From sergio Mon Jun 19 17:27:21 1995 Return-Path: Received: from slab16.unr.edu (slab16.seismo.unr.edu) by quake.seismo.unr.edu (4.1/1.34) id AA23370; Mon, 19 Jun 95 17:27:21 PDT Date: Mon, 19 Jun 95 17:27:21 PDT From: sergio (Sergio Chavez) Message-Id: <9506200027.AA23370@quake.seismo.unr.edu> To: louie Subject: McMechan Content-Length: 10927 X-Lines: 217 Status: RO John, Here goes McMechan's reply concerning the document we talked about. Best regards, Sergio Return-Path: Received: from utdallas.edu by quake.seismo.unr.edu (4.1/1.34) id AA14079; Wed, 14 Jun 95 09:39:32 PDT Received: (from root@localhost) by utdallas.edu (8.6.11/8.6.11) id LAA22403; Wed, 14 Jun 1995 11:39:00 -0500 Date: Wed, 14 Jun 1995 11:39:00 -0500 From: George A Mcmechan Message-Id: <199506141639.LAA22403@utdallas.edu> Received: from ariel.utdallas.edu by utdallas.edu (Brelay v6.01) with BLIMP; Wed, 14 Jun 1995 11:38:59 CDT To: sergio@quake.seismo.unr.edu Subject: expectations & reprints Cc: mcmec@utdallas.edu Status: R Sergio, as you requested, below is the list of expectations for students; certainly you may use it, provided you acknowledge me as the source. I originally wrote these up after finding that I was repeating the same information to many students....it's easier to write it out once & then give them a copy (some of them still don't read it though!!). The reprints of the inversion paper just arrived, so I'm sending a package with all the reprints you requested earlier (it was in the mail yesterday). OK, have a good day!! George McMechan UTD 6/14/95 ------------------------------------------------------------------------ THINGS TO KEEP IN MIND (FOR GRADUATE RESEARCH ASSISTANTS) 1. Your first PRIORITY should be your research. All other academic activities (including classes) should be thought of mainly as providing you with the tools & information you need to do research. Choose your classes and term projects so that they support your research. It's necessary to keep class grades up to receive continued support, but keep them in pers- pective. RA support is for research, not for taking classes. 2. Research Assistantships are obtained from grants which have comittments for results in a specific time period (usually a year). Any reasonably-well motivated student who works steadily will easily make significant PROGRESS in a year. The expectation is that a project will be completed every 1.0 to 1.5 years, as continued support depends on grant renewals which are mainly dependent on previous performance. Thus it is important to write a paper as soon as the research has produced good results. 3. Do not think in terms of writing a thesis or dissertation. Think in terms of writing PAPERS for technical journals. An MS thesis is one paper; a PhD is normally four (5+ if any of them are 'notes' rather than 'papers', and 3 if any of them involve extensive field work or large data sets). 4. Your graduation date is independent of when you WANT to graduate; it depends entirely on the production of the publication- quality papers described in #3. In view of the work required to put a dissertation together (even after the papers are done), to prepare the defense, and to do the revisions required by the journal reviewers before publication, and by the dissertation committee, all papers are expected to be submitted for publica- tion prior to the start of the semester in which you intend to graduate. 5. The only research that you get credit for is that which is COMPLETED. Projects begun but not finished have no long term value. A project is not complete until the paper is accepted for publication and the software are fully documented, cleaned up and archived. Being a hard worker and putting in long hours is a necessary but NOT SUFFICIENT aspect of doing a dissertation. 6. Not everything that you try is expected to be a viable solution or algorithm (that's the nature of research). Evaluate your work CRITICALLY, both at the beginning, and throughout the project. If you discover a serious flaw, don't take it personally, just admit it, stop, and try something else. The faster you discover a 'dead-end', the faster you will get to a real solution. There is nothing more wasteful of your time and energy than continuing work in a non-productive direction. 7. A viable research project is on a topic that is ORIGINAL, SIGNIFICANT, and that is at the LEADING EDGE of current knowledge in the field. This requires reading the current research literature to find out what others have recently done, and what the most productive approaches are. There are good projects that are not possible to do at this time because of the lack of computing capacity or a key idea; try to find one that has just now crossed the threshold from being impossible to being possible (i.e. right at the 'leading edge'). 8. Define your research TOPIC in terms of some question to be answered, a hypothesis to be tested, some phenomenon to be investigated, a new technique to be evaluated, etc. Then, see what tools you need to to solve this problem, and whether they are available or need to be produced. The reverse direction (starting with a technique and trying to fit a problem to it) is too restrictive of the range of possibilities and does not allow for full use of your creative imagination, which is the usual source of innovation. 9. Many projects involve development of TOOLS, but this by itself is not a research project. For example, you may write a program for solution of the 3-D viscoelastic wave equation using the pseudospectral method on a staggered grid; this is new, but is NOT a research contribution and has very little value until it is used to solve some scientific problem. 10. Don't 'reinvent the wheel'. Use the available resources, especially existing software, to your advantage. If a tool already exists (such as finite-difference modeling), then it is a waste of your time to do it again; make sure that you fully understand what the program does and its limita- tions before you use it, but then go BEYOND that to investigate some new question (that is, to do research). When using software produced by someone else, remember to give them CREDIT for their contribution; others will do the same for you when using your programs later. 11. A research project will normally have the following elements: a) CONCEPTUALIZATION of a new idea or algorithm b) IMPLEMENTATION of the idea through production of new software or modification of existing software c) TESTING the process using synthetic data d) APPLICATION to field data e) error ANALYSIS f) write & submit the PAPER g) SOFTWARE documentation & archiving 12. The best SOURCES for good research topics are the past 5 years of the main research journals, discussions with others, and observed features in field data. The key journals for seismology are: a) Geophysics, for exploration b) Geophysical Prospecting, for exploration c) Geophysical Journal International, for theoretical topics d) Journal of Geophysical Research, for data analysis e) Bulletin of the Seismological Society of America, for earthquakes and engineering f) others, such as the Journal of Geophysics, Wave Motion, the Journal of the Acoustical Society of America, the Journal of Seismic Exporation, IEEE Transactions of Geoscience and Remote Sensing, and many others, are also potential sources. Math, Physics, Engineeering & Medical Journals can also often give good ideas. 13. TALK to each other! Find out what the other students are doing, and be willing to SHARE your experiences & knowledge with them. One of the best resources you have is the unique perspectives of your colleages. While you are expected to do the key parts of your research INDEPENDENTLY, you are also expected to contribute to the health of the group as a whole; in a COOPERATIVE atmosphere, everyone benefits. 14. Take RESPONSIBILITY for the smooth operation of your environment. (don't leave piles of plots or printouts in public areas (what you do at your own desk is up to you)); fill up the copier trays when they run out; let the secretary know when supplies of anything are getting low (don't wait until there is none); sign out books borrowed from the reading room, and return them as fast as possible; be tolerant & respectful of your neighbors (someday the person sitting next to you may be your employer!); etc. 15. Produce a 'TIME-LINE' in which you set achievable goals for completion of each sub-project (each paper) with target dates for completion. Evaluate how you are doing on a regular basis. 16. Work steadily so it gets done in reasonable time; projects left until deadlines are usually of inferior quality. Any work that is not of professional (publishable) quality will have to be redone or abandoned and so is not worth doing at all. 17. Good research always raises new, interesting questions; make a LIST of these ideas as they arise, so they are not forgotten, but DON'T leave your current project to persue them; remember #5 above! It's OK (even helpful) to work on a second (or even a third) project in parallel, but give high priority to the central project so that it will get done. When starting a new project, look at your list and ask yourself which of the pos- sibilities is the BEST (not the easiest or the fastest) science that you can do at that time; that will be your next project. 18. Develop a PROFESSIONAL attitude; think your project through in considerable detail to ensure it meets the criteria listed above so that your work is good science, and your progress is steady rather than a series of crises. 19. The best solution is usually the simplest one. 20. After reading this, don't be preoccupied by it; let your motivation come from your project itself. G. McMechan Lithospheric Studies