（能力）zt 套磁的看法

北京的争争

来美国三年了,应该说作为一个研究生的经验也积累了不少了,因为平时和老板处得也不错
,所以也耳濡目染了一些美国研究生院招生的程序和规则.这几天没事情逛了逛国内几个大
的留学申请论坛,看见大家都在讨论陶瓷,心里有一些话不得不说.

6 {9 g. \* K( I1 O. p7 ^, u2 G( ^陶瓷这个词现在已经几乎成了所有留美申请朋友的关键词乐,没有人不知道的.甚至还能看
见一些申请文科如教育，心理学的xdjm也在bbs上面问陶瓷有没有用，怎么陶瓷等等。这关
6 L4 W' a$ r6 @3 {键的原因是在于，每年都有大批的中国留学生通过和教授联系，也就是陶瓷，拿到offer。
但是，看今年的神情形式，陶瓷已经有泛滥的迹象，甚至听说有同学雇佣中介发大批的一
4 o  J: ?% c: d) J( S0 w抹一样的陶瓷信给教授，以求“普遍撒网，重点培养”。还有的同学同时给一个学校的三
个以上的教授发陶瓷信；还有的同学的陶瓷信肉麻之极，照老美的说法一看就是个'kiss-
; ?) k9 C$ F2 S8 G# r5 Kass'...这些不仅大大影响了中国留学生的整体形象，对于个人的申请也是非常非常的不利
。我接下来就会给大家讲讲其中应该注意的地方。
0 w( }0 j+ A- q, G% k7 I
, r! T9 S* S8 X# k& Y6 }) J第一，所有的申请人都有权利给自己感兴趣的科研组负责人，也就是系里的教授，发联系
" h) ~- J0 i& ~- g6 A7 f信，询问是否有意向招学生。很多朋友对这一认识有误区，认为发这样的信就是对他人的
# W0 H9 c* k5 U& ^$ U; J% i一种骚扰，是不对的，所以很鄙视。其实不然。你既然申请了这个学校或者对这个学校感
! H( Q3 h8 F0 @  ]兴趣，当然首先要确定你感兴趣的研究组里面是不是有可能有你的位置。这样的询问信是
4 X, q* E4 U6 `1 L很自然，很职业，也是很平常的事情。但是要注意，有些系或者有些教授在网页上面会说
清楚，不欢迎任何形式的私下联系，那么就只能通过网页获取你想知道的信息。不要再勉
8 O5 Q* s& E0 c强他人。同时,在寄出询问的email之后，如果几天都没有收到消息，那就说明教授对你确
实不感兴趣，那么就不要在发信骚扰了。很多美国教授对这个头痛就是因为很多学生在发
& Z$ M7 I& i8 r9 r% i陶瓷信上面确实是“孜孜不倦”，有的教授甚至不得不回一封措词比较强硬的信，彻底断
5 C4 i( S9 ]  e4 W3 i! [% g了这个学生的念头，来逃避他的骚扰。。。所以，请所有陶瓷的朋友，在寄出第一份信
& [' `8 u/ h2 V1 Q! W之后，做一个记录，如果没有回复的话，也不要在发一模一样的信了。email丢失的可能性
+ G" I' b& t1 |! Z: R& ~是很小的，你没有收到回复的原因就是教授不想找人，或者对你没有兴趣。

. P: ?7 C% u5 x" e8 j第二，陶瓷信不是请求信，双方是基于平等的交流。很多人的陶瓷信简直肉麻之极。要知
: V5 @+ s' ]4 }道，这样的语气其实只会招人反感，因为这一看就是没有经验的人。真正有科研经历，有
和其他学校教授交流经验的申请人，绝对不会用低人一等的语气给教授写信。再次强
调一边，陶瓷是基于平等的，你自己低人一等了，别人不来理睬你更是理所当然的事情。
我还看见有的朋友在论坛上面不停的问，教授这样回复我了，我之后应该怎么继续套。真
5 _2 }( z8 }% C4 ^. B$ W是谬之大误啊！首先，随随便便在公共论坛上面公开私人信件就已经不合适了，何况你对
与教授的联系一点计划都没有，那你当初干吗还要陶他的瓷呢？这样的朋友摆明就是所有
的教授都掏一下，“宁可错杀一千，不可漏杀一个”；撞大运般的先撞一个再说。我敢说
，像这样的朋友能够最后套到offer的可能性是很小的----你可以这样陶瓷，别人当然也
3 n, I- }- l/ Y. z0 A: D5 H可以了！如果你没有什么特别的东西让教授亲睐你，那么对不起，为什么他就要找你做学
6 v# c  [# t; F生呢？另外，陶瓷之前，一定要看清楚对方的科研背景，以及“最近”正在进行的科研活
* `# ^4 Y: ]  f# D& E动，注意一定要看最近的。比较有用的办法是看publication里面'submitted'或'to  
* Z% L; P# e& S2 kappear'的论文，有机会的话下载下来仔细研读一下也是非常有用
的。很多朋友陶瓷信过去就是直接问是不是有钱，说到自己就是gpa多少，拿过多少奖，
1 g9 {( F, U# ?0 j5 g+ \  R好一点地会介绍一下自己的科研经历以及发表过的论文等等，但是人家真正感兴趣的是
你可以在他的领域里面有多大的成就。注意，是他的领域。你不好好研究一下对方的背
景就贸贸然陶瓷信过去，反而对你不利。因为教授看见这样一封没有丝毫针对性地陶瓷信，
8 f! m* l6 b7 T1 a. }会认为你是在用一个通用的模板信批量发信。如果遇上较真一点的教授，会直接把你的
+ A7 Q4 K" V; D4 q' n- e4 ?email发给招生主任（通常也是系里的教授），那么对不起，你就别再指望这所学校了。
; J% g4 ~* r! c  l# H" H
第三，不要在陶瓷信里面把话说司。很多朋友同时联系好几个学校，email答应了一个教授
/ q3 P8 y% {* X7 l2 c的offer之后来了新offer反而有烦恼了，因为在陶瓷信里面把话说得太死，以至于很难反
悔去新的学校。在这点上，我就要说，你这个时候做的决定，以及对教授的答复不仅仅是
4 i$ w) k8 R9 Z/ U* k0 m代表你自己，对很多教授来说，也代表了全体中国学生的整体形象。你应该坦诚以告，而
且不要拖了几个月再告诉别人，而应该一旦确定offer之后马上就应该告诉别的学校，这样
# }1 w, P( C. W$ Q/ i3 d不仅方便别人，也对自己在美国的将来有好处。要知道，美国这里很多科研圈子是很小的，
今天在这个会议于上了，明天在那个会议碰到了。如果这个时候一着不慎，就很有可能找来
臭名。何况，如果你处理得不谨慎，那么至少你这个学校后来的学弟学妹可就惨了，这个
3 l% K6 a& }5 S. G教授不会再信得过了。没办法，虽然美国强调机会平等，不能歧视。但是教授也是人，他
7 S! _( e! R5 e8 |/ o( B3 i3 w对一个学校出来的某一个申请人的判断很大程度上也要基于对这个学校出来的学生的整
9 J# ]: b# K4 O3 t体评价。那么有些朋友就会说，“那又怎么样？反正我也出来了，offer也拿到了，后面
& a4 H- R) k; D) Y) Y  f1 W人的事情管我什么事？”呵呵，首先，像你这样不顾他人，不顾后果的人到了美国是中国
2 G! }7 r. Q: \. r的幸运，美国的倒霉；其次，你这样做了一次，以后就会做第二次，栽跟斗是迟早的事情。

# g" v4 H' Z& [+ v1 ~$ V5 ^最后我转铁kansas state univ 电子工程系的一个印度assistant 教授对陶瓷的看法和建
议。我也是最近偶尔看见他的这个网页的。有兴趣看原文的可以google他的主页。这位刚
# [( ^; V7 y& k: U  W- `0 X4 h刚博士毕业的印度年轻人意见很尖锐，很多地方语气甚至都不太友好。但是，认认真真地
% R5 v$ j8 J; x; G看完绝对对你的申请道路有着非常非常大的帮助。


International applicants seeking assistantship/admission: please do not send  
% w  ^1 `2 Z* x5 j! Gme emails blindly without reading this  
   
7 A/ g' C/ ]+ |9 O, IEvery day I receive emails from international candidates for information,  
for advice, and especially requesting financial assistance. Unfortunately,  
( U8 Z0 X/ a0 D  Z" r/ i5 v# _I delete most of them without even reading them carefully. I get so many  
! v1 X/ l5 }- r8 Uirrelevant emails that cannot reply to each. Extremely rarely has one  
2 W" T- k' e5 p8 a0 {1 _caught my attention. Please do not send me emails without reading this  
first. I am only interested in corresponding to candidates who are  
exceptional, and have a strong interest and experience in my immediate area  
+ S4 c: l' K# r8 g, j: `of specialization, and I highly prefer ones with a Master's degree.  
   
1 T  J- t4 `+ K7 O# I3 ^3 lFor your benefit, I have included in this page, some writing tips, when  
. z6 h0 d. m$ m8 kcorresponding to me, and perhaps also in writing to other faculty members  
in US Universities. When corresponding, do not mention your age/date of  
$ J9 g6 {& M2 ^1 Hbirth, your gender, nationality etc. These are superfluous information.  
   
+ h3 E  ~- `8 X
+ ?, b& C; P0 f# _8 f. s
I am SOMEWHAT interested in: Your awards/academic honors, academic records,  

GPA, GRE score (2200+), experience.  
% x  k1 ~) H+ p  ?1 \   
I am HIGHLY interested in: Your publications, research interests (have  
supporting evidence for your claim, don't just claim you are interested,  
unless you actually have done some work in the area).  
' W* V# B) {0 d) \8 B) s6 {   
% n3 q3 d  ~5 w, |*(Incidentally, I do not know what NOVELL or Microsoft certification are.  
  S0 h8 h6 Q$ F: G9 RResearch is more concerned with things like advanced mathematical skills,  
creativity etc. Too much software skill actually may even count negative in  

my group).  
& u0 m. G; B; N/ D+ g   
   
. N' l9 W) S1 |2 g   
1 z6 ^1 Y/ z1 g" _% oInformation I CANNOT provide  
2 f. y8 Y9 l6 f5 j% p   
1. Chances of getting assistantship in the department (other than my own  
available research grants - I do not have any right now, although not very  
likely, it may change in future).  
7 h- u: Y5 v6 W6 i4 |   
2. Deadlines for assistantship/admission.  
   
3. Any information on admission procedures and policies.  
   
2 s5 H( R% |; w8 q! r1 J" F4. Minimum GRE/TOEFL scores required.  
9 S) j+ [9 Q* t- p% d. I   
5. How to get application forms.  
! ^) P  K8 {6 }4 w0 D& L   
6. Whether a decision for assistantship or admission was made.  
   
6 s$ }4 V2 d6 T3 }6 m4 L7. Other faculty who would be interested in a prospective student's  
background (you can easily see faculty research interests from the  
2 U' K8 A2 f. W8 y; H( G$ _department website).  
   
8. When an admission/assistantship decision will be made.  
   
9. Any other information, other than that below,  
  H, X% ?) L  R   
   
   
  _6 c) T( n4 Y, \; qThe only information I CAN provide  
   
+ u7 A: s7 @6 }; B, }- A) G& X1. Compatibility of your interests/background with my own research area.  
* I+ a2 w- Q' n  _5 V/ t" U) i9 t   
! g( o2 T# b7 R2 C2. Possibility of getting research support from my own (not departmental)  
funds. I am always trying to get external funding for my research projects.  
$ @+ Q! L* H- N' f1 y
   
3. Possible research areas you could work on with your background, with me.
# S  o7 Y6 g+ A% d6 X
" {2 R/ w' t; q
0 F* r( o8 Y+ V% q/ _; mI know how interested some students are in coming to the USA for a higher  
& s% f3 H  N5 [! p& c4 e) G: fdegree. I have written this as a useful guide for you. I have also written  
. q9 _/ `2 X! s+ F+ u, j0 Lthis to let prospective international students try and focus their own  
+ o4 t2 F$ F; v- d5 O6 ~% Iefforts properly, instead of wasting their energy sending me emails. Here  
2 J- f! g+ u8 Fare two sample letters.  
   
A sample letter that will never get the candidate an assistantship with me:  

  i. l/ g5 b2 Z. c! k8 a5 X* {   
( E" M$ S3 l. U$ ?7 Y" C6 hUnwanted email  
/ h5 U7 P5 s: s) k  a   
7 M+ Y; a) F" a& F: y9 ^" _$ QRespected sir,  
   
; H% f  P: m* C5 t" B# w: o9 DI am very interested in going to the US for higher studies. I am seeking  
admission and financial assistance in your esteemed University. I am very  
interested in artificial intelligence and computer architecture. I have  
$ ^' |8 E" m0 a5 ]6 Wprogramming experience in C++, C and Java. I am also familiar with several  
software packages in the Windows environment (MS Word, Excel Spreadsheet  
! D3 f5 _! _6 ]etc.). I also have networking certification from NOVELL. My GRE score is  
- G+ R: i$ J( ?2050, and I got 99 percentile in the TOEFL exam.  
6 O: v1 [( x) g, W2 s. S4 O   
I request you to spare a few moments from your precious time to go through  
) I! w( d& }. O5 }; Smy attached resume.  
   
Sir - if you are kind enough to offer me assistantship, I promise to work  
+ W! ?3 I- s$ v( S8 B9 z- f6 Cvery hard for you. I have a very good academic record since primary school.  
+ [; z$ W8 H7 d+ @' s
) ]  }8 W) H7 D! XAlternately, can you tell me what are my chances of getting assistantship  
4 [' x- Q* b3 J- D* `from your department?  
- |2 ~2 G& h/ ~- Z   
" J: L3 r8 s$ x/ \. XCan you kindly inform me about the last dates for Summer term 2002 before  
which I can apply?  
1 ^; o0 y. }( z- \: m6 ^: ]   
Thank you for your valuable time,  
. `. P9 I" k. x$ ~3 x. o9 x- J   
Yours obediently  
4 Q$ A# e8 i/ H9 B3 Y1 s9 {   
7 ]+ o% J+ K0 R7 @ABC  
   
2 y: Z  s6 a. K6 G; [1 DMy comments: I simply delete these emails, which I consider garbage. The  
, c" ~/ [0 {3 u5 Sperson seems to be too obsequious and flattering. In the emails that I do  
( H& H' `1 l. ~+ L) q; H1 ^read (occasionally), I do a first scan for keywords like "precious",  
"respected" etc. deleting the email if I find one. Obviously the person is  
not professional enough. The person is interested in artificial  
8 M" y0 p8 v. ^4 y: u* Gintelligence and computer architecture - these are quite unrelated areas,  
making me think that he is trying to cover as many professors as possible  
# _/ c$ F5 I% H) U+ y9 Uto get aid from any possible source. Besides, he does not have anything in  
# o8 t, B- d9 ~/ U) lsupport of his claim (about his interest in artificial intelligence). He  
emphasizes on his software skills, making me think that his goal is to get  
a programming job in the US and higher studies is only to allow him access  
1 K4 E/ ]4 E4 ?/ v1 {4 bto the US job market. His GRE scores are OK, but not exceptional. He asks  
me to tell him what his chances of getting an assistantship are, elsewhere  
in the department. I have absolutely no way of finding that out, and even  
if I knew, I would not bother to reply back. He also asks me to give him  
. G5 x! ~3 u& _6 P. {. Qinformation about summer deadlines. How am I supposed to know that? (I  
" X! c" B7 y& {1 H, \1 L$ L7 Mwould typically expect that this information should be available from the  
& Q- d- i% W& ?: n$ ?University web site, or from the grad school, but my guess is as good as  
. c+ U7 E8 T( {, {- w9 }4 oyours.) Finally, this letter is too generic. He probably uses exactly the  
% s2 l' A' M, K' vsame format for hundreds of email letters to various professors throughout  
the country, leaving no stone uncovered. I receive hundreds of email like  
2 u7 W2 ]* u$ N$ g/ Ithis. Regretfully, I cannot answer everyone, and these applications do not  
  h9 o2 ^1 c( ?interest me at all.  
2 T) b+ W2 p8 j$ O- w+ O: ?, D   
   
   
: Q$ e4 k+ ^& BI have even received emails from students asking me to evaluate their  
chances of getting admission, based on their GRE scores, TOEFL scores and  
GPA. Here is an example:  
/ f7 a9 h' t: m% Z, I   
Highly undesirable email  
   
- Q2 M* Z4 Q$ s! @- CRespected Sir,  
8 T: W5 J7 J0 P" v   
7 p; d/ A: B* c5 b& vI am applying to a few US Universities for higher studies, including your  
esteemed University.  
+ o" ~5 N2 _' a7 h  A1 Z   
I am a fourth year electronics engineering student at XYZ University,  
India. My undergraduate GPA is 3.2/4.0. I obtained 2000 in my GRE  
5 o1 \8 P1 h& E/ cexamination (V: 500, Q: 800, A: 700), and 456 in the TOEFL.  
- C. W) k! c, W- J   
( [6 {: p+ \+ x% c; M. HCan you spare some valuable time to assist me in determining what my  
! v. e0 i$ x6 h6 `- achances are for assistantship in your department and in other Universities?  

: ^9 K6 m4 g2 QI will appreciate any help from you.  
7 M, v' `; V, X, U# Q& g) @  P8 i   
/ T! S7 }1 X% oSincerely,  
) L8 [# m' T2 m2 q4 {, f7 s& _   
" U. j' {( L- W" N7 a) vPQR  
' b) y" g7 w6 e) ]   
My comments: This letter is at least as bad as the previous one. It too  
! E# o6 S; A7 K/ wlooks excessively obsequious and unprofessional with words like 'respected'  

) d) @) b: a4 i+ ?and 'esteemed'. I will not respond to queries like this, even if I could  
possibly "detemine their chances". As I mentioned earlier, my mailbox gets  
" h% v1 `+ J5 dflooded with these queries such as this each week, and I have no time to  
$ I/ B: S; z6 o' v4 E( ]respond. But more significantly, I do not have the slightest way of knowing  

4 F, H8 c8 o/ k( C& y0 A# u! Dthis, not even in KSU EECE department, leave aside other Universities. You  
are much better positioned to do that, based on your own peers'  
experiences. Take an example, how much information can you, as a student,  
; J- g& b) l7 a2 B+ d4 W9 Ogive me about the criterion for promotion from associate professor to full  
  r  _' M5 a. A- f, hprofessor in your University? Likewise, I have no idea what policies  
# y: G, y! b; Ograduate schools in the US follow to admit/denying admission to prospective  

students. I barely know the format of the GRE exam, leave aside TOEFL. I do  

not have the slightest idea how to distinguish between a 'good' TOEFL score  

4 z1 g! K2 h: \2 F" o* c+ [and a 'bad' one. Besides, I know that most Indian Universities do not  
0 w" ^& I3 _3 C# M3 _+ ofollow a GPA system. His "GPA" is most likely inaccurate, obtained by  
1 q8 R( s. k+ a+ x8 Z# b  bapplying some arbitrary conversion formula devised largely to scale up the  
overall percentage score. Use GPA to fill in the official admission form,  
: R" }% T: V& D: Eif you please, since there is no space for marks, but do not give that  
information to me unless your University actually follows a 4 point GPA  
( F+ v! w$ B$ S( J* p  q1 h" i& \system. (Unfortunately, I cannot distinguish a good percentage score from  
, W& p5 h. }# D: q9 Z/ [an average one either.)  
   
& g3 c# x$ C$ y( R   
   
. h7 [0 H$ \  @$ `' ZI have even received letters with personal items unrelated to the  
profession - something which I consider to be highly negative (examples -  
"I looked at your family pictures in your homepage", or "I am from the same  
* r4 e% Q8 C/ u" t" ^
+ R5 A5 L- d, y9 s- {& I6 Vpart of the country as you are" etc.).  
   
  g- x0 W- C5 ]+ i; x& A' T   
   
; F5 \( P$ s% i2 U% Y# m  e+ I* ]Now, here is a letter that would DEFINITELY attract my attention:  
; T* Y5 h8 e# ]- @! k. Q6 ^   
Excellent email  
. f" y! u4 q. H4 V1 o6 T   
Dear Prof. Das,  
   
I am applying to KSU (amongst other places) for a Ph.D., and I am very  
interested in your research areas. I would like to explore the possibility  
of doing my research with you, with financial support.  
6 u0 z8 n; S3 j6 q4 G1 r; ?# P   
One possible area that would interest me a lot is in applying Tikhonov  
regularization to your problem in computational neuroscience, since I  
+ ~5 V" |/ l  U/ ?( }understand that neural data can be very noisy, and yet sparse. Alternately,  
9 b* \! b) g9 n4 g, I5 S
3 p* Y/ D4 F' D6 H' I( b9 d- ^I am also interested in applying evolutionary algorithms to your genetic  
$ F$ N+ M6 w6 Y: X2 H! o, lneural network parameter estimation. One of my other interests in in  
- U  ]' k6 [2 [4 e; z; `8 sparallel processing. Are you considering parallel implementations for any  
of your optimization/estimation algorithms? Considering that the projects  
that you list in your web site would probably be very computationally  
intensive, that may be a worthwhile strategy to pursue.  
; _! P$ J0 n# a6 _8 }% c! e   
  H5 Y/ q- I/ ~- T1 T- rI presented a paper entitled "Hybrid Gradient Descent Based Training of  
4 c* X6 f2 K' R; ~& x# Q6 `Probabilistic Recurrent Networks" in the International Conference on Neural  
: y1 O" ]  F/ O. f* w+ k  I
Networks and Image Processing, Bangalore, last year. It is a reviewed  
9 G& |$ y# I, ]( npublication. If have attached a copy of the article in pdf format.  
   
$ x6 m9 E+ o$ V; x; I0 c3 fIf you have available funds, and are interested in my application please  
; [; i4 a% H: Ilet me know soon.  
   
XYZ  
, ^/ m8 g! g9 {/ s! d' [   
0 |1 H/ [  C  k0 Y; H2 fMy comments: This is a straightforward letter from someone with a genuine  
interest in my area. He is an exceptional candidate. He has evinced a keen  
desire to learn, and already has considerable mastery over my area of  
8 n* ^7 l/ ~( m' W( b9 ~research. If I have any funding available, I would definitely try my very  
BEST to get this person. This person may have an MS degree, while many  
applicants have only a BS degree.

0 V: F5 z& c4 T! f6 ~7 K/ v


I know how interested some students are in coming to the USA for a higher  
! @% j7 g' |5 T# N  Ddegree. I have written this as a useful guide for you. I have also written  
this to let prospective international students try and focus their own  
efforts properly, instead of wasting their energy sending me emails. Here  
are two sample letters.  
   
A sample letter that will never get the candidate an assistantship with me:  
9 ~6 N7 J1 |/ N% i- K9 a' F
   
Unwanted email  
1 _6 O2 C7 O2 V/ K* A* [$ w5 j. h+ S   
Respected sir,  
   
- z9 j3 G4 I0 c& p- _/ SI am very interested in going to the US for higher studies. I am seeking  
admission and financial assistance in your esteemed University. I am very  
' Q0 f. m% x/ C! D0 y0 o% binterested in artificial intelligence and computer architecture. I have  
' U/ P7 ?2 ]0 ]6 r" nprogramming experience in C++, C and Java. I am also familiar with several  
& W2 C; [* B' k2 Q. q; ]software packages in the Windows environment (MS Word, Excel Spreadsheet  
8 [8 j) V, }2 ?1 c; ^& U  Ketc.). I also have networking certification from NOVELL. My GRE score is  
2050, and I got 99 percentile in the TOEFL exam.  
   
I request you to spare a few moments from your precious time to go through  
! `1 y6 d1 E8 xmy attached resume.  
" @, U3 J  a. ]( m' w6 R, Q   
+ W; o- R% g9 a* e. D8 `Sir - if you are kind enough to offer me assistantship, I promise to work  
5 \7 d5 m" D! _! B, Z8 u' }# xvery hard for you. I have a very good academic record since primary school.  

Alternately, can you tell me what are my chances of getting assistantship  
from your department?  
* o' B0 P$ V! `1 j   
. }$ A/ r5 L- r$ H  q; K3 W+ ACan you kindly inform me about the last dates for Summer term 2002 before  
which I can apply?  
   
) k+ V$ t) T, Q# G; z- g, I% q% [Thank you for your valuable time,  
4 }, a2 ?, m0 E' x/ r3 d   
Yours obediently  
) l8 I" y, w, {$ f" `$ q. W   
; v3 ^: J- e( N- gABC  
   
My comments: I simply delete these emails, which I consider garbage. The  
person seems to be too obsequious and flattering. In the emails that I do  
/ `! D1 Q; Y5 t2 rread (occasionally), I do a first scan for keywords like "precious",  
& _$ S: s) x; {" N4 M2 l# {9 s"respected" etc. deleting the email if I find one. Obviously the person is  
+ i. X& c$ {4 Jnot professional enough. The person is interested in artificial  
/ s# X( F: W* pintelligence and computer architecture - these are quite unrelated areas,  
0 c# k1 `& ~1 g1 I: @making me think that he is trying to cover as many professors as possible  
. h2 z7 ?, I# |- {! nto get aid from any possible source. Besides, he does not have anything in  
support of his claim (about his interest in artificial intelligence). He  
emphasizes on his software skills, making me think that his goal is to get  
a programming job in the US and higher studies is only to allow him access  
to the US job market. His GRE scores are OK, but not exceptional. He asks  
me to tell him what his chances of getting an assistantship are, elsewhere  
1 W- [0 U( Q5 v) G: }) ~- ?in the department. I have absolutely no way of finding that out, and even  
if I knew, I would not bother to reply back. He also asks me to give him  
. ~; I4 f6 {- p! @) [1 einformation about summer deadlines. How am I supposed to know that? (I  
would typically expect that this information should be available from the  
# j! Q+ a. [  X2 l! vUniversity web site, or from the grad school, but my guess is as good as  
yours.) Finally, this letter is too generic. He probably uses exactly the  
same format for hundreds of email letters to various professors throughout  
0 T) V( Y& m, c+ A) I9 O& Zthe country, leaving no stone uncovered. I receive hundreds of email like  
this. Regretfully, I cannot answer everyone, and these applications do not  
- c, P% ^7 e/ P; cinterest me at all.  
# Q. D, o" R% H; j, \( E   
   
# j( H& I9 Y; V6 l   
9 B6 b8 W9 x4 _8 D- i+ G( II have even received emails from students asking me to evaluate their  
! |; k" |; u7 |( @6 _8 zchances of getting admission, based on their GRE scores, TOEFL scores and  
, B0 n! v$ B/ a3 cGPA. Here is an example:  
   
Highly undesirable email  
7 Y; R* q& c) q8 f   
9 B' _- ^: W: r  q# f/ FRespected Sir,  
   
I am applying to a few US Universities for higher studies, including your  
esteemed University.  
. S- E+ m' d8 i' J# A   
" K( r9 Y9 G6 ]6 _( Q; ?  eI am a fourth year electronics engineering student at XYZ University,  
  o8 z  T, \& H6 u# n' j9 w+ n4 E3 Z- \( wIndia. My undergraduate GPA is 3.2/4.0. I obtained 2000 in my GRE  
examination (V: 500, Q: 800, A: 700), and 456 in the TOEFL.  
   
Can you spare some valuable time to assist me in determining what my  
8 q$ ~8 D5 o* I- d8 nchances are for assistantship in your department and in other Universities?  

# L4 [6 [8 U; T: ]/ m" j+ S5 S# sI will appreciate any help from you.  
+ E2 n2 f7 M# N) H' B. M+ {) w   
8 [6 T! {2 K+ a1 k3 qSincerely,  
   
- K0 G, j0 i# j0 Z  F0 t% R& ~PQR  
5 p7 u9 u: f) p) b   
+ B5 ?7 }% d- L& `$ GMy comments: This letter is at least as bad as the previous one. It too  
looks excessively obsequious and unprofessional with words like 'respected'  

( m! M: _1 v( j# H/ Fand 'esteemed'. I will not respond to queries like this, even if I could  
7 f* k  i9 t' E8 xpossibly "detemine their chances". As I mentioned earlier, my mailbox gets  
flooded with these queries such as this each week, and I have no time to  
respond. But more significantly, I do not have the slightest way of knowing  

this, not even in KSU EECE department, leave aside other Universities. You  
are much better positioned to do that, based on your own peers'  
experiences. Take an example, how much information can you, as a student,  
" @3 d* h" X  t0 [give me about the criterion for promotion from associate professor to full  
professor in your University? Likewise, I have no idea what policies  
graduate schools in the US follow to admit/denying admission to prospective  
% S; z9 Z) J' F1 V
students. I barely know the format of the GRE exam, leave aside TOEFL. I do  

% o* T8 L( y* G" J9 j  G1 h5 @not have the slightest idea how to distinguish between a 'good' TOEFL score  

! d3 d- M0 q. W# P5 W4 pand a 'bad' one. Besides, I know that most Indian Universities do not  
follow a GPA system. His "GPA" is most likely inaccurate, obtained by  
  s; F8 D. l( ]2 I2 Z6 Q% B; Kapplying some arbitrary conversion formula devised largely to scale up the  
overall percentage score. Use GPA to fill in the official admission form,  
: E9 @+ \+ J) w' iif you please, since there is no space for marks, but do not give that  
information to me unless your University actually follows a 4 point GPA  
system. (Unfortunately, I cannot distinguish a good percentage score from  
an average one either.)  
& R7 y9 [4 |" `+ @   
+ Q1 O, b) A# L4 \   
   
$ f# x1 `+ i; t0 WI have even received letters with personal items unrelated to the  
profession - something which I consider to be highly negative (examples -  
7 g0 J% q% e* B2 A"I looked at your family pictures in your homepage", or "I am from the same  
4 i5 C9 Q% c! ?5 ^( n9 K
part of the country as you are" etc.).  
   
% E" }4 f. Z0 Z   
2 q0 }6 O& ?2 y) S8 t   
: }% W& T/ |% qNow, here is a letter that would DEFINITELY attract my attention:  
   
# x1 X6 t' C; ]Excellent email  
   
( W2 N* f3 ~& E& z. M" _Dear Prof. Das,  
5 U2 ]1 ^9 Z$ e" T/ e   
, Q0 q6 h, o  b; FI am applying to KSU (amongst other places) for a Ph.D., and I am very  
8 v7 W+ X5 u- P, j" _  o6 Binterested in your research areas. I would like to explore the possibility  
of doing my research with you, with financial support.  
   
4 G9 K8 p" Y+ t! T$ mOne possible area that would interest me a lot is in applying Tikhonov  
regularization to your problem in computational neuroscience, since I  
understand that neural data can be very noisy, and yet sparse. Alternately,  

  r/ V" i: c+ c% f0 A* OI am also interested in applying evolutionary algorithms to your genetic  
9 }! E5 ?/ E7 w1 n6 R  R8 }neural network parameter estimation. One of my other interests in in  
parallel processing. Are you considering parallel implementations for any  
of your optimization/estimation algorithms? Considering that the projects  
& B/ q. Z" f% \that you list in your web site would probably be very computationally  
# E4 ^' t* L; `7 @intensive, that may be a worthwhile strategy to pursue.  
* U+ R5 v3 t' {   
+ Y" w" n1 v6 d  Q0 ?I presented a paper entitled "Hybrid Gradient Descent Based Training of  
Probabilistic Recurrent Networks" in the International Conference on Neural  
. p* ~+ T. ~" `; [
Networks and Image Processing, Bangalore, last year. It is a reviewed  
publication. If have attached a copy of the article in pdf format.  
   
If you have available funds, and are interested in my application please  
3 |* |7 G3 G5 `! z, E! p! rlet me know soon.  
   
8 J& v. ^# j6 \3 {2 GXYZ  
7 P/ Y  R' B$ a; G* p* A6 }3 {   
' O2 {+ F; Y% V4 k% F3 fMy comments: This is a straightforward letter from someone with a genuine  
interest in my area. He is an exceptional candidate. He has evinced a keen  
. s( U' c, n+ F0 Q0 l' @desire to learn, and already has considerable mastery over my area of  
# h1 c5 H* ^! {! h. Q, dresearch. If I have any funding available, I would definitely try my very  
BEST to get this person. This person may have an MS degree, while many  
; _. k3 v( x! V2 s6 ~1 lapplicants have only a BS degree.
4 O: M& n# V& u1 p0 z& L
; x0 S' K+ `3 J% |- v: [[ 本帖最后由 北京的争争 于 2006-6-5 03:03 AM 编辑 ]

北京的争争

Here is a letter from a fresh undergraduate that looks very good:  
   
  w4 Y  b, r. y% U& u* f' [Good email  
   
  \4 e& {6 L' y9 ?9 V" N% J* }$ KDear Dr. Das,  
   
I recently applied the Ph.D. program in your department at KSU.  
# G2 x" P5 W( s  K8 {4 r   
5 H' k$ K4 ^4 E8 q* h+ \7 {I  looked at your home page, and your research areas were very interesting  
* N! s' j0 ~  }- v' M5 h4 T0 R6 bto me. I also read your research article that appeared recently in the  
journal "Signal Processing & Data Mining (note: this is a hypothetical  
! v% A  I) Y  l3 fjournal)". I am very interested in your idea of applying independent  
component analysis to data mining. My own undergraduate project was in a  
6 f, [$ q* q1 j) c1 r- w( n& jclosely related area and should be of interest to you. It was entitled  
7 }8 [0 G' d. [8 B, S# {7 ~"Applications of LVQ algorithms to speech recognition".  Furthermore, I  
recently joined XYZ labs in Mumbai, (India) as an assistant scientist, and  
I my work is in designing neural network algorithms for knowledge discovery  
3 H4 o) A; P( F# c. X& D
applications.  
) B/ i5 k( s6 C   
My GRE score was 2270, and I also took the GRE engineering subject test  
1 _8 |. e; J8 n" Z0 c& Mwith a score of 96 percentile.  I stood first in the the University in my  
" g) g4 ~4 |7 [$ ubatch (245 students).  
   
# ~* f( f: \% X' WPlease consider me for any financial assistantship that you may have. I  
  A% s  O2 C7 J6 Phave attached a 5 page postscript format curriculum vitae.  
   
Sincerely,  
7 f) q% d# ?: z% ]   
ABC  
   
My comments: I would try to get some funding for this candidate, although I  
, A+ g7 b! B7 z3 H% o
9 A* z. B# c3 Y0 B! i3 o9 F( W/ C3 Cstrongly prefer students with an M.S. degree.  
   
I recommend that fresh BS students who wish to apply, do the following:  
7 t  O1 ?' w6 a0 l0 L. z0 tDecide on what area you are really interested in. Be very specific about  
what you want. Then first spend a few months researching material in that  
area. Look at conference and journal papers. If you do not have access to a  

1 [( g7 a& |4 p' w# s3 Bdecent library, then surf the web, and read as much on the subject as  
3 B7 B! u+ u$ R6 _- \$ e" b: O% kpossible. Even try and present a paper at a national conference, perhaps  
7 O& h! a" t4 j: A9 `$ Z" Ywith a University professor there. When you are finally ready, start  
getting information about faculty interested in that area and write to them  

only. This is the only way you ever stand a chance for a research  
assistantship. I think it is very unlikely that a professor in the USA  
1 L9 K* s! ?- B9 _: awould take you blindly based on your GRE scores, your undergraduate record  
and a few letters of recommendation.  
! d6 \, d: M9 _1 p  v' @   
If you are not willing to do this, then focus on a teaching assistantship  
9 e4 T) l+ a, b# A/ {: m/ Wonly. For teaching assistantships, you should not even be sending emails to  
% I7 s, y- ]# N0 [0 j* F' ?$ ?8 j
  N7 R5 a" {' a7 N" S2 S" k* {/ n, Ovarious professors, but only to the department chair/graduate  
; m+ w, a* N/ H# W& v- N; M0 t; _coordinator/graduate advisor. Individual faculty members have no control  
4 Y( `, S, B$ ~7 O, J$ G+ hwhatsoever in deciding how teaching assistantships are going to be awarded.  
  |; L2 v, [! ?6 {4 {+ x4 x' f, V& |8 K
For these positions, your undergraduate performance, GRE scores, etc. count  
# K+ x: N0 @: ]% v1 u) v2 @* c
% {. X% r0 r- C1 E& G& Q: p2 Ua lot.  
7 @2 d8 R* L, o+ m   
   
   
7 \, @! g! r; u4 h  yHere is a letter from someone from a country where English is not spoken:  
+ J7 v( n& e7 g* ]1 j5 @4 H) u   
Another excellent email  
4 d& C+ F: F7 \2 p. O" g1 a* o   
7 T3 n" k4 m: @( W( }% O- {Dear Dr. Das,  
( s) Y( a. i2 ?8 v7 c, N1 ]   
I studied at a famous University in Beijing, PRC in Electronics Engineering  
: g7 n! }' g5 \' Q! R/ H
in 2001. I intention is to travel to US for a Ph.D. I find your area very  
& `" U! k- \% R( vinteresting such as neural networks and genetic algorithm. I published a  
paper in fuzzy logic after research in my university. This paper is in  
Chinese. But I submitted another English paper for a conference last month.  

; ^& o# r6 R. Q& U8 W& a7 XIt is application of neurofuzzy system to 3 joint robot arm control. My GRE  

$ Z) G" @2 C$ ~" j8 a% W! S- Kscore is 2100, (Quantitative 800, Analytical 800, Verbal 500).  
; r* K" s. [' d/ Q   
I am very interested in Kansas State University. Do you have assistantship  
available in next semester to teach a Ph.D. student in neural network  
1 c2 ~) M/ F4 e" |application to pattern recognition or sensor fusion? Please see my resume  
$ H6 @# |$ i4 M0 W4 ]8 tif I am suitable for you.  
   
) K  _( ?, b  ?/ N! Q7 f& A. P3 g% C! }Thank you,  
   
0 ], f  E% G; S4 v* d5 PJKL  
   
This letter looks exceptional.  I receive many letters from candidates from  
) b1 r" D- {$ L4 k4 |
Asian countries where English is not emphasized. That is no consideration  
from my point of view, as long as you qualify for grad school. A lack of  
+ G7 J2 v& F7 Y" PEnglish should not discourage someone from applying. Students from several  
  K( H2 i( ]7 R; n7 r  Z1 Hcountries with bare English capabilities have known to be highly successful  
- V8 w& C; k2 v9 X" y) A3 r. M9 F
in the US because they are so hard working and sincere that they more than  
6 E3 o0 f7 D& j1 s( w8 Dcompensate for this minor disadvantage. I will only care about things such  
as your mathematical skills, and research record.  This person's GRE  
analytical and quantitative scores are excellent. But most important is the  
$ g& s& Q. p  G8 `2 K! y0 o
fact that he has two publications, even if one is in Chinese. I will give  
credit to this student for writing a paper in English, even if he is not  
5 a; o) w  V% lfully comfortable in that language. It shows how dedicated the person is. I  

certainly would like to have this person in my research team.  
9 P" Y0 d( D' S! A8 S   
! F6 r5 |  a/ p# r  y   
   
I have heard that there are agencies which send out emails for you to  
various professors in the USA. In fact I and my colleagues have received  
exactly the same generic email several times, with only the names changed.  
7 n- Y  D7 `# I/ H! j2 qSuch an email can only have a NEGATIVE impact on your chances. That is a  
total waste of money and a very bad idea.  
   
Finally, don't be discouraged if you cannot make it to the USA. With  
economies in India and China skyrocketing, I expect in about a decade, the  
& d7 _1 m- _( X5 S! w' ?job prospects in these countries could even be better than in the USA. Even  
! `. R$ _+ F' f
8 i% }/ |1 a: z& B+ f: vtoday, I am pleasantly surprised to see engineers, software personnel,  
doctors, scientists and managers in India enjoying more or less the same  

6 `) b9 u+ b) o, f8 Wlifestyle that one does in the USA (and probably better off than their  
- _" B1 Z7 x$ Scounterparts in some other countries like Britain). I am sure the standard  
! m: m2 @. R9 t0 xof living in some other Asian countries, such as China is excellent too.  
   
   
" V# D# p/ \8 R0 H4 L/ [0 d1 G   
2 H# \! q/ u2 ~2 _7 W! d9 POne last word. If, and only if you get my attention (which is extremely  
. A, ?# }# K9 h* erare), I will keep in touch with you, push hard to increase your interest,  
% w; |. W4 ^( d. U: O1 b0 ltry my best and do EVERYTHING possible to get you. Currently I do not have  
much funding available (or have not updated this page), but I hope that  
0 ?: m4 R& l4 W% y- @4 Nwill change soon.

若海

那你能告诉我怎么套你吗？
) t) B& j$ _7 j) U0 r可不要给我讲外文啊，我是一句也不会的。