-
Notifications
You must be signed in to change notification settings - Fork 0
/
out
executable file
·27 lines (27 loc) · 14.3 KB
/
out
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
./Chapter4/c_4_data_as_facts.tex:131:The participant is a researcher external to the company. Data is embodied in reality and can be discovered through measurement. Data represents the fundamental relationship of things to one another.
./Chapter4/c_4_data_as_facts.tex:134:This interview also involved someone external to the company. Their conception of data represents something with a strong sense of embodiment; data is physically present in the world, in relations between matter. Measurements and observations can {\em discover} data, but not create it out of subjective whole cloth.
./Chapter4/c_4_data_as_facts.tex:156:At the same time, data has different natures according to the roles that need the data. \quotation{The geologist says I want the sample. The chemist says I want the analysis.} The participant notes that for other roles, data's substance can be process data or the discovery of relations within the set of process data.
./Chapter4/c_4_data_as_facts.tex:184:Data-as-facts is a narrower category, indicating that a piece of data is a scientific fact, used for the revelation of knowledge, discovering how the universe works by looking at the relationships and arrangements of things.
./Chapter4/c_4_analysis.tex:9:I am testing two questions of interest, which guide the course of my analysis: whether people have different realities of data and whether my methodologies can discover someone's personal construction of data. By guiding the abductive research process with these two questions of interest, I may be able to limit my conclusions to the most simple and credible story possible from the evidence.
./Chapter4/c_4_interview_conclusion.tex:10:\item I have discovered constructions of data that profoundly differ from each other and that agree with other, related research.
./Chapter4/c_4_interview_conclusion.tex:12:\item There is evidence for both Ackoff's and Tuomi's hierarchies of data, information, and knowledge. Ackoff's work has penetrated modern thinking, especially management thinking, quite deeply. At the same time, the more \quotation{modern} relativistic philosophies of science seem to correspond quite well with Tuomi's ontology. Far more evidence is needed to create suggested \quotation{trading zones} that can bridge these and other discovered conceptions of data.
./Chapter4/c_4_interview_conclusion.tex:14:\item Both the interviews and the surveys discovered different realities of data, but not without failures. More research is necessary to explore the utility of my methodology and to refine it for different purposes.
./Chapter4/c_4_raw.tex:798:Participant: Let’s go intuition. You might discover what I actually think along the way.
./Chapter4/c_4_raw.tex:2056:Participant: Information. Set relationship -- if we know a priori yes, {noun 74} is related to {Noun 12}, that’s Information. ??? But it may have to be something I have to discover. It depends where I -- where that relationship comes from. Sometimes I can just know where it comes from or I have to discover it through further manipulation.
./Chapter4/c_4_raw.tex:2062:Interviewer: The Information is that there is a relationship here? So, we have relevant observations and data sets. Some data sets have relational context internal to the set. Others have context external too the problem. Well, all have context external to the problem. Some have internal context based on prior Knowledge turned into Information. Given that we have some that don’t have an internal context that we have to discover, we could classify this as the {position 58}ing problem? Walk me through this.
./Chapter4/c_4_raw.tex:2068:Participant: This same problem? Ultimately, I’m interested in the {adjective 78} and distribution of that {noun 60}. There are a bunch of things which I immediately know are directly relevant. {adjective 79} rate, the {Noun 12}. ... I can do all of that -- any {position 58} should be able to do that. There is something about the nature of the {noun 84} that’s probably important. That’s probably going to have to do with: what {noun 97}es that produce the {noun 84}? What {noun 97}es that those raw materials were... It’s not -- I may suspect that they are related. I may not. There was an argument about that. I have to discover that relationship. In a nice case... the number of times when people don’t have any idea about an {position 58}ing problem. They go "right. Here’s the thing I’m trying to control. We currently have the {noun 85} table, there are 500 columns in the {noun 85} table, let’s join that with that table and let’s do bivariant plots to every one of them and get the r-squared. And we go through -- I’m not kidding. Ask [name] about {Noun 23} and {noun 28}. This is one of the only -- if the only tool you have is a hammer, everything looks like a thumb. It’s just like that. Can I do anything more sophisticated? Hopefully I can winnow out some of those 500 odd columns, because I know something. I have to be careful, because often -- the thing that’s driving this might be related to this. We don’t measure that. We measure some other things that are related to that. Hopefully I can get a weak relationship between them and the thing of interest. And maybe I’ll discover this thing one day. That’s sort of the {noun 97} where we’re trying to work out what drives this. And there are quite a few things we have that where we’ll find a subset of 5 things that can produce a decent linear model but allows us to describe, predict, what that’s going to do. And then time goes by and they don’t work so well anymore which tends to suggest that it was this thing down here that they were related to. And some other things that were also important are changing and we don’t really know what’s going on. That’s actually the key we’re trying to address: what research is trying to do and what the -- technology department solve today’s problems, solve next month’s problems. Research department, solve /the/ problem. Not the same thing. That’s the implication that we’re actually getting to the -- we’re aiming towards laws of nature. What is actually the real reason? As opposed to "How can I get a useful working model that allows us to make more money?" That’s more the technology {position 58}’s job. Which means we’re going to have overlap and conflict because we have different agendas.
./Chapter4/c_4_raw.tex:2813:Interviewer: From what you described, the refinement of theories and the research of theories is a research programme. Because you’re not really questioning your inner assumptions of “this is {Noun 23}, this is what it does. We know it, we’re not going to go: ‘hey, I’ve discovered a new property of {Noun 23}’ we’re going ‘hey, here’s this refinement of the outer shell.’
./Chapter3/c_3_conducting.tex:55:Participants, if not uncomfortable from the unusual thinking demands of the \SDFN, generally engaged in a self-reflective discussion. It was vital to ask open questions that build a scaffold for the participant's self-discovery.
./Chapter7/c_7_db.tex:111:Miss G would be most interested in the objective and discrete changes in product availability. These are based on objective business measurements that do not require interpretation. As such, the analysis of buying habits is something of an imponderable that is a useful pipe-dream but not discoverable from the \quotation{database} they have. Data, to her, is a purchase or a price.
./Chapter7/c_7_db.tex:164:Fixing errors in a production system is difficult as \quotation{the cost of fixing errors grows exponentially as a function of elapsed time to discovery.}\cite{Wand2002} Since these errors exist in a production system that has run for some time, the cost of fixing the errors is prohibitive at the stage when the errors are found. The only way to insure that the system model of the database designers keeps the system in line with reality is to inspect the database and insure that it corresponds with reality within the required specifications.
./c_conclusion.tex:5:This research aimed to understand how users perceive the nature of data. In a more limited sense, I wanted to demonstrate that users have different conceptions of data and a method for discovering hints as to how they use and understand data. In that light, I have been successful. This research should serve as the foundation to more specific and focused studies into the different social constructions of data present in our societies and what epistemological and ontological basis data may have.
./c_conclusion.tex:21:The Social Data Flow Network has great utility as an reality-discovery methodology. The \SDFN, inspired by the data flow diagram methodology of systems design, is quite able to cause participants to differentiate between practical definitions and theoretical definitions when discussing a participant’s construction of data. And is therefore better than more focused requirements gathering methodologies at this task.
./Chapter1/c_1_execSummary.tex:8:My research demonstrates that people have different constructions of data. The methodology of the \infull{SDFN}, created for this dissertation, has proven able to probe those understandings. The \infull{SDFN}, loosely based on a \infull{DFD} and combined with ideas from \infull{SNA}, provides a way of discovering practical definitions of hard-to-operationalize terms like {\em data}. The process of repeatedly categorizing various items as data allows the methodology to explore how participants actually use the term, rather than relying on theoretical dictionary-based definitions.
./Chapter1/c_1_methodologicalSummary.tex:5:The primary result of this thesis is the methodology of the Social Data Flow Network. The SDFN uses repeated categorization to explore how individuals group informational or communicative flows into categories. By eliciting categorizes that focus on data, information, and knowledge, the participants use the categorization to operationalize their epistemological understanding of data: they indicate what is and is not data and how it becomes information and knowledge. This elicitation helps both the interviewer and the participant to discover their own situational conceptualization of data.
./Chapter1/c_1_results.tex:42:My interview analysis discovered three different conceptions of data. It would be hard to deny that interviews I and II have data as communication, III and IV have subjective observations (with IX hinting at them) and the rest considering data as objective fact. With these broad differences evident, I feel question of interest 1 has been satisfied.
./Chapter1/c_1_results.tex:48:Data, in the communicative sense, merely requires signs and things to communicate with those signs. The data can be rendered as bits or marks on paper, but it is seen as a factor of semiotic import rather than as something to be discovered or filtered.
./Chapter2/c_2_hierarchiesOfData.tex:16:Ackoff was a management consultant and former professor of management science at the Wharton School specializing in operations research and organizational theory. His article formulating what is now commonly called the Data-Information- Knowledge-Wisdom hierarchy (or DIKW for short) was first given in 1988 as a presidential address to the International Society for General Systems Research. This background may help explain his approach. Data in his terms are the product of observations, and are of no value until they are processed into a usable form to become information. Information is contained in answers to questions. Knowledge, the next layer, further refines information by making \quotation{possible the transformation of information into instructions. It makes control of a system possible} (Ackoff, 1989, 4), and that enables one to make it work efficiently. A managerial rather than scholarly perspective runs through Ackoff's entire hierarchy, so that \quotation{understanding} for him connotes an ability to assess and correct for errors, while \quotation{wisdom} means an ability to see the long-term consequences of any act and evaluate them relative to the ideal of total control (omnicompetence). While a scholarly perspective on this hierarchy might prioritize the processes of inquiry and discovery, Ackoff does not account for them. But his concept of omnicompetence, which refers to \quotation{the ability to satisfy any and every desire} (Ackoff, 1989, 8), does encompass the satisfaction of user-defined needs.
./Chapter2/c_2_litReviewIntro.tex:7:Concept elicitation methodologies are a subset of knowledge elicitation methods, a tool used in many disciplines to \quotation{obtain the information needed to solve problems}\cite{burge1998}. Knowledge elicitation, in the main, is focused on direct problem solving: exploring requirements and understanding the meanings of those requirements. However, by turning the techniques of knowledge elicitation onto epistemological questions of category, we can discover not the direct meaning behind requirements, but some of a person’s semiotic models of the constructions behind those requirements.
./Chapter2/c_2_justification.tex:27:I also want to create a method that can help extend \HCI\ design practice. This methodology should be applicable to all sorts of design, as it is a tool for rendering clients' realities and not a specific kind of technical reality. The discovery of practical meaning of terms, ideas, and affordances\cite{Norman1999} of data is another tool with which HCI designers can understand how to render data presented in an interface. A tool that can make elements of private jargon explicit, and that is focused on that task (rather than treating it as a happy byproduct) can significantly contribute to the HCI design cycle.
./Chapter2/c_2_justification.tex:31:My basic discoveries, both methodological and philosophical, should have pragmatic results. I hope to create a methodology that improves communication and database design. I explore how we socially construct and use the term \quotation{data}. From this investigation, I can offer potential insights into how we create trading zones between different cultures of data use. While true understanding of the nature of data may be outside the scope of this present research, the construction of a foundation is not. Any methodology created must be robust enough to provide useful observations and a compelling story.
./Chapter2/c_2_linksWithLiterature.tex:23:My work profoundly agrees with the discoveries he made, though my research focuses far more on data and differentiates three different orders of data to his two.