diff --git a/demystify.pdf b/demystify.pdf index fba7f54..ef747af 100644 Binary files a/demystify.pdf and b/demystify.pdf differ diff --git a/demystify.tex b/demystify.tex index e261ccf..c982af8 100644 --- a/demystify.tex +++ b/demystify.tex @@ -557,7 +557,9 @@ \subsection{Why CS depts do not waive the application fee?} In gneral, application fee is typically a requirement of the university. Individual departments and programs do not have the flexibility to waive the application fee, even if they want to. \begin{commentbox}{Vu:} -In my opinion, requiring applicants to pay the fee helps ensure their seriousness, as it filters out non-serious candidates. Most CS programs already receive many applications and would be overwhelmed if the application process were free. Note that if you have financial difficulties, you can ask the department for a waiver, but this is typically only granted in exceptional cases. +In my opinion, requiring applicants to pay the fee helps ensure their seriousness, as it filters out non-serious candidates. Most CS programs already receive many applications and would be overwhelmed if the application process were free. Even with application fee the competition is already very tough, imagine if the application is free and the number of applications doubles or triples.\\ + +Note that if you have financial difficulties, you can ask the department for a waiver, but this is typically only granted in exceptional cases. \end{commentbox} @@ -633,7 +635,9 @@ \subsection{Generic Letters are Bad}\label{sec:generic-letters} \subsection{Self-written Letters are Bad}\label{sec:self-letters} Many international students write letters themselves, typically due to the requests of their profs. or letter writers. Such letters have \emph{little value} and are considered weak by reviewers---why can you not even find someone who cares or knows enough about you to write a candid personal reference letter? Instead of the ref. writer talking about you, in this case it is you who write about yourself (and they just sign the letter). Experienced reviewers can easily spot (because the student often has little experience in writing LoRs) and \red{red flag} such letters. -Most well-known and well-respected profs would \emph{not} ask you to write your own letter (in fact, even not well-known ones wouldn't do this to students they care about). This might be a common practice at specific universities and the students do not have a choice as they need the letter. However, think about this: if a prof. does this often, then they either don't know how to write a LOR (more common than you would think) or simply do not know or care enough about you. In any case, such LoRs are not useful and might even hurt your application. So if you are in this situation, you should find someone else to write for you. +\begin{commentbox}[Vu:] + Well-known and well-respected profs would \emph{not} ask you to write your own letter (in fact, even not well-known ones wouldn't do this to students they care about). This might be a common practice at specific universities and the students do not have a choice as they need the letter. However, think about this: if a prof. does this often, then they either don't know how to write a LOR (more common than you would think) or simply do not know or care enough about you. In any case, such LoRs are not useful and might even hurt your application. So if you are in this situation, you should find someone else to write for you. +\end{commentbox} \subsection{LoRs from Admin People}\label{sec:admin-letters} @@ -649,7 +653,6 @@ \subsection{LoRs from Admin People}\label{sec:admin-letters} - \subsection{Waiving Your Right} Choosing not to look at a reference letter is pretty standard in school and job applications. When you waive your right to see the letter, it adds a layer of trust, showing you're confident in your choice of referees and that you're not trying to twist their words. It's also about keeping things open and honest between you and your letter writers and encourages them to be real about your strengths and qualifications. Plus, it keeps things private. If you do not waive your right, the letter writer might refuse to write for you or write a generic letter that does not help your case. Reviewers also might raise concerns about a letter that is not waived, e.g., if you do not trust your letter writers, then you should find someone else to write for you. In short, it's a standard practice and a way of keeping things straightforward and respectful in the whole recommendation game. @@ -661,7 +664,7 @@ \subsection{Waiving Your Right} Choosing not to look at a reference letter is p \end{commentbox} -\subsection{Help your LOR Writer}]\label{sec:help-your-LOR-writers} +\subsection{Help your LOR Writer}\label{sec:help-your-LOR-writers} As mentioned above in \autoref{sec:self-letters} and \autoref{sec:generic-letters}, do not write your own letter and generic letters do not give much value. Thus, to help your writer to write a strong, customized LoR, you should provide them details or unique things about yourself. For example, let them know about your GPA, research and work experience, papers (if any), or anything you want them to mention. If the GPA in your program is highly competitive (\autoref{sec:gpa}) and they know that, remind them to talk about it in the LOR. You can also provide them with a draft of your SOP so that they can see what you are saying about yourself and complement that with their own perspective. @@ -669,17 +672,38 @@ \subsection{Help your LOR Writer}]\label{sec:help-your-LOR-writers} Sometimes your writer will explicitly ask you for such information, but if not, you should provide it anyway (especially if you have not interacted with them much or have not done much research with them). -\subsection{Remind Your Writers} +\subsection{Remind Your Writers}\label{sec:remind-writers} After you submit your application, you should tell your writers about that and let them know they will soon receive an email from the university to submit their letters. You should also remind them when you submit your application and ask them to submit their letters on time if they haven't done so. Note that most places only have deadlines for the applicant, but are very flexible with the letter writers (in many cases do not even give them any deadline). Also, many places do not begin the admission review process right after the deadline and work on application reviews in the next semester (mid-January). Thus you do want to send reminders because professors can be quite busy and might forget to submit their letters, especially when there is no explicit deadline. However, do not send too many reminders as that can be annoying to the writers. +\subsection{Thank Your Writers}\label{sec:thank-writers} + +After your application is submitted and your writers have submitted their letters (i.e., the wait begins), you should send a quick thank you note, which serves both as an acknowledgement that you know they have submitted the letters and as an appreciate for their help. You should also let them know the outcome of your application, regardless of whether you are admitted or not. In addition to being a common courtesy, this can also help maintain a good relationship with your writers, which can be useful in the future (e.g., if you need another letter for another round of applications or for a job reference). + + % \begin{commentbox}[Vu:] % For Vietnamese students, it's worth mentioning the \href{https://vef2.org/}{VEF2.0 program}, which has helped many students gain admission to top CS PhD programs in the US. VEF2.0 invites US faculty members from leading institutions to conduct rigorous interviews with VEF students and subsequently provide reference letters on their behalf. Despite the limited interaction between the interviewer and the interviewee (primarily just the interview), these reference letters are generally effective and have helped many students get into good PhD programs. % \end{commentbox} + +\subsection{How do you write LoR? Is it lots of work?}\label{sec:lor-writing} + +If a student asks me to write letter for them, I will agree as I believe it is part of my job. However, I will let them know if I won't be able to write a strong letter for them (e.g., they just take my class and I don't know them well enough or they have not done much research with me), and suggest they find someone else. If they insist, then I will write the letter, but it will likely be a weak one. However, I will still try to write something positive about the student, e.g., they are hardworking, or that I heard they are working on a research project, etc. I usually ask the student to provide me with some details about themselves, e.g., their GPA and transcript, and use that to write the letter. Regardless, the letters will still be short and weak (\autoref{sec:generic-letters}) and usually take about 5-10 minutes to write. + +For students who I know well, i.e., strong letters, then it will take a lot longer as it will be personalized. While I have a general template of such strong letters, it still can take me an hour or more to write such a letter. I also often ask the students to provide information (\autoref{sec:help-your-LOR-writers}) and what they think I should highlight in the letter. If they already have a draft of their SOP, they can also provide that to me so I can complement what they say with my own perspective. I also offer to review their SOP and provide feedback. + +%I do ask the students to waive their right to see the letter, as it is a standard practice and a way of keeping things straightforward and respectful in the whole recommendation game. I also let them know that I will not show them the letter, even if they do not waive their right. I also let them know that I will not write a letter for them if they do not waive their right, as I believe that such letters are not useful and might even hurt their application. + + + + +%\subsection{Will it annoy them if they have to submit many letters? }\label{sec:many-letters} +%Profs. are used to write letters for their students + + \section{Statement of Purpose (SOP)}\label{sec:sop} \sectioninfo{\emph{SOP is important}. Write it in such a way that makes you \emph{stand out}.} @@ -806,7 +830,8 @@ \subsection{School} Graduating from top universities \emph{that we recognize} he However, if we do not know much about schools in your country, then we are uncertain about the quality of your school and likely treat your school as \emph{``unknown foreign''}, which can be a minus point. -If your school is the top in your country but you think it might not be well-known internationally, you should talk about it in your SOP (and also ask your LOR writer to do that). Of course, if you're interested in working with Vietnamese, consider \href{https://github.com/dynaroars/dynaroars.github.io/wiki/Viet-CS-Profs-US}{CS programs in the US that have Vietnamese professors}. %It might also be helpful to have your CS dept to put itself on CSRankings so that others know about the dept, its faculty, and research. +Many international students mistakenly believe that their school is well-known, but in fact, it is not (e.g., many Vietnamese students believe that ``Bach Khoa'' is well-known internationally, but it is not). So you should explicitly mention that in your SOP and ask your LOR writers to do that as well. +Of course, if you're interested in working with Vietnamese, consider \href{https://github.com/dynaroars/dynaroars.github.io/wiki/Viet-CS-Profs-US}{CS programs in the US that have Vietnamese professors}. %It might also be helpful to have your CS dept to put itself on CSRankings so that others know about the dept, its faculty, and research. \begin{commentbox}[Vu:] Sometimes PhD adcom in the US will share a document such as \href{https://github.com/dynaroars/dynaroars.github.io/wiki/Foreign-Top-Schools}{this one}, which lists the top schools in several countries. We also ask other faculty and students if we think they know about the place. For example, when I was a postdoc at UMD, members of their CS PhD adcom asked me to evaluate applicants from Vietnam. During my time at UNL and now here at Mason, I have looked at Vietnamese applications (whether they are assigned to me or not) and provided input to their reviewers, e.g., X is the top tech school in Vietnam and so it should be \emph{top} instead of \emph{unknown} foreign, which makes a huge difference. @@ -832,8 +857,8 @@ \subsection{Grades}\label{sec:gpa} \paragraph{Bad Grades} While having good GPA might not help much (again, because of other much more important factors), -having bad GPA can be a \red{red flag}. In fact, many universities have a minimum GPA requirement (e.g., > 3.0) and will automatically reject applications with lower GPAs. -If you have bad grades, you should explain them in your SOP or have your LOR writers explain them in their letters. +having bad GPA can hurt your application. Many universities have a minimum GPA requirement (e.g., $> 3.0$) and will automatically reject applications with lower GPAs. +If you have bad grades, you should explain them in your SOP or better yet, have your LOR writers explain them in their letters if they know the reasons. Moreover, having bad grades in relevant courses, e.g., Math and CS, can be a \red{red flag}. Adcom members often scan through transcripts (\autoref{sec:ievaluate}) looking for C and lower grades in Math and CS courses and might raise concerns if they see several of them. @@ -1291,19 +1316,19 @@ \subsection{Schools that offer PhD in CS} \subsection{Selecting and Ranking Schools}\label{sec:selecting-ranking-schools} \begin{center} - \includegraphics[scale=0.6]{files/c1.png} + \includegraphics[scale=0.5]{files/c1.png} \end{center} -International students often put schools into two bins: (i) top schools that they dream about, and (ii) everything else. They commonly use rankings from US News, which is not transparent and questionable\footnote{\url{https://cra.org/cra-statement-us-news-world-report-rankings-computer-science-universities/}}. Sometimes they evaluate based on the reputation of the school's undergrad program or the reputation of the school's non-CS programs such as medical, math, or physics. -They even rank universities based on popular places they know in the US, e.g., California, Texas, and New York. +Many students just put schools into two bins: (i) top schools that they dream about, and (ii) everything else. They commonly use rankings from US News, which is not transparent and questionable\footnote{\url{https://cra.org/cra-statement-us-news-world-report-rankings-computer-science-universities/}}. Sometimes they evaluate based on the reputation of the school's undergrad program or the reputation of the school's non-CS programs such as medical, math, or physics. +Many international students rank universities based on popular places they know in the US, e.g., California, Texas, and New York. -Instead of these superficial criteria, you should consider the CS program and the research interests of faculty members. +Instead of these superficial criteria, you should specifically consider the CS program and the research interests of faculty members. You can learn about these using resources such as \href{https://csrankings.org}{CSRankings.org}. You will be very surprised to learn that a school that you didn't know much about can have very strong research in your topic (and vice versa, a school you thought highly about might have no faculty working in the research field you're interested in). This is also a good way to learn about individual faculty, e.g., who works on what, and well-known CS conferences\footnote{In CS (and probably only in CS), conferences, not journals, are often the main venue to publish research findings (see \href{https://homes.cs.washington.edu/~mernst/advice/conferences-vs-journals.html}{why here}).}. %\autoref{sec:ranking} gives the top 50 CS programs in the US according to CSRankings. \begin{commentbox}[Dat:] Most Vietnamese students, including those from top schools, \textbf{do not know} about CSRankings. Maybe applicants who worked at top research places such as VinAI would know about it. \end{commentbox} -\paragraph{What matters to you?} While many find CSRankings useful, it is not perfect, and in general, rankings are superficial, and you need to do more research to be informed and make better decisions. For example, you might prefer places with a large community from your country (e.g., Northern VA has a large population of Vietnamese). You might prefer places with high-tech companies (Seattle or Silicon Valley), with many outdoor activities (e.g., hiking, skiing), or with better weather (\emph{``PhD can be depressing, so would you rather be depressed in California or New York?''}). You might need to think about the cost of living, e.g., living in California is more expensive than in Virginia, or places with high crime rates (note that while some universities might be in a high-crime city but the campus itself is safe). +\paragraph{What matters to you?} While many find CSRankings useful, it is still superficial as every other ranking. You should consider other factors that matter to you. For example, you might prefer places with a large community from your country (e.g., Northern VA has a large population of Vietnamese). You might prefer places with high-tech companies (Seattle or Silicon Valley), with many outdoor activities (e.g., hiking, skiing), or with better weather (\emph{``PhD can be depressing, so would you rather be depressed in California or New York?''}). You might need to think about the cost of living, e.g., living in California is more expensive than in Virginia, or places with high crime rates (note that while some universities might be in a high-crime city but the campus itself is safe). If you get admission to several places, you should consider attending Open Houses (\autoref{sec:accepted}) and contact profs. that you're interested in at those places and talk to them. They would be more willing to chat with you now that you have been admitted. Ask questions about \href{https://github.com/dynaroars/dynaroars.github.io/wiki/Answers-to-Ph.D-Advisor-Guide}{their work}, how they manage students, and their expectations. You can even ask to contact their students. @@ -1572,8 +1597,9 @@ \section{How long does it take to evaluate an application?}\label{sec:ievaluate} For each application, our system compiles a single PDF file, which consists of a summary (degrees, GPAs, etc), transcripts, test scores, LoRs, a CV, SOP, and writing samples. I usually read in this order. I start with the \emph{summary}, checking for low GPAs or test scores below the university minimum (\autoref{sec:standard-tests}). I then skim the \emph{transcripts} for low grades in relevant courses, noting issues like ``many low grades in main courses or unknown international school with good GPA''. These are not as important as LORs or SOP but I read them first because the review system has questions about them (e.g., ``is GPA good? is IELTS sufficient?'') -I read carefully \emph{strong LoRs} and skim weaker ones, noting either strong letters from well-known professors/researchers or generic content. I skim over \emph{CV} and look for publications, research experiences, and notable achievements, noting things like ``published papers in top venues'' or ``gold medal in an international competition.'' -I read strong \emph{SOPs} thoroughly (and skim weak ones). I note whether the SOP is exciting, research-oriented, standing out, e.g., applicant is minority (\autoref{sec:urm}), and tailored to our program (e.g., if they are familiar with the work of some faculty or have talked to them). These notes are entered into the evaluation system. +I read carefully \emph{strong LoRs} and skim weaker ones, noting either strong letters from well-known professors/researchers or generic content. I skim over \emph{CV} and look for publications, research experiences, and notable achievements. I take notes of things like ``published papers in top venues'' or ``gold medal in an international competition''. + +I skim weak \emph{SOPs} but read strong ones thoroughly. I note whether the SOP is exciting, research-oriented, standing out, e.g., applicant is minority (\autoref{sec:urm}), and tailored to our program (e.g., if they are familiar with the work of some faculty or have talked to them). These notes are entered into the evaluation system. Finally, I enter my decision, which is usually either a rejection or an offer of admission with full funding (e.g., from TA). I also recommend very strong candidates for the University Presidential fellowship. Note that while the system has other decision options, eg., admit without funding, provisional admission (e.g., if they need to take some courses), I rarely use them, simply because we want to fully fund our PhD students. @@ -1764,8 +1790,7 @@ \chapter{Domestic Students}\label{sec:domestic-students} \epigraph{\vspace{-0.2in} I’m not a bad guy! I work hard, and I love my kids. So why should I spend half my Sunday hearing about how I’m going to Hell?}{\textsc{The Simpsons}} -Most of what is written in this document applies to both domestic\footnote{For simplicity domestic means you did your undergrad (or MS) at a US university.} and international students. However, there are some differences and benefits that domestic students should be aware of and can leverage to improve their chances of admission. - +Most of what is written in this document applies to both domestic\footnote{For simplicity domestic means you did your undergrad (or MS) at a US university. Different universities might have different definitions, e.g., permanent residents or US citizens.} and international students. However, there are some differences and benefits that domestic students should be aware of and can leverage to improve their chances of admission. \paragraph{Standing out \autoref{sec:improve-your-chance}} There are \emph{few} domestic applications compared to international ones, i.e., domestic students are the minority (\autoref{sec:urm}) in the CS PhD application pool. Many US universities want to increase the number of domestic students in their programs (and as mentioned later, there are specific fellowships and funding for domestic students). That makes you stand out and can help your case. @@ -1789,7 +1814,7 @@ \chapter{Domestic Students}\label{sec:domestic-students} \chapter{MS Admission}\label{chap:ms} \chapterinfo{MS focuses on coursework and prepares you for \emph{industry}, while PhD focuses on research and prepares you for academia or research.} -While both MS and PhD programs are graduate degrees, they are \emph{very different} in terms of objective, admission requirements, course requirements, duration, and funding. This section provides guidance on applying to MS programs in CS. +While both MS and PhD programs are graduate degrees, they are \emph{very different} in terms of objective, admission requirements, course requirements, duration, and funding. This section discusses the differences and provides guidance on applying to MS programs. \section{Differences between PhD and MS} @@ -1831,7 +1856,7 @@ \section{Differences between PhD and MS} \section{MS Admission}\label{sec:ms-admission} -MS programs are much less competitive than PhD programs, i.e., you're likely to get in if you can afford it. Some refer to MS programs as a \emph{cash cow} for the department, as MS students are often not funded and have to pay tuition. +In most cases MS CS programs are much less competitive than PhD programs, i.e., you're likely to get in if you can afford it. Many think of MS programs as a \emph{cash cow} because students are often not funded and have to pay tuition. While admission requirements are similar to PhD programs (e.g., GPA, SOP, LORs), research is not a focus in MS programs. In most cases, the main requirements are that you have sufficient background in CS, e.g., through your undergrad degree. This does not mean MS programs are easy to get admitted, but the requirements are much lower compared to PhD programs. @@ -1866,9 +1891,9 @@ \subsection{Application Materials} \subsection{Funding}\label{sec:ms-funding} Unlike PhD programs, which often have funding (\autoref{sec:funding}), MS students are typically not funded. The reason is because MS students do not focus on research and thus are not funded through RA and PhD students have priority for TA positions. \begin{itemize} -\item (RA) Profs. are not willing to take MS students as RAs because they are not around long enough to be productive. It can take a while for a student to get up to speed and start being productive, and by that time, they are about to graduate. Moreover, the goal of most MS students is to get a job, not to do research, so they are not as motivated to do research. +\item[\textbf{RA}] Profs. are not willing to take MS students as RAs because they are not around long enough to be productive. It can take a while for a student to get up to speed and start being productive, and by that time, they are about to graduate. Moreover, the goal of most MS students is to get a job, not to do research, so they are not as motivated to do research. -\item (TA) MS students are typically not given TA positions because PhDs are given priority. Many depts. do not have enough TAs for PhD students and so they cannot afford to give them to MS students. +\item[\textbf{TA}] MS students are typically not given TA positions because PhDs are given priority. Many depts. do not have enough TAs for PhD students and so they cannot afford to give them to MS students. \end{itemize} However, there are always exceptions, e.g., if you have a strong background and can demonstrate that you can be productive in research, then a prof. might be willing to take you as their RA. @@ -1887,7 +1912,7 @@ \section{Selecting and Ranking Schools}\label{sec:selecting-ranking-schools-ms} \item \textbf{Ranking:} Unlike PhD programs where the adviser and research are perhaps the most important factor, in MS programs the ranking of the school is more important. Thus, you should consider the ranking of the school in CS, especially in the specific area you're interested in. For example, if you're interested in software engineering, you might want to consider schools with strong SWE programs. -\item \textbf{Living Cost:} MS students are typically not funded, thus living cost can be a big hassle. You should consider the tuition and living expenses of the area. Note that sometimes living costs, e.g., renting, seems scary at first but students often find way to make it work, e.g., by sharing an apartment with other students. +\item \textbf{Living Cost:} MS students are typically not funded, so living cost can be a big factor. You should consider the tuition and living expenses of the area. Note that sometimes living costs, e.g., renting, seem scary at first but students often find way to make it work, e.g., by sharing an apartment with other students. \end{itemize}