Dissertation Brainstorming


2008-08-28 -17:27 -0700

One-third of the work in the M.ScIT program is a dissertation project.  Once I have completed 90 units (6 modules at 15 units each), I can begin developing the dissertation proposal and project.   According to the current plan, I will have completed 90 units on April 1, 2004.  I also have a 4-week break at that point, and I will use it for initial preparation and discussion of dissertation topics.  By June 3, 2004, I will be in action on developing a proposal, enrolling a sponsor, and finding a supervisor.  This is so I may begin the dissertation work on August 19, 2004, following my 8th and last course module.  That gives me until February 28 or so to complete the dissertation and have it submitted, accepted, and graded.  I want to make sure that my completed degree work is submitted to the Board of Examiners in May, 2005, and that I can be at the July, 2005 commencement in Liverpool.

The dissertation includes a project.  The project has a sponsor and the result of the project must benefit the sponsor.  The project can be a research project, with one of the faculty or instructors as sponsor.  The project should have an Information Technology thrust.  The project is not the learning outcome for this part of the M.ScIT effort.   

Since I am retired and self-employed, I will not have an employer sponsor, but I will still look for external sponsors.  There are also different candidates for supervisors, and these are potentially candidates for sponsorship as well.

Here is initial thinking as it comes up.

-- Dennis E. Hamilton
Seattle, Washington
2003 September 20

External Sponsor



Daniel Dennett?   Macro/MicroScope in Computer Education.  Several years ago, I saw work that one of Dennett's students did to show a simulated computer and provide a way to drop through the layers of abstraction of the operating computer, down to electronic circuits.  That has stayed with me all of this time, and I am looking for some way to apply this kind of dynamic, animated drill-down in other aspects of Computer Education and also as a way to illustrate the way that computation is used to manifest and coordinate layers of abstraction.  This is about CS and IT education, especially around the introductory parts.  And it can be modularized, since it goes all the way through OOP, MDA, and the Requirements Gap in Software Engineering.  
Steve McConnell?, David Parnas?.
Kent Beck?
Gail Miles Web-based light-weight structures for the management and engineering of software projects, especially those with web-centric delivery and deployment mechanisms.  Looks at incremental development and the web site as both integration structure, documentation, and prototype confirmation fixture.  Expand on the ideas that were started in the MSC-SE module on scaffolding and incremental management along with incremental development. 
"   Team Projects in IT/SE Education.  I found that the conduct of team projects, especially as part of a software-engineering course, was an amazing problem.  There are many things to consider, and the purpose of this project is to look at conducting projects in a way where there is attention on the accomplishment of teamwork and having a team result.  This seems to be missing from the design of project-centered courses, and I suspect that the faculty actually don't know how to be team players either. This is about the culture and lore of software development and about who we are in project situations more than whatever the content subject matter happens to be.  What is there to provide here?  What would be a meaningful project in this area?
Microsoft, the Trustworthy Computing Group?   Situating Trust: Development of the "trust point" aspect of behavioral architecture, and also.
The Open University Folk?   Collaborative Participation systems -- coordinated activity across related and different ontologies: the feuding lexicographers problem.
Doug Englebart   OHS - The Open Hypertext System - carve out some deliverable piece from this vision and take it on.
Peter Denning, a CSE or ITE group?   [dh:2004-02-05] A lot of what I deal with bears on computer education and basic principles.  I can see something around their popularization, as in the more playful parts of nfoWare.  I am not sure how this could be raised to match up with having IT education be principled and also tied into Great Principles as well as computing as a profession.  The current interview in Ubiquity has me wonder more about that, especially the collapse of programming with computer science.
    [dh:2004-02-07] I was watching a problem in our computer-mediated communication process because of a breakdown in the FirstClass software used on Embanet to serve our on-line courses.   There are visible "moments-of-truth" with the system that would suggest that a major breakdown and deterioration of the arrangement is inevitable.  One is that users already have to fight with whatever model the implementation realizes because of conceptual-model dissonance and metaphor shear.  This experience has me wonder about how one establishes requirements and practices for having distributed communication be as easy to troubleshoot (and to recover in) as is sometimes possible with mainframe or other local systems.  I was motivated by a question from instructor Kathleen Kelm to look at this as a dissertation opportunity.  I am reminded of Van Horn's dicta and how much apologizing is done for software, and how much we train ourselves to tolerate it.  I am wondering what extension of Van Horn's dicta would support coherence and confirmable experience in distributed operation, whether e-mail breakdowns or something like the conceptual and performance difficulties with FirstClass.  I am also wondering how to cast this as an IT problem.  There are a variety of ways I can see, but I am not sure that is what I care about.
    [dh:2004-02-08] I was making some notes on the activities of W3C and I turned my attention to the Web Accessibility Initiative. I started to tear up about the prospects for having computer-mediated communication and instruction be at a level where anyone could take advantage of it, without any barriers to accessibility.  I have avoided digging into this for my web sites or for my creation of Java libraries and such.  There is certainly an opportunity for me to do more, though I don't know if it fits as an M.ScIT dissertation project.  Maybe even having an accessible Wiki would be a major contribution.  I am not sure where to start, but it brings me back to having been a fouding member of the Rochester SIGCAPH chapter in the early 80's and in promoting David Boylan's creation of Braille support tools for programmers at Sperry Univac in the late 60's.  I don't have the expertise or sense of this, but I can see taking it on like a social-service leave project, except that I can volunteer my efforts independently.

[dh:2004-02-09]  I also noticed a beautiful way that classmate Susan Abu Azab mentioned WAI in conjunction with the W3C Commitment to having universal access to the Web.  She and other students mentioned some related activities, including voice, device independence, multi-modal operation, and internationalization.  These are all interesting aspects to be identified.  I also recall that accomodation for disabilities is a highly-customized activity.  For example, physical limitations have a wide variety of expressions and characteristics that defy generic approaches, so customization and adjustment of adaptive technologies is important.  This reminds me of the ductility issue in security.

    [dh:2004-02-09] Classmates Dieter Heimlich, responding to a question from Mark Chao, wonders whether his router should be upgraded to a router/firewall combination.  This reminds me of Bruce Schneier's injunction to not have brittle defenses and, as part of ductility have multiple layers of security so that a system bends under attack but does not break in a way where there is suddenly complete exposure and no clue what exploits have been conducted, data lost, and systems subverted.  This is useful to describe under my centrale configuration support.  There is a thesis topic here around how does one know what the condition of the security systems are, no when there is a failure (exploit or not) and how to respond.  I can see treating it as a risk management exercise.  But with regard to programmatic procedures, there is the interesting question of how to restore/upgrade protection.  For example, if the operating-system key and password safe were penetrated, that would be pretty brittle it seems to me.  The other aspect of this is how does one deal with social engineering and staff not being vigilant with regard to security concerns -- where are passwords written down, etc.  In my response to Dieter I will mention the value of outgoing protection and the desire to demonstrate good corporate citizenship.  I don't know if behavior around hoaxes fits in here.  

Hard Hat Area


You are navigating Orcmid's Lair

created 2003-09-20-13:42 -0700 (pdt) by orcmid
$$Date: 13-08-22 13:00 $
$$Revision: 31 $