as well as the provisioning. However, it is still a
virtual machine depending on Virtual Box (or others) with a huge overhead.
It requires you to have a hard drive file that can be huge, it takes a lot
of ram, and performance can be not very good.
Docker on the other hand uses kernel cgroup and namespacing via lxc. It
means that you ar... 阅读全帖 |
|
p*u
发帖数: 2454 | 13
"天天做电脑前写java,c++的"
coding could be very interesting and challenging, or very tedious and
repetitive. i wouldn't mind trading some money for interesting work, better
with great potential. |
|
h*i
发帖数: 3446 | 14 不是大牛，我学Clojure以前没有接触过任何LISP。所以不懂scheme, 据我听说，
Clojure不是很纯粹的LISP, 除了list之外，还引入了其他核心数据结构，map, vector
, set，就是{}, [], #{}, 不像其他LISP只用()，所以把数据弄来弄去的倒腾很容易，
很直观；没有reader macro, 社区也不鼓励用macro, macro主要是高手们用来作轮子用
的，一般应用程序猿基本不用宏，所以有些LISP程序员不喜欢Clojure。不过Clojure的
用户也不是他们，而是我们这种本来用Java的人。
其他的，主要就是Clojure的数据缺省是强制性immutable的，没有变量的存在，scheme
应该不是这样的。这样的话，一般程序员学Clojure一开始，可能会觉得自己都不会编
程了，因为不能用for loop啥的。Clojure的for是list comprehension， 和for loop
完全不同， Clojure的loop recur倒是有点像for loop, 不过也是被皱眉头的。
当然Clojure用熟了，核心的那些map,... 阅读全帖 |
|
h*i
发帖数: 3446 | 15 我的主打语言是从Pascal -> C -> Java -> Clojure一路走过来的，我用Java的年头最
长。
总的说来，一旦过了FP一开始的那道坎，用FP编程和用IP编程其实区别真不大，高层的
思维是类似的，而底层都要干很多tedious shit，FP也没有什么magic的东东。
Clojure的好处，我个人觉得在于少一些worry。特别是程序大了之后，感觉改逻辑很容
易，一个个的函数而已，相互是没有关系的，不太怕搞出没想到的问题。
Java程序感觉更rigid，东西都是close-coupled，即使用了DI也改善不多，感觉改逻辑
很不容易，不可能只动很少的几个地方。 |
|
k******t
发帖数: 1498 | 16 模板确实有这个问题。自己从头开始，PSD->html+css,也有其他的问题，主要是respon
siveness处理麻烦。总之前端是非常恶心的东西。我一个2B的网站，功能说起来简单的
和小孩过家家一样，后台从架构开始到实现测试，API，不到一个月。运行将近10个月，
维护改进的时间不超过50个小时。前端beta花了好几百个小时才上线。上线半年改进累
计又是几百个小时，才算有个MVP。反反复复的各种tedious shit，各种back and fort
h。做起来还尼玛贼没有成就感。 |
|
d*******r
发帖数: 3299 | 17 看了下大家对 Meteor 的评论
https://www.quora.com/What-are-the-downsides-of-building-with-MeteorJS
- It has its own package system. The amount of wasted effort happening in
the community trying to resolve how to get the ###'s of existing NodeJS and
Bower packages to work with Meteor is just horrible.
- The way they handle hot-code reload using an Apache Cordova(-like?) mobile
web server imho is not a very good architecture, and I see it having long
term supportability and compatability issues. It also mea... 阅读全帖 |
|
d******c
发帖数: 2407 | 18 pandas rule of thumb: have 5 to 10 times as much RAM as the size of your
dataset
There are additional, hidden memory killers in the project, like the way
that we use Python objects (like strings) for many internal details, so it's
not unusual to see a dataset that is 5GB on disk take up 20GB or more in
memory. It's an overall bad situation for large datasets.
The 10 (really 11) things are (paraphrasing my own words):
Internals too far from "the metal"
No support for memory-mapped datasets
Poor p... 阅读全帖 |
|
d******c
发帖数: 2407 | 19 pandas rule of thumb: have 5 to 10 times as much RAM as the size of your
dataset
There are additional, hidden memory killers in the project, like the way
that we use Python objects (like strings) for many internal details, so it's
not unusual to see a dataset that is 5GB on disk take up 20GB or more in
memory. It's an overall bad situation for large datasets.
The 10 (really 11) things are (paraphrasing my own words):
Internals too far from "the metal"
No support for memory-mapped datasets
Poor p... 阅读全帖 |
|
z****e
发帖数: 54598 | 20
我们先找一个共识
就是在一个core上做
很容易，因为没有lock
对吧？这个是所有人都认可的
然后在这个基础之上，我们有两条路可以走
一种是直接上coroutine，要求coroutine解决多核环境下调度的问题
这就是go走的这条路，我相信这条路不好走
也需要“牛得不得了的人”才能搞定，很有可能
另外一条路是这样，你先放下这个问题，我们继续往前走
看看有没有其他的路，找啊找
诶，突然发现，有一条路叫做actor model
可以通过eventloop这种方式，把原来一大票线程给降下来
为什么呢？原理很简单，原先io的时候，同步api无论如何
要占用一个线程，给我等着
异步api的好处就是可以马上释放调用线程
这样我们可以复用同一个线程，然后对于我们常见的场景
也就是那些大量crud的业务场景
基本上都是io bound的场景，所以可以通过少量线程就能host住大量的请求
比如原来用tomcat，要几千个线程的线程池
现在只需要20多个就行了，甚至用1core 1g的机器，也就是2个eventloop，记住这个关
键字
以后你会经常遇到，spring都开始eventloop，spr... 阅读全帖 |
|
r*****k
发帖数: 9 | 21 相信不少读者和我一样受过或正在受自动广告窗口的折磨，特别是那种即使不上网也会跳
出来骚扰的广告更让人受不了。今天我正好有空，所以清理了一番，估计这往下几个月我
的机子会安静一段时间了，希望在国会出新的反自动广告窗口法案前会安然无事。下面说
说我是怎样做的。
1. Go to download.com to download Bazooka Spywar Scanner. It's free for one
month.
2. Install and run it. It gives a list of those bugs hidden in my local
machine and it tells me how to manually fix them, but unfortunately it doesn't
fix them for me. It doesn't matter! I fixed them one by one by myself. It's
tedious and a little bit technically challenging 'cause I had |
|
b*m
发帖数: 124 | 22 【 以下文字转载自 Physics 讨论区 】
发信人: eventhorizon (洒水车), 信区: Physics
标 题: Chinese name in revtex4
发信站: BBS 未名空间站 (Mon Sep 8 00:31:17 2008)
It looks quite tedious to me. Can any one post a tutorial or demo? I'm using
textshop on Mac. |
|
c*****t
发帖数: 1879 | 23 The argument against rm -i:
1. It gives false sense of security.
2. It is extremely tedious.
3. Sooner or later, one gets used to hit y automatically after
rm -i, defeating the original purpose.
On user's end, one should backup often.
Another lesson I learned the hardway was
rm * .o
I put such commands in scripts/Makefiles since then... |
|
w*******g
发帖数: 51 | 24 there is a tedious way that may make it.
first, find SUNWxcmf(?) package from solaris install disks (software disk 2 if solaris 8).
then extract it on your unix account, cpio, bz2.....
extract all the song**.pcf.Z to song**.pcf
transfer the files back to your PC, put hte fonts into the fonts diretory of your Xmanger or X-win32 |
|
s****l
发帖数: 78 | 25 Rich win and CStar are supposed to be able handle such chinese characters ba
h?
If it still can't, you can install a linux mount the fat32 used by win98 in
linux,
copy such chinese-name files into linux, under it change the chinese names t
o be
english ones, and put them somewhere on the internet, or other friends' mach
ines.
and enter win98, copy such files back.
somebody said in dos mode you can del the chinese-name files.
It's a very tedious way, but it surely works. |
|
m******3
发帖数: 983 | 26 Fund accounting
Higher pay compare to other corporate accounting positions.
Very busy and very tedious
Narrow future job market and narrow accounting skills. |
|
m********y
发帖数: 1867 | 27 各有利弊吧-
小firm的好处就是上手快,可以很快就cover很多accounts;但是小firm的客户规模小--
虽然五脏俱全,但终究是只麻雀;而且做audit testing也不是很多,经常只是vouching.
Big4不好地方就是林子大分工细,刚进去常常要做routine or tedious work,such as
footing, making copies, confirmations, etc. 但是客户规模大,见多了猪跑将来
inteview的时候也能说几句;而且audit testing方式比较多也比较好,能学些东西. 重
要的是放在resume上的经历好看.
当然,个人能力更重要,小的里面有出类拔萃的人,大的里面也有啥都学不到最后只能抱
怨的人. |
|
O****v
发帖数: 383 | 28 Just got back from the exam. Somewhat easier than FAR I have to say. Not
much tested on Biz Law, no need to memorize the exemptions. My 3rd testlet
was somewhat out of the ballpark, lots of theories I haven't seen when
studying, I suppose I did alright on the first 2 testlets to get a tricky
3rd one. Simulations are okay, not as tedious as FAR's. I did however have
to guess some of the answers simply becuz they weren't covered by my
study materials AT ALL! I just put 0 for everything I don't kno |
|
d********t
发帖数: 7 | 29 3/31 YE??
Seems like you are only looking at staff level. Try to learn from Partners,
Senior managers. How they think, how they talk, how they lead. Yes, I agree
there are some adminstrative and tedious work to do, its part of the job.
Hopefully you are lucky not to be on these jobs. Yes, I agree that there are
some ego idots in the firm but in 5 years, see if they are still with the
firm. Good luck.
percentage
required
not
is
intelligence |
|
O****v
发帖数: 383 | 30 IA is a dead end job, I've been stuck with it for almost 3 years... As much
as management likes to tell you that you can transfer to other departments/
positions down the road, it's just not that easy. Why? Most business units
hate dealing with internal auditors, cuz we always ask for stuff, sometimes
stuff that is so tedious even I feel bad for them... Alot of people end up
being career auditors, or going back for a MBA to try to change fields. I'm
not saying it's impossible to get out of IA, p |
|
m********y
发帖数: 1867 | 31 SEC reporting is very tedious and boring, and because of the strict
reporting deadline, it will require a lot of late nights at each quarter-end
/year-end. Think twice before you go for it.
Asset management/PE/Hedge is a good option. After a couple of years
experience, you will easily switch from public firm to industry, working as
fund accountant or fund administrator. |
|
m********y
发帖数: 1867 | 32 Blackstone, BlackRock,KKR这些都是大的PE,一点也不小....
很多大的bank也有自己的PE funds,比如JP Morgan,Insurance也有自己的PE,比如AIG...
PE/VC的fund accounting和一般的accounting不太一样,manufacturing accounting的
经验也不applicable,当然accounting basic rule/principle还是差不多的.所以不能
说"想去PE/VC作Accountant一般并不需要你以前做过。"
如果去banking/insurance,大部分都是SEC reporting,做过SEC reporting的都知道10K
,10Q有多麻烦,多tedious,很多都是cosmetic/wording change. |
|
s*******i
发帖数: 741 | 33 I think that unless you could get in a top PhD school. It is not worth
going.
To get in a good research school, you need to be from a top PhD program and
also really good at research. If you go to a teaching school, the salary is
not very high compared to you can get at industry with 10 years big four
experience. A PhD program will take at least five years for a good PhD
program. You will be 40 which may put you in a disadvantage because a lot of
research schools will doubt your research potent... 阅读全帖 |
|
s*******i
发帖数: 741 | 34 I think that unless you could get in a top PhD school. It is not worth
going.
To get in a good research school, you need to be from a top PhD program and
also really good at research. If you go to a teaching school, the salary is
not very high compared to you can get at industry with 10 years big four
experience. A PhD program will take at least five years for a good PhD
program. You will be 40 which may put you in a disadvantage because a lot of
research schools will doubt your research potent... 阅读全帖 |
|
s******u
发帖数: 225 | 35 我觉得international tax做的东西比较有意思，很多都需要law background。
partnership/financial service 很多时候很tedious，刚开始做就是process很多K-1
什么的，只有到了比较高层才能接触高级的内容。corporation tax做provision什么的
能学到很多，而且也比较好跳槽 |
|
i*******n
发帖数: 227 | 36 I hate this job more and more, which is so boring but with big
responsibility. I always spend days pulling hair to find out where the
balance difference is....Nobody cares about how tedious and difficult this
jobs is but only pushes you to get things done by deadline or way before
dealine especially when it's about the holidays....The bit*** boss makes you
input, reverse, fix the journal entries again and again like crazy, but has
zero tolerance about your mistake, which they consider as "potent... 阅读全帖 |
|
i*******n
发帖数: 227 | 37 I hate this job more and more, which is so boring but with big
responsibility. I always spend days pulling hair to find out where the
balance difference is....Nobody cares about how tedious and difficult this
jobs is but only pushes you to get things done by deadline or way before
dealine especially when it's about the holidays....The bit*** boss makes you
input, reverse, fix the journal entries again and again like crazy, but has
zero tolerance about your mistake, which they consider as "potent... 阅读全帖 |
|
l*****k
发帖数: 587 | 38 Well, I am sure I am not the first to do this, but I do want to share it
Sometimes, cloning multiple things into a vector can be tedious as you
have to transfer the fragments through several vector to get the
desired one, usually because the existence of the same Restriction site
in the vector and the fragment you want to clone.
say a gene has 2 frags in two separate vector, now you wanna get
them ligated into the same vector.
EcoRI------------XhoI Frag I (has interal BamHI site)
XhoI- |
|
o2
发帖数: 2764 | 39 Yes, they can do smaller size crystals, but what else? Do they still need
crystals? Yes. As other people pointed out, grow smaller crystals are not
much easier than big ones.
Plus, they will need tons of small crystals at the same size range, is that
easy? No. So it has some advantage, but I did not see any big breakthrough.
People still need to generate crystals, which is hard and tedious. |
|
d*****r
发帖数: 2583 | 40 I think the bottle-neck is not bioinfo itself, or statistical
methods/algorithm, or even math if we put aside the philosophical
question about whether biology can be understood by math.
The bottle-neck is high-throughput methods to acquire the data,
including
large-scale perturbing/imaging tools on molecular/cellular/animal level.
This requires huge progress on
instrumentation/nanotechnology/microfluidics/automation/material
science/surface chemistry, etc. That's why Feng Zhang was so popular on... 阅读全帖 |
|
b*******1
发帖数: 127 | 41 Data, samples, platforms, and even experiments are becoming commodity. You
can easily buy all these online. A lot of them are even free. As long as you
can propose important biological/clinical questions, you can get all
process done without any single experiment by yourself. For example, with
several gene/markers in hand and offer money and co-authorship for
validation, any famous specialized lab would happily take the offer. These
commodities freed us bioinformatics scientists from doing tedio... 阅读全帖 |
|
H****s
发帖数: 301 | 42 Usually people achieve this by phage/yeast antibody display technology. You
need to have a very good antibody library and yeast/phage library panning
capability. Once you have a pool of selected binders, you can screen them by
ELISA. In your case, two ELISA can be performed in parallel, one coated
with your target protein, the other coated with negative target protein. The
ELISA signal will give you a rough idea about target specificity. If you
have access to flow cytometry, combining flow sorti... 阅读全帖 |
|
e*****t
发帖数: 642 | 43 give your whole RNA extraction to seq facility for library preparation. it's
tedious to do it by yourself.if you have time, spend more time on
downstream bioinfo. that would earn you more no matter in publication or job
hunting in the future. |
|
g****1
发帖数: 261 | 44
That's the way that science is supposed to be done. let others do the
tedious labor work :) |
|
s****t
发帖数: 461 | 45 虽然这种工作没有太多的创新成分, 但是毕竟对research community还是很有用的。
要是不能发一片好一点的paper,这种tedious dirty work就更没有人愿意做了。
not every nature/science paper is the same anyway.
Col-
of
assessed
genes |
|
h*****s
发帖数: 153 | 46
My previous boss for my undergraduate project course is a cholesterol person
, it would be easy to get cell samples from his lab to do some lipidomics
analysis. My supervisor has connections with the Cancer research center
affiliated to our uni, some potential collaborations might be involved.
I quite like biology, but the wet experiment is just so tedious for me. |
|
n********k
发帖数: 2818 | 47 depends on your exp...anyway, the most tedious and tiring work ever for me
as much as isotope work...the problem is you have to be 120% careful all
time...I actually don't trust to let others do it for me... |
|
K******S
发帖数: 10109 | 48 this thread is interesting. ppl here advocate quiting bio field everyday and
claim it's tedious, labor intensive. Shouldn't you be happy to see what's
happening in China as you discribed? |
|
