TDDD89 Scientific Method
Seminar 4
Seminar 4
Purpose
- To understand how to formulate and discuss results, and critically review results and methods employed in a thesis.
- To understand how to put a thesis in a wider context.
IT systems in a wider context
As engineers, we have a great responsibility in designing IT systems as they will have far greater effects than we often imagine. During your master's thesis work you will get to analyze, develop and evaluate the effects of IT systems. Some of you will conduct studies on the design and requirements of software, others on the maintainability of software, and yet others will study the technical feasibility of certain technical platforms. Depending on the type of work that you do, you will need to place your work in a proper wider context and reason about the implications of your work. For this purpose, you will need to prepare by understanding some of the general effects of IT systems (see the links below), and by selecting one of the publications in the reading list.
IT systems have profound effects on users, organizations and society. IT systems have enabled us as a society to transform how we find partners, how we find work and learn, how we communicate and make our purchases. However, the promises of the digital economy to reduce energy demands [1] have not materialized [2]. Although digitalization efforts are promoted by the highest levels of government [3] in part to promote a more efficient economy, waste from constant upgrade cycles [4] and the business processes in both software and hardware industries to leave functional hardware platforms behind in upgrade processes known as "planned obsolescence" [5] which can be caused by software as well as hardware [6].
As for the electronics themselves, IT systems are built using minerals from the conflict-ridden regions and cause tensions and violence [7], use critical raw materials [8] that grow ever more scarce in supply and are hard to replace [9], they may be constructed in such a way as to consume enormous amounts of power [10a] and water [10b], [10c]. Electronic waste is becoming the fastest growing waste stream in the world fueled mainly by "higher consumption rates of electric and electronic equipment, short life cycles, and few options for repair" [11].
Among the effects on people, we know that social-media use may lead to social ills [12]. Also, more interconnected systems present us with more vexing security problems as we no longer have full control over where sensitive data is stored in the cloud [13], and we must be much more cautious about how we provide access to information [14]. The prospects for misuse of AI-technologies for surveillance have prompted scientists to post open letters calling for restriction of use [15]. Machine Learning algorithms are intended to harness trends in data, which can be prone to bias (e.g., ethnic bias) [16] and thus can have severe consequences if used e.g. in criminality prediction [17], and may be used to increase social inequalities and prey on vulnerable user groups [18]. When using algorithmic trading of financial instruments, we may hide the fact that financial markets are social meeting grounds where people find agreements based on their subjective appreciation of assets being traded and ultimately one another [19], or tilt the balance of power in favor of big institutions [20].
A good summary of such aspects of IT impact on society and the corresponding ethical issues can also be found in Section 2.6 (The Societal Dimension, pp. 110-134) of the HiPEAC Vision 2019 roadmap document [21a-g] (select any of the subsections 2.6.1-2.6.7).
Preparations
Read the chapters* covering Method, Results, Discussion and Conclusion
of the
sample Master's thesis pertaining to your team.
* Note that the chapter structure in a master thesis needs not
be formally structured in a generic and breadth-first way like this;
depending on the topic there can be good reasons for a problem-specific
chapter structure, e.g. with a more
depth-first presentation of some results directly followed by their discussion
before other results are presented,
which is often easier to read and matches the usual flow in technical research articles.
Each individual should read at least one text from the references [1]-[21] from the above section on the wider effects of IT systems in society, or a similar text obtained from another resource on the wider implications of the kind of systems that you plan to contribute to. (In case the references above contain many entries, or are excessively long, select a section of text of approximately the length of a newspaper article to read.)
Each individual will read also one of the scientific papers from below that you find most relevant; where there is a choice, try not to choose the same as the paper read by your team mate, and try to avoid that all in your group read the same paper. Hence, discuss and agree in your group before the seminar (e.g., at the end of Seminar 3) who reads which papers.
If no suggested paper seems appropriate for understanding the wider implications of the type of thesis that you intend to write, you can choose another (peer-reviewed) paper, if it is intended to provide guidance on understanding the wider implications of the work you intend to do in your final thesis.
Take notes while reading the paper, and make sure that you can present the main contents of the paper to the peers in your team.
Finally, read the ACM Code of Ethics and Professional Conduct and the IEEE Code of Ethics.
In sum, each individual should read:
(A) a (newspaper or HiPEAC Vision) text from the section on IT systems in a wider context above,
(B) one scientific paper selected from the following options:
-
(B) For teams with thesis proposals on maintaining and developing software
(applicable for thesis themes 1, 2, 3):
- Zoya Durdik, Benjamin Klatt, Heiko Koziolek, Klaus Krogmann, Johannes Stammel and Roland Weiss: "Sustainability guidelines for long-living software systems." 28th IEEE International Conference on Software Maintenance (ICSM). IEEE, 2012.
- Paris C. Avgeriou, Davide Taibi, Apostolos Ampatzoglou, Francesca Arcelli Fontana, Terese Besker, Alexander Chatzigeorgiou, Valentina Lenarduzzi, Antonio Martini, Athanasia Moschou, Ilaria Pigazzini, Nyyti Saarimäki, Darius Sas, Saulo Soares de Toledo, Angeliki Tsintzira: "An Overview and Comparison of Technical Debt Measurement Tools". IEEE Software, vol. 38, no. 3, pp. 61-71, May-June 2021.
- (B) For teams with thesis proposals on developing, evaluating and optimizing (parts of) software systems
(applicable for thesis themes 2 and 6):
- Sebastien Lafond, Simon Holmbacka and Johan Lilius. "Energy aware software: Issues, approaches and challenges." 7th International Green and Sustainable Computing Conference (IGSC), Hangzhou, 2016, pp. 1-8. IEEE.
- Miyuru Dayarathna, Yuanlong Li, Yonggang Wen and Rui Fan: "Energy consumption analysis of data stream processing: a benchmarking approach." Software Practice and Experience 47:1443-1462, Wiley, 2017.
- (B) For teams with thesis proposals on designing (parts of) hardware systems
(applicable for thesis theme 6):
- Christoffer Åleskog, Håkan Grahn, Anton Borg: Recent Developments in Low-Power AI Accelerators: A Survey. Algorithms 2022, 15(11), 419, 2022.
- Mona Komeijani, Erinn G. Ryen, and Callie W. Babbitt. "Bridging the Gap between Eco-Design and the Human Thinking System." Challenges 7.1 (2016): 5.
- (B) For teams with thesis proposals on
defining requirements for and designing
software (applicable for thesis themes 1, 3, 5):
- C. Becker, S. Betz, R. Chitchyan, L. Duboc, S. M. Easterbrook, B. Penzenstadler, N. Seyff, and C. C. Venters (2016). Requirements: The key to sustainability. IEEE Software, 33(1):56-65.
- (B) For teams with thesis proposals in
theoretical computer science:
- William P. Thurston: "On proof and progress in mathematics." Bulletin of the American Mathematical Society 30.2 (1994): 161-177.
- (B) For teams with thesis proposals on investigating machine learning or AI
techniques (applicable for thesis theme 4):
- A. Birhane, P. Kalluri, D. Card, W. Agnew, R. Dotan, and M. Bao: "The values encoded in machine learning research". In: Proceedings of the Conference on Fairness, Accountability, and Transparency (FAccT'22), June 2022, Pages 173-184. ACM.
- Alexandra Chouldechova and Aaron Roth: "A snapshot of the frontiers of fairness in machine learning." Communications of the ACM 63(5):82-89, May 2020.
- Radosvet Desislavov, Fernando Martínez-Plumed, José Hernández-Orallo: Trends in AI inference energy consumption: Beyond the performance-vs-parameter laws of deep learning, Sustainable Computing: Informatics and Systems, Volume 38, 2023.
- (B) For teams with thesis proposals on IT
security (applicable for thesis theme 5):
- Oscar Lasierra, Gines Garcia-Aviles, Esteban Municio, Antonio Skarmeta, Xavier Costa-Perez: European 5G Security in the Wild: Reality versus Expectations, Proc. 16th ACM Conference on Security and Privacy in Wireless and Mobile Networks (WiSec'23), May 2023, ACM.
- Stephen B. Wicker and Dipayan Ghosh: "Reading in the panopticon - Your kindle might be spying on you, but you can't be sure." Communications of the ACM 63(5):68-73, May 2020.
-
(B) For teams with thesis proposals on maintaining and developing software
(applicable for thesis themes 1, 2, 3):
(C) The remaining chapters (on method, results, discussion and conclusion) of the published Master's thesis pertaining to your team.
(D) The ACM Code of Ethics and Professional Conduct
and the IEEE Code of Ethics.
Questions
For the newspaper text/online resource that you read (A), answer the following questions:
1. What were the main arguments made? Present the argument and how they are supported in the text.
2. Do you agree with the author? Justify why or why not.
3. Should you as an engineer consider what is presented in the article? If so, how?
For the scientific paper that you read (B), answer the following questions:
4. What are the main arguments made in the paper? Present the arguments and supporting evidence from the text.
5. How can you make use of the material in the scientific paper when framing your own thesis work in a wider context?
6. How would you define the context and possible wider consequences of your thesis? Justify given the literature.
For the sample thesis that you read, each individual should answer the following questions. Justify your answer by referring to the course literature including the papers that you had read before Seminar 2.
7. How can you explain the results, and how they have been obtained through the method described? Do they seem to address the research questions properly?
8. How is the replicability, validity and reliability of the results discussed?
For definitions of validity and reliability, see for
example the Instructions for final thesis reports
(
Swedish/
English) and Section 5.2.3 of
Runeson and Höst (2009).
9. Overall, how would you assess the sample thesis chapter(s)/section(s) about Method of the thesis based on the grading rubric?
10. Are any ethical and societal considerations considered in the sample thesis?
If yes, where and how?
If not, do you have any suggestion of such aspects that could have been considered?
Refer to Section 3.3 in
Runeson and Höst (2009)
for descriptions of some ethical considerations that are of
importance during case-study research, as well as the papers in the reading list above.
If ethical and societal considerations are not considered appropriately, explain which
considerations seem relevant for the thesis at hand and how you would have liked them
to be taken into account.
Finally, consider the ACM Code of Ethics and Professional Conduct and the IEEE Code of Ethics (D).
11.
Which of the two documents would you, in principle, be ready to sign with your name? Why or why not?
Based on these documents, do you see any potential ethical conflicts with respect to your chosen thesis topic in a wider context?
Submissions
Your answers need to be available in the Seminar4 folder in LISAM at the seminar start.Note that all answers in this seminar are individual, not per team.
The seminar
During the seminar, you will discuss your answers to Questions 1-11 within the seminar group.After the seminar
Individually, send a 4-5 lines informal reflection about the seminar discussion from your own point of view to your seminar leader by email, by the end of the seminar day. Focus on one insight that is important for you. (Your seminar participation will not be fully approved in webreg before also this reflection has been received.)
Page responsible: Christoph Kessler
Last updated: 2024-12-02