Date: Thu, 23 Mar 2000 17:36:59 -0600 From: John Womersley Subject: Re: D0 policy document on graduate student thesis topics To: Declan Keane Cc: weerts@fnal.gov Dear Declan, thank you for your enquiry about the policy for thesis topics in a large experiment like D0. I am sending you a rather detailed description of the process, written for incoming graduate students by our Physics Coordinator in 1997. I think this will address most of your questions. Let me know if you need any more help. best regards John Womersley co-spokesman, D0 Folder: GENERAL From: strovink ( Mark Strovink ) Subject: Advice to D0 graduate students on choosing a thesis topic, etc. Date: 16-DEC-1997 20:51 16 Dec 97 To: D0 graduate students From: Mark Strovink Subj: Friendly advice on choosing your Ph.D. thesis topic, and related matters Here are some FAQs relevant to your choice of Ph.D. thesis: 0. WHAT IS THE RELATIONSHIP BETWEEN A STUDENT AND D0? The D0 experiment is a research instrument, which is made available to you as a graduate student when you join the experiment. You expect to be able to use D0 data for research for your Ph.D. thesis. From interactions with people in the D0 collaboration you expect a substantial improvement in your training and education which prepares you uniquely for your future. On the other hand D0 (the collaboration) expects a contribution from you. It is all the work that you do in the collaboration, consisting of contributions to designing, building, commissioning and running of detectors and/or software, analysis of data and its documentation. It is essentially everything that is needed from the conception of an experiment to the publication of a physics result. Having experienced and contributed to all these phases of an experiment is what defines an experimental physicist. Work on the detector is particularly important. The knowledge thus gained is useful not just because you may have to build a detector yourself, some day in the future. Rather, by coming into intimate and serious contact with the detector, your understanding of the data deepens. D0 is a large and complex experiment, but it does not need to be intimidating. Among the tools and facilities which the collaboration makes available for doing physics are a world-class detector and a full mechanism for identifying, recording and reducing interesting data. There are extensive shared analysis projects on many critical aspects of the data such as calibrations, luminosity, acceptance, and related physics channels. On the other hand, the application of those resources and tools is made in a unique way to particular data by each thesis student. There is a lot of help and support available, as well as much scope for individual creativity. 1. WHO SHOULD BE INVOLVED IN CHOOSING A THESIS TOPIC? You, your thesis advisor, and the leadership (convenors) of the physics group in which you plan to do your analysis. Give very careful consideration to advice from the latter. The convenors have a broad and deep understanding of the physics program in their group; often, when you get really stuck, they will be helping you to get back on track. Find out from the convenors of all five physics groups which topics are ripe for your consideration, before you focus in on one or two groups. Outstanding opportunities usually are available; I'd be happy to offer advice as well. Keep in mind that, if you're working on a project to which, in view of its physics interest, D0 has a special commitment, you'll get more help, more visibility, and more resources. There will still be plenty of opportunity for you to contribute independently. Overall, you'll have a more exciting and rewarding experience. 2. WHEN SHOULD A THESIS TOPIC BE CHOSEN? D0 policies don't allow physics analysis topics to be "staked out" by anyone (see below). You shouldn't feel the need to "reserve" a topic well in advance of the point at which you can work on it nearly full time, applying the necessary level of experience and expertise. Instead, you gain by waiting to choose until you are ready to really plunge in, because it will be clearer at that point which topic will be best for you. How might the first few years of graduate school be spent by a D0 student? This is only an example -- your mileage may vary: Years 1-2: Primarily at home institution: taking courses, passing exams, possibly TAing. Try to spend time at D0, for example a summer or two. Begin contributing to D0 as your primary responsibilities permit. Year 3: Unless you are deeply involved in a Run 2 construction project that is centered at your home institution, this is the most important year to be resident at D0, if at all possible. Here's a unique opportunity to help build a major physics instrument [the D0 Run 2 detector, with all new hardware inside the calorimeter, much new hardware outside, and all new (modern) software]. I say "unique" because I think it's easier for you, as a grad student, to take responsibility (and credit) for key pieces of the D0 upgrade than would be the case in a project, like the LHC detectors, where the responsibilities are more intricately subdivided. You can dive in and make a real impact on Run 2 now! D0's future depends on it, and (see below) yours probably does too. On the path to choosing a thesis topic, shop during the first few months of Year 3 in the various analysis groups. Then select one or two of them; attend the meetings, get to know the people, and become familiar with the group's physics program and the thesis topics that are possible within that group. As your responsibilities in building the Run 2 system permit, accept finite and well defined analysis tasks in the group; get up to speed on using the analysis tools that you'll need for your thesis. Year 4: Primarily at D0 if possible. By this time your role in the D0 upgrade is well established. You're broadly knowledgeable about the D0 physics program and you're quick with D0 software tools. The beginning of year 4 is a good time to choose a Ph.D. thesis topic. It isn't necessarily the case that the earlier you choose a topic, the earlier you will finish. A thesis topic chosen near the beginning of year 4 is likely to be more topical, better conceived, more practical to attack, more able to build on the work of others, and better integrated with the rest of the physics group's analysis program. Moreover, if you've already made a solid contribution to the D0 upgrade, you'll be able to work more intensively on your thesis analysis at this point. 3. WHO NEEDS TO BE SATISFIED THAT MY THESIS ANALYSIS IS CORRECT? Keep in very close contact with your advisor and with the physics convenors and other members of D0 who are interested in your analysis. This is easier to accomplish, but not guaranteed, if you reside at D0; if you are back at your home institution, you should have facilities (travel, electronic conferencing, etc.) that make it possible to maintain close contact with the physics group. Around the time that you are finishing your thesis analysis and writing it up, your objectives (in rough time order) are: A. Finish the analysis to the satisfaction of the physics group and your advisor. It's vitally important to maintain a working version of an analysis note, which contains the present state of the analysis during the process of finalizing it. Studying this note is the most effective means available to your advisor and physics group for the purpose of understanding what you are doing. It's in your interest for them to reach that understanding as early as possible, so that you are all on the same wavelength and analysis steps don't need to be repeated. B. Finish the detailed analysis note. Get the convenors to request an EB, and address their concerns. In the background, write the boilerplate sections of your thesis (The Standard Model, detector appendices, etc.) (Strictly speaking, EBs must approve only what is disseminated outside D0; in principle you could graduate without any EB involvement. In practice, however, you want your results to be public D0 results -- this is why I emphasize your interaction with an EB at this stage.) C. Import the EB approved analysis note into your thesis as its main component. Address the concerns of your thesis committee. While waiting for laggard thesis readers, put your analysis note in shape for publication. Around this time, give strong consideration to scheduling a formal presentation of your analysis to the D0 collaboration (preferably in a D0PAM, otherwise in a practice talk for a conference presentation). This will broaden the appreciation within the collaboration for the work you have done, facilitating their approval of your publication. D. Graduate and publish, more or less at the same time. 4. WHAT ARE SOME OF THE PITFALLS TO BE AVOIDED? - You plunge full-time into Ph.D. thesis analysis before (and at the expense of) establishing a strong role in the D0 upgrade. This is likely not to have a positive effect on your career. Future leaders in our field need to have a broad understanding of what it takes to plan and carry out a key measurement. You need a deep appreciation of how detectors, electronics, and software infrastructure really work; the most straightforward way to get that understanding is to be involved deeply in the Run 2 construction project. And, as explained above, an "early plunge" doesn't necessarily speed up your degree. - At D0, or elsewhere, you become so absorbed in your analysis work that you lose track of what is going on in HEP as a whole. Fermilab has a variety of seminars (in general science, theory, astrophysics, accelerator physics and experimental HEP). Your university does as well. This is the place to keep up with new ideas in the field and see what you will be doing in the future. - You perform your thesis analysis in excessive isolation. For example, you convince your thesis advisor and committee that the analysis is complete -- but the physics group and EB ask some penetrating questions requiring more work to answer. Meanwhile, you move on to your first real job. The questions don't get answered completely, no one else has time to climb your learning curve, so your work gets buried. This is a tragedy for all concerned. Your career suffers because your thesis analysis was never published. Unpublished work is forgotten work; you, D0 and humanity lose the benefit of your analysis, and D0 loses the benefit of its effort in helping you. Unfortunately, this pitfall is not rare in D0; I consider avoiding it to be one of the major challenges facing the D0 physics groups. 5. WHAT IS D0'S POLICY ON "STAKING OUT" ANALYSES? "If an analysis is 'reserved', as a result of an agreement especially including the active participation of the analysis group leadership, this means that D0 supports the notion that the reservee should work on the 'reserved' analysis, and that D0 intends to facilitate this work. D0 expects that the reservee will work on this analysis in a way that does not delay publication. This does mean that D0 management will not necessarily discourage others from pursuing analysis topics which are similar or the same, especially in cases where a timely completion (=publication) is desired. This policy is meant to include cases in which the 'reserved' analysis is a Ph.D. thesis topic." I should add that the spokespersons maintain a list of thesis topics and who is doing them. As you see from the above, your presence on the list does not mean that you have exclusive rights to the topic appearing next to your name. Neither does it force you to stick to that topic if your interests change.