Category Archives: measuring usability

DAAI V4 FRONTPAGE

eJournal: uffmm.org,
ISSN 2567-6458, 12.May – 18.Jan 2020
Email: info@uffmm.org
Author: Gerd Doeben-Henisch
Email: gerd@doeben-henisch.de

HISTORY OF THIS PAGE

See end of this page.

CONTEXT

This Theory of Engineering section is part of the uffmm science blog.

HISTORY OF THE (D)AAI-TEXT

See below

ACTUAL VERSION

DISTRIBUTED ACTOR ACTOR INTERACTION [DAAI]. Version 15.06, From  Dec 13, 2019 until Jan 18, 2020

aaicourse-15-06-07(PDF, Chapter 8 new (but not yet completed))

aaicourse-15-06-05(PDF, Chapter 7 new)

aaicourse-15-06-04(PDF, Chapter 6 modified)

aaicourse-15-06-03(PDF, Chapter 5 modified)

aaicourse-15-06-02(PDF, Chapter 4 modified)

aaicourse-15-06-01(PDF, Chapter 1 modified)

aaicourse-15-06 (PDF, chapters 1-6)

aaicourse-15-05-2 (PDF, chapters 1-6; chapter 6 only as a first stub)

DISTRIBUTED ACTOR ACTOR INTERACTION [DAAI]. Version 15.05.1, Dec 2, 2019:

aaicourse-15-05-1(PDF, chapters 1-5; minor corrections)

aaicourse-15-05 (PDF, chapters 1-5 of the new version 15.05)

Changes: Extension of title, extension of preface!, extension of chapter 4, new: chapter 5 MAS, extension of bibliography and indices.

HISTORY OF UPDATES

ACTOR ACTOR INTERACTION [AAI]. Version: June 17, 2019 – V.7: aaicourse-17june2019-incomplete

Change: June 19, 2019 (Update  to version 8; chapter 5 has been rewritten completely).

ACTOR ACTOR INTERACTION [AAI]. Version: June 19, 2019 – V.8: aaicourse-june 19-2019-v8-incomplete

Change: June 19, 2019 (Update to version 8.1; minor corrections in chapter 5)

ACTOR ACTOR INTERACTION [AAI]. Version: June 19, 2019 – V.8.1: aaicourse-june19-2019-v8.1-incomplete

Change: June 23, 2019 (Update to version 9; adding chapter 6 (Dynamic AS) and chapter 7 (Example of dynamic AS with two actors)

ACTOR ACTOR INTERACTION [AAI]. Version: June 23, 2019 – V.9: aaicourse-June-23-2019-V9-incomplete

Change: June 25, 2019 (Update to version 9.1; minor corrections in chapters 1+2)

ACTOR ACTOR INTERACTION [AAI]. Version: June 25, 2019 – V.9.1aaicourse-June25-2019-V9-1-incomplete

Change: June 29, 2019 (Update to version 10; )rewriting of chapter 4 Actor Story on account of changes in the chapters 5-7)

ACTOR ACTOR INTERACTION [AAI]. Version: June 29, 2019 – V.10: aaicourse-June-29-2019-V10-incomplete

Change: June 30, 2019 (Update to version 11; ) completing  chapter  3 Problem Definition)

ACTOR ACTOR INTERACTION [AAI]. Version: June 30, 2019 – V.11: aaicourse-June30-2019-V11-incomplete

Change: June 30, 2019 (Update to version 12; ) new chapter 5 for normative actor stories (NAS) Problem Definition)

ACTOR ACTOR INTERACTION [AAI]. Version: June 30, 2019 – V.12: aaicourse-June30-2019-V12-incomplete

Change: June 30, 2019 (Update to version 13; ) extending chapter 9 with the section about usability testing)

ACTOR ACTOR INTERACTION [AAI]. Version: June 30, 2019 – V.13aaicourse-June30-2019-V13-incomplete

Change: July 8, 2019 (Update to version 13.1 ) some more references to chapter 4; formatting the bibliography alphabetically)

ACTOR ACTOR INTERACTION [AAI]. Version: July 8, 2019 – V.13.1: aaicourse-July8-2019-V13.1-incomplete

Change: July 15, 2019 (Update to version 13.3 ) (In chapter 9 Testing an AS extending the description of Usability Testing with more concrete details to the test procedure)

ACTOR ACTOR INTERACTION [AAI]. Version: July 15, 2019 – V.13.3: aaicourse-13-3

Change: Aug 7, 2019 (Only some minor changes in Chapt. 1 Introduction, pp.15ff, but these changes make clear, that the scope of the AAI analysis can go far beyond the normal analysis. An AAI analysis without explicit actor models (AMs) corresponds to the analysis phase of a systems engineering process (SEP), but an AAI analysis including explicit actor models will cover 50 – 90% of the (logical) design phase too. How much exactly could only be answered if  there would exist an elaborated formal SEP theory with quantifications, but there exists  no such theory. The quantification here is an estimate.)

ACTOR ACTOR INTERACTION [AAI]. Version: Aug 7, 2019 – V.14:aaicourse-14

ACTOR ACTOR INTERACTION [AAI]. Version 15, Nov 9, 2019:

aaicourse-15(PDF, 1st chapter of the new version 15)

ACTOR ACTOR INTERACTION [AAI]. Version 15.01, Nov 11, 2019:

aaicourse-15-01 (PDF, 1st chapter of the new version 15.01)

ACTOR ACTOR INTERACTION [AAI]. Version 15.02, Nov 11, 2019:

aaicourse-15-02 (PDF, 1st chapter of the new version 15.02)

ACTOR ACTOR INTERACTION [AAI]. Version 15.03, Nov 13, 2019:

aaicourse-15-03 (PDF, 1st chapter of the new version 15.03)

ACTOR ACTOR INTERACTION [AAI]. Version 15.04, Nov 19, 2019:

(update of chapter 3, new created chapter 4)

aaicourse-15-04 (PDF, chapters 1-4 of the new version 15.04)

HISTORY OF CHANGES OF THIS PAGE

Change: May 20, 2019 (Stopping Circulating Acronyms :-))

Change: May 21,  2019 (Adding the Slavery-Empowerment topic)

Change: May 26, 2019 (Improving the general introduction of this first page)

HISTORY OF AAI-TEXT

After a previous post of the new AAI approach I started the first  re-formulation of the general framework of  the AAI theory, which later has been replaced by a more advanced AAI version V2. But even this version became a change candidate and mutated to the   Actor-Cognition Interaction (ACI) paradigm, which still was not the endpoint. Then new arguments grew up to talk rather from the Augmented Collective Intelligence (ACI). Because even this view on the subject can  change again I stopped following the different aspects of the general Actor-Actor Interaction paradigm and decided to keep the general AAI paradigm as the main attractor capable of several more specialized readings.

ENGINEERING AND SOCIETY: The Role of Preferences

eJournal: uffmm.org,
ISSN 2567-6458, 4.May 2019
Email: info@uffmm.org
Author: Gerd Doeben-Henisch
Email: gerd@doeben-henisch.de

FINAL HYPOTHESIS

This suggests that a symbiosis between creative humans and computing algorithms is an attractive pairing. For this we have to re-invent our official  learning processes in schools and universities to train the next generation of humans in a more inspired and creative usage of algorithms in a game-like learning processes.

CONTEXT

The overall context is given by the description of the Actor-Actor Interaction (AAI) paradigm as a whole.  In this text the special relationship between engineering and the surrounding society is in the focus. And within this very broad and rich relationship the main interest lies in the ethical dimension here understood as those preferences of a society which are more supported than others. It is assumed that such preferences manifesting themselves  in real actions within a space of many other options are pointing to hidden values which guide the decisions of the members of a society. Thus values are hypothetical constructs based on observable actions within a cognitively assumed space of possible alternatives. These cognitively represented possibilities are usually only given in a mixture of explicitly stated symbolic statements and different unconscious factors which are influencing the decisions which are causing the observable actions.

These assumptions represent  until today not a common opinion and are not condensed in some theoretical text. Nevertheless I am using these assumptions here because they help to shed some light on the rather complex process of finding a real solution to a stated problem which is rooted in the cognitive space of the participants of the engineering process. To work with these assumptions in concrete development processes can support a further clarification of all these concepts.

ENGINEERING AND SOCIETY

DUAL: REAL AND COGNITIVE

The relationship between an engineering process and the preferences of a society
The relationship between an engineering process and the preferences of a society

As assumed in the AAI paradigm the engineering process is that process which connects the  event of  stating a problem combined with a first vision of a solution with a final concrete working solution.

The main characteristic of such an engineering process is the dual character of a continuous interaction between the cognitive space of all participants of the process with real world objects, actions, and processes. The real world as such is a lose collection of real things, to some extend connected by regularities inherent in natural things, but the visions of possible states, possible different connections, possible new processes is bound to the cognitive space of biological actors, especially to humans as exemplars of the homo sapiens species.

Thus it is a major factor of training, learning, and education in general to see how the real world can be mapped into some cognitive structures, how the cognitive structures can be transformed by cognitive operations into new structures and how these new cognitive structures can be re-mapped into the real world of bodies.

Within the cognitive dimension exists nearly infinite sets of possible alternatives, which all indicate possible states of a world, whose feasibility is more or less convincing. Limited by time and resources it is usually not possible to explore all these cognitively tapped spaces whether and how they work, what are possible side effects etc.

PREFERENCES

Somehow by nature, somehow by past experience biological system — like the home sapiens — have developed   cultural procedures to induce preferences how one selects possible options, which one should be selected, under which circumstances and with even more constraints. In some situations these preferences can be helpful, in others they can  hide possibilities which afterwards can be  re-detected as being very valuable.

Thus every engineering process which starts  a transformation process from some cognitively given point of view to a new cognitively point of view with a following up translation into some real thing is sharing its cognitive space with possible preferences of  the cognitive space of the surrounding society.

It is an open case whether the engineers as the experts have an experimental, creative attitude to explore without dogmatic constraints the   possible cognitive spaces to find new solutions which can improve life or not. If one assumes that there exist no absolute preferences on account of the substantially limit knowledge of mankind at every point of time and inferring from this fact the necessity to extend an actual knowledge further to enable the mastering of an open unknown future then the engineers will try to explore seriously all possibilities without constraints to extend the power of engineering deeper into the heart of the known as well as unknown universe.

EXPLORING COGNITIVE POSSIBILITIES

At the start one has only a rough description of the problem and a rough vision of a wanted solution which gives some direction for the search of an optimal solution. This direction represents also a kind of a preference what is wanted as the outcome of the process.

On account of the inherent duality of human thinking and communication embracing the cognitive space as well as the realm of real things which both are connected by complex mappings realized by the brain which operates  nearly completely unconscious a long process of concrete real and cognitive actions is necessary to materialize cognitive realities within a  communication process. Main modes of materialization are the usage of symbolic languages, paintings (diagrams), physical models, algorithms for computation and simulations, and especially gaming (in several different modes).

As everybody can know  these communication processes are not simple, can be a source of  confusions, and the coordination of different brains with different cognitive spaces as well as different layouts of unconscious factors  is a difficult and very demanding endeavor.

The communication mode gaming is of a special interest here  because it is one of the oldest and most natural modes to learn but in the official education processes in schools and  universities (and in companies) it was until recently not part of the official curricula. But it is the only mode where one can exercise the dimensions of preferences explicitly in combination with an exploring process and — if one wants — with the explicit social dimension of having more than one brain involved.

In the last about 50 – 100 years the term project has gained more and more acceptance and indeed the organization of projects resembles a game but it is usually handled as a hierarchical, constraints-driven process where creativity and concurrent developing (= gaming) is not a main topic. Even if companies allow concurrent development teams these teams are cognitively separated and the implicit cognitive structures are black boxes which can not be evaluated as such.

In the presupposed AAI paradigm here the open creative space has a high priority to increase the chance for innovation. Innovation is the most valuable property in face of an unknown future!

While the open space for a real creativity has to be executed in all the mentioned modes of communication the final gaming mode is of special importance.  To enable a gaming process one has explicitly to define explicit win-lose states. This  objectifies values/ preferences hidden   in the cognitive space before. Such an  objectification makes things transparent, enables more rationality and allows the explicit testing of these defined win-lose states as feasible or not. Only tested hypothesis represent tested empirical knowledge. And because in a gaming mode whole groups or even all members of a social network can participate in a  learning process of the functioning and possible outcome of a presented solution everybody can be included.  This implies a common sharing of experience and knowledge which simplifies the communication and therefore the coordination of the different brains with their unconsciousness a lot.

TESTING AND EVALUATION

Testing a proposed solution is another expression for measuring the solution. Measuring is understood here as a basic comparison between the target to be measured (here the proposed solution) and the before agreed norm which shall be used as point of reference for the comparison.

But what can be a before agreed norm?

Some aspects can be mentioned here:

  1. First of all there is the proposed solution as such, which is here a proposal for a possible assistive actor in an assumed environment for some intended executive actors which has to fulfill some job (task).
  2. Part of this proposed solution are given constraints and non-functional requirements.
  3. Part of this proposed solution are some preferences as win-lose states which have to be reached.
  4. Another difficult to define factor are the executive actors if they are biological systems. Biological systems with their basic built in ability to act free, to be learning systems, and this associated with a not-definable large unconscious realm.

Given the explicit preferences constrained by many assumptions one can test only, whether the invited test persons understood as possible instances of the  intended executive actors are able to fulfill the defined task(s) in some predefined amount of time within an allowed threshold of making errors with an expected percentage of solved sub-tasks together with a sufficient subjective satisfaction with the whole process.

But because biological executive actors are learning systems they  will behave in different repeated  tests differently, they can furthermore change their motivations and   their interests, they can change their emotional commitment, and because of their   built-in basic freedom to act there can be no 100% probability that they will act at time t as they have acted all the time before.

Thus for all kinds of jobs where the process is more or less fixed, where nothing new  will happen, the participation of biological executive actors in such a process is questionable. It seems (hypothesis), that biological executing actors are better placed  in jobs where there is some minimal rate of curiosity, of innovation, and of creativity combined with learning.

If this hypothesis is empirically sound (as it seems), then all jobs where human persons are involved should have more the character of games then something else.

It is an interesting side note that the actual research in robotics under the label of developmental robotics is struck by the problem how one can make robots continuously learning following interesting preferences. Given a preference an algorithm can work — under certain circumstances — often better than a human person to find an optimal solution, but lacking such a preference the algorithm is lost. And actually there exists not the faintest idea how algorithms should acquire that kind of preferences which are interesting and important for an unknown future.

On the contrary, humans are known to be creative, innovative, detecting new preferences etc. but they have only limited capacities to explore these creative findings until some telling endpoint.

This suggests that a symbiosis between creative humans and computing algorithms is an attractive pairing. For this we have to re-invent our official  learning processes in schools and universities to train the next generation of humans in a more inspired and creative usage of algorithms in a game-like learning processes.

 

 

 

 

ACTOR-ACTOR INTERACTION ANALYSIS – A rough Outline of the Blueprint

eJournal: uffmm.org,
ISSN 2567-6458, 13.February 2019
Email: info@uffmm.org
Author: Gerd Doeben-Henisch
Email: gerd@doeben-henisch.de

Last corrections: 14.February 2019 (add some more keywords; added  emphasizes for central words)

Change: 5.May 2019 (adding the the aspect of simulation and gaming; extending the view of the driving actors)

CONTEXT

An overview to the enhanced AAI theory  version 2 you can find here.  In this post we talk about the blueprint  of the whole  AAI analysis process. Here I leave out the topic of actor models (AM); the aspect of  simulation and gaming is mentioned only shortly. For these topics see other posts.

THE AAI ANALYSIS BLUEPRINT

Blueprint of the whole AAI analysis process including the epistemological assumptions. Not shown here is the whole topic of actor models (AM) and as well simulation.
Blueprint of the whole AAI analysis process including the epistemological assumptions. Not shown here is the whole topic of actor models (AM) and as well simulation.

The Actor-Actor Interaction (AAI) analysis is understood here as part of an  embracing  systems engineering process (SEP), which starts with the statement of a problem (P) which includes a vision (V) of an improved alternative situation. It has then to be analyzed how such a new improved situation S+ looks like; how one can realize certain tasks (T)  in an improved way.

DRIVING ACTORS

The driving actors for such an AAI analysis are at least one  stakeholder (STH) which communicates a problem P and an envisioned solution (ES) to an  expert (EXPaai) with a sufficient AAI experience. This expert will take   the lead in the process of transforming the problem and the envisioned  solution into a working solution (WS).

In the classical industrial case the stakeholder can be a group of managers from some company and the expert is also represented by a whole team of experts from different disciplines, including the AAI perspective as leading perspective.

In another case which  I will call here the  communal case — e.g. a whole city —      the stakeholder as well as the experts are members of the communal entity.   As   in the before mentioned cases there is some commonly accepted problem P combined  with a first envisioned solution ES, which shall be analyzed: what is needed to make it working? Can it work at all? What are costs? And many other questions can arise. The challenge to include all relevant experience and knowledge from all participants is at the center of the communication and to transform this available knowledge into some working solution which satisfies all stated requirements for all participants is a central  condition for the success of the project.

EPISTEMOLOGY

It has to be taken into account that the driving actors are able to do this job because they  have in their bodies brains (BRs) which in turn include  some consciousness (CNS). The processes and states beyond the consciousness are here called ‘unconscious‘ and the set of all these unconscious processes is called ‘the Unconsciousness’ (UCNS).

For more details to the cognitive processes see the post to the philosophical framework as well as the post bottom-up process. Both posts shall be integrated into one coherent view in the future.

SEMIOTIC SUBSYSTEM

An important set of substructures of the unconsciousness are those which enable symbolic language systems with so-called expressions (L) on one side and so-called non-expressions (~L) on the other. Embedded in a meaning relation (MNR) does the set of non-expressions ~L  function as the meaning (MEAN) of the expressions L, written as a mapping MNR: L <—> ~L. Depending from the involved sensors the expressions L can occur either as acoustic events L_spk, or as visual patterns written L_txt or visual patterns as pictures L_pict or even in other formats, which will not discussed here. The non-expressions can occur in every format which the brain can handle.

While written (symbolic) expressions L are only associated with the intended meaning through encoded mappings in the brain,  the spoken expressions L_spk as well as the pictorial ones L_pict can show some similarities with the intended meaning. Within acoustic  expressions one can ‘imitate‘ some sounds which are part of a meaning; even more can the pictorial expressions ‘imitate‘ the visual experience of the intended meaning to a high degree, but clearly not every kind of meaning.

DEFINING THE MAIN POINT OF REFERENCE

Because the space of possible problems and visions it nearly infinite large one has to define for a certain process the problem of the actual process together with the vision of a ‘better state of the affairs’. This is realized by a description of he problem in a problem document D_p as well as in a vision statement D_v. Because usually a vision is not without a given context one has to add all the constraints (C) which have to be taken into account for the possible solution.  Examples of constraints are ‘non-functional requirements’ (NFRs) like “safety” or “real time” or “without barriers” (for handicapped people). Part of the non-functional requirements are also definitions of win-lose states as part of a game.

AAI ANALYSIS – BASIC PROCEDURE

If the AAI check has been successful and there is at least one task T to be done in an assumed environment ENV and there are at least one executing actor A_exec in this task as well as an assisting actor A_ass then the AAI analysis can start.

ACTOR STORY (AS)

The main task is to elaborate a complete description of a process which includes a start state S* and a goal state S+, where  the participating executive actors A_exec can reach the goal state S+ by doing some actions. While the imagined process p_v  is a virtual (= cognitive/ mental) model of an intended real process p_e, this intended virtual model p_e can only be communicated by a symbolic expressions L embedded in a meaning relation. Thus the elaboration/ construction of the intended process will be realized by using appropriate expressions L embedded in a meaning relation. This can be understood as a basic mapping of sensor based perceptions of the supposed real world into some abstract virtual structures automatically (unconsciously) computed by the brain. A special kind of this mapping is the case of measurement.

In this text especially three types of symbolic expressions L will be used: (i) pictorial expressions L_pict, (ii) textual expressions of a natural language L_txt, and (iii) textual expressions of a mathematical language L_math. The meaning part of these symbolic expressions as well as the expressions itself will be called here an actor story (AS) with the different modes  pictorial AS (PAS), textual AS (TAS), as well as mathematical AS (MAS).

The basic elements of an  actor story (AS) are states which represent sets of facts. A fact is an expression of some defined language L which can be decided as being true in a real situation or not (the past and the future are special cases for such truth clarifications). Facts can be identified as actors which can act by their own. The transformation from one state to a follow up state has to be described with sets of change rules. The combination of states and change rules defines mathematically a directed graph (G).

Based on such a graph it is possible to derive an automaton (A) which can be used as a simulator. A simulator allows simulations. A concrete simulation takes a start state S0 as the actual state S* and computes with the aid of the change rules one follow up state S1. This follow up state becomes then the new actual state S*. Thus the simulation constitutes a continuous process which generally can be infinite. To make the simulation finite one has to define some stop criteria (C*). A simulation can be passive without any interruption or interactive. The interactive mode allows different external actors to select certain real values for the available variables of the actual state.

If in the problem definition certain win-lose states have been defined then one can turn an interactive simulation into a game where the external actors can try to manipulate the process in a way as to reach one of the defined win-states. As soon as someone (which can be a team) has reached a win-state the responsible actor (or team) has won. Such games can be repeated to allow accumulation of wins (or loses).

Gaming allows a far better experience of the advantages or disadvantages of some actor story as a rather lose simulation. Therefore the probability to detect aspects of an actor story with their given constraints is by gaming quite high and increases the probability to improve the whole concept.

Based on an actor story with a simulator it is possible to increase the cognitive power of exploring the future even more.  There exists the possibility to define an oracle algorithm as well as different kinds of intelligent algorithms to support the human actor further. This has to be described in other posts.

TAR AND AAR

If the actor story is completed (in a certain version v_i) then one can extract from the story the input-output profiles of every participating actor. This list represents the task-induced actor requirements (TAR).  If one is looking for concrete real persons for doing the job of an executing actor the TAR can be used as a benchmark for assessing candidates for this job. The profiles of the real persons are called here actor-actor induced requirements (AAR), that is the real profile compared with the ideal profile of the TAR. If the ‘distance’ between AAR and TAR is below some threshold then the candidate has either to be rejected or one can offer some training to improve his AAR; the other option is to  change the conditions of the TAR in a way that the TAR is more closer to the AARs.

The TAR is valid for the executive actors as well as for the assisting actors A_ass.

CONSTRAINTS CHECK

If the actor story has in some version V_i a certain completion one has to check whether the different constraints which accompany the vision document are satisfied through the story: AS_vi |- C.

Such an evaluation is only possible if the constraints can be interpreted with regard to the actor story AS in version vi in a way, that the constraints can be decided.

For many constraints it can happen that the constraints can not or not completely be decided on the level of the actor story but only in a later phase of the systems engineering process, when the actor story will be implemented in software and hardware.

MEASURING OF USABILITY

Using the actor story as a benchmark one can test the quality of the usability of the whole process by doing usability tests.

 

 

 

 

 

 

 

 

 

 

 

AAI THEORY V2 – MEASURING USABILITY

eJournal: uffmm.org
ISSN 2567-6458, 6.February 2019
Email: info@uffmm.org
Author: Gerd Doeben-Henisch
Email: gerd@doeben-henisch.de

CONTEXT

An overview of the enhanced AAI theory  version 2 you can find here.  In this post we talk about the tenth chapter dealing with Measuring Usability

MEASURING  USABILITY

As has been delineated in the post “Usability and Usefulness”   statements  about the quality of the usability of some assisting actor are based on some  kinds of measurement: mapping some target (here the interactions of an executive actor with some assistive actor) into some predefined norm (e.g. ‘number of errors’, ‘time needed for completion’, …).   These remarks are here embedded in a larger perspective following   Dumas and  Fox (2008).

Overview of Usability Testing following the article of Dumas & Fox (2008), with some new AAI specific terminology
Overview of Usability Testing following the article of Dumas & Fox (2008), with some new AAI specific terminology

From the three main types of usability testing with regard to the position in the life-cycle of a system we focus here primarily on the usability testing as part of the analysis phase where the developers want to get direct feedback for the concepts embedded in an actor story. Depending from this feedback the actor story and its related models can become modified and this can result in a modified exploratory mock-up  for a new test. The challenge is not to be ‘complete’ in finding ‘disturbing’ factors during an interaction but to increase the probability to detect possible disturbing factors by facing the symbolically represented concepts of the actor story with a sample of real world actors. Experiments  point to the number of 5-10 test persons which seem to be sufficient to detect the most severe disturbing factors of the concepts.

Usability testing procedure according to Lauesen (2005), adapted to the AAI paradigm
Usability testing procedure according to Lauesen (2005), adapted to the AAI paradigm

A good description of usability testing can be found in the book Lauesen (2005), especially chapters 1 +13.  According to this one can infer the following basic schema for a usability test:

  1. One needs 5 – 10 test persons whose input-output profile (AAR) comes close to the profile (TAR) required by the actor story.
  2. One needs a  mock-up of the assistive actor; this mock-up  should  correspond ‘sufficiently well’ with the input-output profile (TAR) required by the  actor story. In the simplest case one has a ‘paper model’, whose sheets can be changed on demand.
  3. One needs a facilitator who is receiving the test person, introduces the test person into the task (orally and/ or by a short document (less than a page)), then accompanies the test without interacting further with the test person until the end of the test.  The end is either reached by completing the task or by reaching the end of a predefined duration time.
  4. After the test person has finished the test   a debriefing happens by interrogating the test person about his/ her subjective feelings about the test. Because interviews are always very fuzzy and not very reliable one should keep this interrogation simple, short, and associated with concrete points. One strategy could be to ask the test person first about the general feeling: Was it ‘very good’, ‘good’, ‘OK’, ‘undefined’, ‘not OK’, ‘bad’, ‘very bad’ (+3 … 0 … -3). If the stated feeling is announced then one can ask back which kinds of circumstances caused these feelings.
  5. During the test at least two observers are observing the behavior of the test person. The observer are using as their ‘norm’ the actor story which tells what ‘should happen in the ideal case’. If a test person is deviating from the actor story this will be noted as a ‘deviation of kind X’, and this counts as an error. Because an actor story in the mathematical format represents a graph it is simple to quantify the behavior of the test person with regard to how many nodes of a solution path have been positively passed. This gives a count for the percentage of how much has been done. Thus the observer can deliver data about at least the ‘percentage of task completion’, ‘the number (and kind) of errors by deviations’, and ‘the processing time’. The advantage of having the actor story as a  norm is that all observers will use the same ‘observation categories’.
  6. From the debriefing one gets data about the ‘good/ bad’ feelings on a scale, and some hints what could have caused the reported feelings.

STANDARDS – CIF (Common Industry Format)

There are many standards around describing different aspects of usability testing. Although standards can help in practice  from the point of research standards are not only good, they can hinder creative alternative approaches. Nevertheless I myself are looking to standards to check for some possible ‘references’.  One standard I am using very often is the  “Common Industry Format (CIF)”  for usability reporting. It is  an ISO standard (ISO/IEC 25062:2006) since  2006. CIF describes a method for reporting the findings of usability tests that collect quantitative measurements of user performance. CIF does not describe how to carry out a usability test, but it does require that the test include measurements of the application’s effectiveness and efficiency as well as a measure of the users’ satisfaction. These are the three elements that define the concept of usability.

Applied to the AAI paradigm these terms are fitting well.

Effectiveness in CIF  is targeting  the accuracy and completeness with which users achieve their goal. Because the actor story in AAI his represented as a graph where the individual paths represents a way to approach a defined goal one can measure directly the accuracy by comparing the ‘observed path’ in a test and the ‘intended ideal path’ in the actor story. In the same way one can compute the completeness by comparing the observed path and the intended ideal path of the actor story.

Efficiency in CIF covers the resources expended to achieve the goals. A simple and direct measure is the measuring of the time needed.

Users’ satisfaction in CIF means ‘freedom from discomfort’ and ‘positive attitudes towards the use of the product‘. These are ‘subjective feelings’ which cannot directly be observed. Only ‘indirect’ measures are possible based on interrogations (or interactions with certain tasks) which inherently are fuzzy and not very reliable.  One possibility how to interrogate is mentioned above.

Because the term usability in CIF is defined by the before mentioned terms of effectiveness, efficiency as well as  users’ satisfaction, which in turn can be measured in many different ways the meaning of ‘usability’ is still a bit vague.

DYNAMIC ACTORS – CHANGING CONTEXTS

With regard to the AAI paradigm one has further to mention that the possibility of adaptive, learning systems embedded in dynamic, changing  environments requires for a new type of usability testing. Because learning actors change by every exercise one should run a test several times to observe how the dynamic learning rates of an actor are developing in time. In such a dynamic framework  a system would only be  ‘badly usable‘ when the learning curves of the actors can not approach a certain threshold after a defined ‘typical learning time’. And,  moreover, there could be additional effects occurring only in a long-term usage and observation, which can not be measured in a single test.

REFERENCES

  • ISO/IEC 25062:2006(E)
  • Joseph S. Dumas and Jean E. Fox. Usability testing: Current practice
    and future directions. chapter 57, pp.1129 – 1149,  in J.A. Jacko and A. Sears, editors, The Human-Computer Interaction Handbook. Fundamentals, Evolving Technologies, and Emerging Applications. 2nd edition, 2008
  • S. Lauesen. User Interface Design. A software Engineering Perspective.
    Pearson – Addison Wesley, London et al., 2005