Writing for Dissertations and Publications in STEM Fields Lawrence O. Hall Distinguished University Professor Computer Science and Engineering, ENG 060 University of South Florida Tampa, FL 33620 email@example.com
Some Background on me Writing in STEM Disciplines Where to Publish: Conference, Journal, Online media Questions? Outline
B.S. Florida Institute of Technology Wrote a little for school newspaper about sports They asked as I (to my surprise) was named outstanding in English in High School Ph.D.Florida State University Wrote several papers and turned Ph.D. Dissertation into a book Have published 90 journal papers and over 290 conference papers with over 23,000 Citations (GS) My Background
Reviewer for conferences, journals Editor, IEEE Transactions on Systems, Man and Cybernetics: Part B Associate Editor, Editor in Chief Vice President, Publications, IEEE Biometrics Council President, IEEE Systems, Man and Cybernetics Society Member of IEEE Publications, Products and Services Board since 2010 Experience with Paper Writing
Why Should We Write? To tell people about our research so they can build on it and use it.
Clear, simply written, Abstract that tells what the problem is, how it was solved and why it matters (best solution ever?, first solution?) Clear Introduction and clear conclusions that summarize advances Use social media to tell people of great work So many papers appear now, that one is only going to have time to read helpful ones How do I get people to read my paper?
Focus is on Factual Technical Content Simple prose, short sentences, itemized lists Number equations, avoid unnecessary math Discuss figures, tables in text; include descriptive captions; Cite sources If a theory paper, give proof If an experimental paper, describe in sufficient detail for others to replicate Provide Code and Data if you can Writing in STEM Disciplines
You can never write something too simply. Ask a colleague if they understand something complex to write up To make code usable (for you too), put in COMMENTS and do a couple of tests on different data configurations before release Writing in STEM Disciplines
Include what didn’t work (unfortunately most papers don’t); science would advance much more rapidly if we didn’t have to repeat what others already tried and failed Do not cut and paste from others work – Plagiarism is a major problem since it is so easy to copy others work today. Use quotes when you must do so (do so sparingly, if ever) and clearly cite quoted material Dos and Don’ts
A Good Paper is Organized into FOUR Parts: A succinct, clear and informative Abstract Tell the reader what you will tell (Introduction) Tell the reader what you accomplished (Body) Tell the reader what you covered (Summary and Conclusions) PLUS - A list of critical references Organization
What is the problem you solved? Why is this an important problem? What have others done about it? What did you do (process, equipment, cost, repeatability, performance comparison)? Why what you did is better than others? What more can be done to make it better? Where have you published? Your Dissertation Should Answer These
Introduction and Problem Statement What is the problem you solved? Why is this an important problem? Literature Survey What have others done about it? Body of Dissertation What did you do (process, equipment, cost, repeatability, performance comparison)? Why what you did is better than others? Conclusions and Future Work What more can be done to make it better? References Publication List (validates that your work is accepted by peers) Where have you published? These translate to your chapters:
NSF Format for Grant Proposals is a possible format for Thesis/Dissertation Proposals Project Summary (one page description of resulting activity including intellectual merits and the broader impacts) Written at a level appropriate for a person interested in science and technology Project Description (15 pages limit) Clear Statement of Work Objectives, Significance, Relation to longer-term goals and state of knowledge General Plan of Work Design, Experimental methods, Broader impacts, Timeline and Milestones References Thesis/Dissertation Proposals
Conferences: Typically short papers; incremental work; less rigorous review (exceptions are some CSE conferences); quick turn around; opportunity to network with other researchers and share your ideas; proceedings may not be widely distributed Networking is always important. However, conference papers are often not completely finished works Where to Publish: Conference, Journal, Online?
Journal: Well developed, more complete, longer papers; rigorous peer review with opportunities for rebuttal; traditionally considered as “archival” work; widely disseminated by publishers Where to Publish: Conference, Journal, Online?
Online (arXiv): Rapid posting; no peer review (except comments); uncertainty about longevity Where to Publish: Conference, Journal, Online?
In IEEE code can be an attachment to a journal article or put in Code Ocean via a docker Github is a great place to crowdsource code Data – In IEEE there is Dataport or as an attachment to a paper. There are various other data repositories around. Curation is an issue Where to Publish Code and Data?
Conference Reputation, Selectivity, Organizers, Popularity, Location, Publisher, Worldwide Distribution, Cost, Lead time, etc. Will the group of people you want to meet or learn from be at the meeting Is it so large you cannot find people Where to Publish: Other factors
Journal Reputation, Circulation, Submission to Publication Time, Editorial Board, Publisher, Worldwide Distribution, Cost, etc. Impact Factor is an objective measure of quality (roughly speaking it is the ratio of the number of citations to number of articles in a specified time period), from Journal Citation Report (JCR), a product of Clarivate Analytics Google Scholar H-index for journal – many papers have no to few cites and then some have lots and lots Article Influence score Where to Publish: Other factors
Reader (Library) Pays (Limits access to subscribers) Publisher distributes peer-reviewed (quality assurance) content and readers/libraries purchase and maintain archives (for ever) – Stacks of bound volumes; traditional model, almost obsolete Author is guaranteed publication regardless of ability to pay Recent variation for digital content – publisher assumes archiving responsibility – libraries/individuals subscribe to access content Author/Research Sponsor Pays (Open-Access Model) This is a new trend to increase access to published research content Some US research sponsors (NIH for example) support this to increase access to results of research sponsored by them But this would make it difficult for researchers with limited resources to publish their findings Which one is better? - This is an Ongoing Debate Who Pays for Archival Dissemination
Strunk, William. The elements of style. Penguin, 2007. Strunk, William and White, E.B., The elements of style, Pearson; 4th edition (May 25, 2019) Truss, Lynne. Eats, shoots & leaves: The zero tolerance approach to punctuation. Penguin, 2004. References