This book sounds rather interesting and has some great photography as well as nicely designed pages. The contributors to the book is also quite impressive, but it is impossible to tell if its content is good. Has anyone checked it out?

Business Model Generation: A Handbook for Visionaries, Game Changers, and Challengers

Business Model Generation is a practical, inspiring handbook for anyone striving to improve a business model or craft a new one. 1) Change the way you think about business models Business Model Generation will teach you powerful and practical innovation techniques used today by leading companies worldwide. You will learn how to systematically understand, design, and implement a new business model or analyze and renovate an old one. 2) Co-created by 470 strategy practitioners Business Model Generation practices what it preaches. Co-authored by 470 Business Model Canvas practitioners from 45 countries, the book was financed and produced independently of the traditional publishing industry. It features a tightly-integrated, visual, lie-flat design that enables immediate hands-on use. 3) Designed for doers Business Model Generation is for those ready to abandon outmoded thinking and embrace new, innovative models of value creation: executives, consultants, entrepreneurs and leaders of all organizations.

http://www.businessmodelgeneration.com/

Yesterday, a friend, who is also doing her PhD related to diagrams (University of Granada), sent me an email asking me that question. She also sent me a kind of research she has been done in Google images. She introduced the word ‘diagram’ in English, Spanish and German in Google images and the results were quite interesting. See below some of the images.

Spanish results of the term 'diagrama' (Paloma Lopez)

English results of the term 'diagram' (Paloma Lopez)

German results of the term 'diagramm' (Paloma Lopez)

Analysing the images, it can be seen that in Spanish the term ‘diagram’ seems to be denoting more complex structures, while most of the English images are linear and simpler layouts. In German, images related to the term ‘diagram’ are basic diagrams, such as pie charts, bar charts, flow charts.

Undoubtedly, what is common in the three languages is that a diagram is composed by a visual (graphic or typographical) structure that employs synthetic and schematic language to represent connections between components, quantities, directionality, functions. In terms of diagrams layout, there is a long list of parameters that can define diagrams structure, for instance the kind of information represented, the media where the diagram is going to be, the purpose of the diagram, its audience, among others.

A 25-minute documentary of the London Underground Diagrammatic Map creation (BBC Four)

Watch the video!

This is the title of Dieter Rams’ exhibition at the Design Museum. Rams was in charged of the German consumer electronics manufacturer, Braun, for more than 40 years, developing a synthetic, rigorous and pure visual language for its products. In this exhibition can be appreciated how his principles of design were applied in and followed by his designs.
Below are explained the 10 design principles defined by Dieter Rams:

Good design is innovative.
Good design makes a product useful.
Good design is aesthetic.
Good design makes a product understandable.
Good design is unobtrusive.
Good design is honest.
Good design is long-lasting.
Good design is thorough down to the last detail.
Good design is environmentally friendly.
Good design is as little design as possible.

As a starting point of his designs, Rams applied a clear and well-defined structure (grid?) to organise the elements (information?) before moving on to the creation stage itself.

Design Museum. Dieter Rams' structure used in the creation of his designs to organise elements. (©Photos Sheila Pontis)

Rams’ objective was to create extremely functional objects where irrelevant information had no place. His designs are examples of simplicity and clarity. (see below). Rams also thought that sketching (analysing?) was an essential part of the design process. In addition, for Rams this stage of the process is independently of the media or technology, and cannot be done with computers. This stage is about taking the necessary time to carefully think about what to do, the characteristics and uses of the product/project.

Design Museum. Wall diagram of Dieter Rams' object. (© Photo Sheila Pontis)

Less and More. The Design Ethos of Dieter Rams. 18 November 2009 to 7 March 2010 – Design Museum London

Given that neither vehicles I eventually went with – a 1999 Mercedes-Benz Vito 108D and a 2001 Mercedes-Benz Sprinter 311CDI LWB – were current models, it was difficult to source manufacturer drawings and dimensions on which to base the plans. So, an ‘on-the-job’ approach should be taken, and many of these drawings are retrospective or mere scribbles. It is advisable to find a way of getting access to a vehicle first – before you buy it or definitely start-building in it – and take measurements if you can’t source them elsewhere. If you have an idea in your head, though, it’s hard to stop yourself from just ploughing in to start.

In planning phase-2 developments for the first van, some photos were taken and made into outline drawings – on which ideas could be scribbled out. The icon to the left links to drawings, the far right are ablank setfor the Vito (the latter are also here as a powerpointfile).

Due to model variations, measurements taken are precise to specific vehicle, but are listed here for reference.

If you’re going to fit a water systemand power system, you also need a lot of planning – not just in terms of what you need, but where it will all fit in respect to your cupboards, etc. The following is a useful article on considerations you should keep in mind during the planning stage. You’ll also find a lot of helpful assistance in the SBMCC forums and in ‘the’ book.

Last Wednesday I attended to one of the RNUAL lectures (UAL) about how to approach writing up problems (essays, papers, chapters, thesis). As a first general conclusion, to approach this problem, the best strategy is to have a clear framework –a plan? – as a starting point. This means, to know exactly what to write and discuss, before start writing and writing without having a define direction. For this, the selection and organisation of raw material is essential.
One of the methods introduced during the lecture was to organise the material/information in columns.

3-column-structure method. This method groups similar information in each column and enables to see new connections between the data. Black lines represent clear data connections, while dot lines represent new data connections, seen as a result of visualising raw material.

I called this method 3-column structure as it classifies raw material in three interconnected columns in a quite useful and clear manner.

So, a possible starting point can be the definition of the main themes (of the chapter/of a paper/of an essay). These are the themes that are key for a subject and that can be used to structure a piece of writing. That is to say, the most relevant issues that will be discussed and explained.

After this, as a second stage, it is important to select the authors related to those themes. This means, who has said/written something significant about that themes. Besides the authors’ names, quotations can also be included into this category. Ideally, each theme should match with one or more authors from the literature and practice review.

A third and last stage would be focussed on defining the specific things of the previous selected themes I would like to write about. For things are understood the individual elements that composed a theme, such as key terms, projects or interviews.

Information/raw material can also be classified following a different stages order. For example, starting with the selection of things and leaving the themes for the end. Or by identifying key authors and then matching them with their main themes and terms. The order of the stages depends on the subject being analysed and the kind of text that has to be written.

Different approaches that the 3-column-structure method can be applied. The stages order depends on the kind of material and the purpose of the task.

In addition, it is important to notice that it was also highlighted during the lecture the importance of using diagrams and mind maps as a way of organising information. This is one of the reasons that this 3-column structure method is useful: it visualises the collected material, emphasising and ordering the important points. Moreover, these diagrammatic structures also show how different parts or components are related and enable to see new points of view that with a traditional organisation (or without it!) are not seen so clear.
Once again, this method shows how powerful tools diagrams are for organising and visualising complex information.

Following this (or any other) way of organising information, as a result ideas and thoughts will be also organised, and the task of writing up (or any other!) would be easier, as there is a previous plan.

I never realised that film manufacturer websites could actually be useful.

At http://www.panavision.com/tools.php you’ll find a collection of charts, tables, diagrams and calculators.. for depth of field (takes into account your sensor size/film gauge and everything), equivalent focal lengths, f-stop fractions, and HMI filming speeds at both 50 and 60Hz. It’s a good resource, and I was quite surprised to find so much free information on a website that’s meant to be flogging panavision gear. So.. congrats to panavision on providing me with more than XCU camera-porn and propaganda (a la RED’s website)

Matt Davey recently wrote about studies of software development diagrams.

In my experience, a pencil and paper design is better than a Visio diagram because it can be created, edited, and iterated much more quickly and the pencil doesn’t get in the way. Rarely do I fight my real life ruler to make straight lines, but I regularly curse Visio.

There is no need for the overwhelming number of Visio software and systems diagrams.  The diagrams would be 10 times as good with 1/10th the effort if the pencil (with liberal allowance for erasers) were used instead.  These diagrams can be easily scanned and kept forever with documentation.

The real issue is that the managers, the people who have no interest in the functionality of the diagrams themselves, insist on diagrams made with tools.  So Visio diagrams get created, they are rarely if ever updated and they take much more time to create with no extra benefit.  Thank you PHB.

Creately Flow Chart Template

Need to create a flow chart or Venn diagram for one of your class projects? Try the free web application, Creately.

There are numerous templates for things as varied as SWOT analysis, concept maps, mind maps, organizational charts, web flow charts, and online sitemaps.

If you go to their Education Graphic Organizer Templates page, you’ll find templates for story boards, cycle diagrams, T-charts for listing pros and cons, and even a comparison/contrast essay template.

Completing a group project? No problem. Creately allows you to add email addresses as invites, and then you can share, publish, and post comments on draft diagrams as a group.

Creately Story Board Template

Creately Comparison Contrast Essay Template

Hans Hoffman, The Gate, 1959-60

I am interrupting the sleep mode of the playground to publish a selfish and self-centered call for papers. For the 14th conference of the European Society for the History of Economic Thought in Amsterdam, I am submitting a session on the use of visual representation in economics, with the following blurb:

The last two decades have witnessed a growing literature on visualization in the history of science following the publication of Lynch and Woolgar’s Representation in Scientific Practice (1990) – see for instance a recent focus section in Isis (March 2006). Despite previous attempts to draw the attention of historians of economics and insightful published papers on the subject – e.g. a ECHE conference in 2002 and a related mini-symposium in JHET in 2003), the use of visual representation in economics remains largely misunderstood. Graphical methods, for instance, are still regarded as a mere subdivision of mathematical analysis, whereas Klein (1995), Cook (2005) and Giraud (2007) have demonstrated that they have been considered distinct from mathematics since the early days of neoclassical economics. More generally, though anyone would concede that graphs, charts, tables, pictures and illustrations are part of the economist’s workaday tools, few efforts have been engaged to understand precisely how they operate within the larger models and theoretical frameworks in which they are used. Failure to recognize the role of visualization in economics is related to the fact that historians of the field tend to focus on the development of theory rather than on the practices in which theorization is entrenched, favoring a foundational approach which undermines cultural specificities. The most recent contributions to the history of science, indeed, have pointed out that the role of visualization in science is best understood within the framework of visual culture – see for instance Luc Pauwels (ed.), Visual Cultures of Science (2006).In this session, we would like to follow this literature by bringing together a set of papers which explore the use of visual representation in connection with peculiar cultures, whether disciplinary or operating at a larger level – the birth of mass-media in the US, for instance. Contributions will focus on the invention of visual devices in relation with specific practices, on the interaction between economists and artists or on how certain visual methods are affected when audiences are different from those they were originally intended for. They need not be focused on theoretical economics but also on the use of visual representation by economic propagandists, state administrations or practitioners operating on markets

I already have two papers for the session, including one by Loic and myself on the visual display of economic information in the US during the interwar period (we draw on the FSA pictorial project and on Otto Neurath’s Isotype method). I would be happy to include one or two other papers. These may not be strictly papers on the history of economics but also papers on the history of management or general history articles which cover economic themes (for instance, economic history, history of measurement and the larger history of social sciences). Beyond the ESHET conference, this session may help launch the discussion on this neglected aspect of scientific practice and to help increase multidisciplinary work on the subject in the near future. If you have an abstract to submit, you can do this directly to me (yann.giraud[at]u-cergy.fr, replace [at] with @), I will re-submit the session as a whole before the papers are individually submitted through the ESHET website. You can also contact me if you have already submitted a paper which you think may fit this session in particular.

Otto Neurath, designer and sociologist, made beautiful pictograms that invoked the data he displayed. If his units represented unemployed people, they might take the form of hunched, depressed silhouettes, hands in pockets. Days of travel across the ocean might be expressed as jagged ocean waves.

We often forget that information graphics need not be cold and clinical.  The book “Otto Neurath: The Language of the Global Polis” serves as a tonic to that tendency.

The blurb for the book: “Austrian sociologist Otto Neurath was a seminal Modernist figure. Much attention has been given to his achievements in the fields of graphic design and philosophy (Neurath was a member of the Vienna Circle, founder of the Museum of Society and Economy, inventor of the ISOTYPE pictorial system and champion of the Unity of Science movement), yet his involvement with urbanism and architecture has been all but ignored. From 1931 onwards, Neurath collaborated with the International Congress of Modern Architecture and its chief exponents–Cornelis van Eesteren, Sigfried Giedion, Le Corbusier and Laszlo Moholy-Nagy–to develop an international language of urban planning and design. More experimentally, throughout the 1930s a fascination with visual media led to an attempt to franchise the Museum of Society and Economy by establishing international satellite museums. This volume contains a text by curator and writer Nader Vossoughian, which offers a fresh perspective on one of the most versatile intellectuals of the twentieth century.”

Some terms related to diagrams (and also to design) strategies, language, layout, and so on are often misused or misunderstood:

- When is a diagram simple?
- When a diagram is simple is because it has few elements… and when it has more elements it becomes complex and less clear..
- Additional information is useless and complicated the message…
- Is additional information irrelevant?
- To make simpler a diagram, the best solution is to delete elements…

These terms are simplicity, complexity, clarity, additional, and irrelevant.

Simplicity. The Oxford dictionary defines simplicity as something being free from complexity or intricacy and absence of ornament. Shedroff (2007) synthesises these concepts explaining that clarity deals with one particular message or goal at a time, rather than the communication of everything at the same time; and that simplicity is misused producing meaningless messages rather than clarifying them. Referring to the concept of design simplicity, Per Mollerup (2005) talks about minimalism. He explains that minimalism deals with the fact of having few elements: simplicity of elements, number of element types, number of elements, and their size. However, both Bertin (1983) and Tufte (1990) explain that simplicity can be reached with an appropriate information organization, rather than with a low number of elements.

Complexity. Complexity deals with the number of dimensions (informational variables, level of reading, levels of information) that can be represented on a 2D design. Similarly than simplicity, something complex is not related to the number of graphic elements employed in a diagram, but with how they relate to each other.
Tufte (1983) describes a diagram as complex when it contains multiple layers of information that create multiple viewing depths and multiple viewing angles. He adds that complex diagrams need to have at least three viewing levels: 1) general structure: What is seen from a distance as a summary of the content, first level of content analysis; 2) main structure: What is seen up close as the primary idea of the content, second level of content analysis; and 3) hidden structure: What is seen implicitly and is understood besides the diagram itself, third level of content analysis.
High-information graphics combine several layers of close reading with an overview.

Clarity. Clarity is defined as the quality of being clear, distinct and easily perceived or understood and free from ambiguity. Clarity deals with the kind of message that a diagram communicates, focussing on the content, and not in the quantity of elements used to represent that content. The number of elements is not a parameter of clarity. A diagram can have only few elements but if all of them are at the same visual level with equal values, equal textures, equal colours, equal shapes, as a result there would be failed communication. When everything is emphasised, nothing is emphasised. If, on the contrary, a diagram has a high number of elements but well organised and differentiated, the result would be an extremely clear diagram. I believe that Tufte’s (2007) definition about this concept is undeniable. He states that ‘it is not about how much of empty space there is, but rather how it is used. It is not how information there is, but rather how effectively it is arranged’.

Additional. The Oxford English Dictionary defines additional as something extra, something that complements, enriches or improves a main subject. Tufte calls ‘evidence’ the information that adds details and enriches a diagram. A difference of meaningless information, additional information can sum details from multiple sources and levels to a diagram. Diagrams that contain additional information do not rely in one type of data or stay at one level of analysis. Detailed information helps to explain a complex content of a diagram, and makes it easier to understand. Furthermore, additional information enriches reading and amplifies content. Tufte suggests that for clarifying, the answer is to add details.
However, it is important to notice not additional information is considered useful. Mollerup explains what redundancy means. He argues that when additional information is purposefully used, it becomes meaningful information, eventhough it can be redundant. Furthermore, he adds that sometimes a message with no additional information is extremely vulnerable to misunderstanding. Nevertheless, not all additional information qualifies as useful redundant information. Sometimes, much additional information is just useless. As an example, both Mollerup and Wurman (2001) classify fashionable and trendy graphic ideas as additional information that causes graphic noise, has no useful meaning and makes the information less understandable.

Irrelevant. On the contrary, irrelevant is defined as something that is not pertinent or applicable, or not closely connected or appropriate to the matter being considered. Irrelevant elements are unnecessary elements. Moreover, a big difference with additional/detailed information, when unnecessary information is given more than once, it becomes meaningless and redundant.

- Bertin, J. (1983). Semiology of Graphics. Diagrams, Networks, Maps. UK: The University of Wisconsin Press
- Mollerup, Per (2005). Wayshowing : a guide to environmental signage : principles & practices. Baden : Lars Müller
- Shedroff, N. (1994). Information Interaction Design: A Unified Field Theory of Design. http://www.nathan.com/thoughts/unified/index.html [accesed on 14th September, 2009]
- Tufte, E. (2006). Beautiful Evidence. Cheshire, Connecticut, USA: Graphic Press.
- Tufte, E. (1990) Enviosing Information. Cheshire, Connecticut, USA: Graphic Press.
- Tufte, E. (1983). The Visual Display of Quantitative Infornation. Connecticut, USA: Graphic Press.
- Tufte, E. (1998) Visual Explanations. Cheshire, Connecticut, USA: Graphic Press.
- Wurman, R.S. (2001) Information anxiety 2. Expanded & updated ed. Indianapolis, Ind. : Que

“Last year, we made a great stride in writing.  The two older children went through the Institute for Excellence in Writing course.  In short, the instructor (on dvd) teaches a method through repetition.  The children go through the same method every time they write.  He gives them tools, so they can write about any subject.  IEW includes a solid method for writing research papers.  As far as English Grammar goes, we follow the classical model of learning… “Classical Conversations” gave me a plan to teach grammar.  The children have memorized the list of prepositions, Helping Verbs, and Linking Verbs.  Also, I discovered a great grammar tool in diagramming sentences!  Last year they studied this form of art using The First Whole Book of Diagrams by Mary Daly.  It brings order to sentences and helps us all to see the parts of speech.  There have been quite a few things in the past that have not worked.  Some of those include workbooks.  My children usually got the correct answers, but that did not translate into good writing.  So, I kept looking.  Finally, we read, read, read!  They read the books suggested in Sonlight for the time period we are studying plus other classical literature and biographies.  We still read aloud to them all.” –Hannah

Data Store is an interesting section of The Guardian where everyone can submit their own diagrams and information graphics. The aim of this section is to make complex information more accessible to people. Facts and figures are published using accessible tools that enable everyone to download raw information for creating visualisations.
Diagrams can be submitted to the Guardian and can be seen in a new Guardian Data Store Flickr group.

This is an effective manner to bring closer diagrams to the people, and show how powerful tools diagrams can be in terms of visual communication and education, making the complex easier and adding new points of view.

Some visualisations created with information of Data Store can be seen below:

Information graphic created by David McCandless, visualising the factors of airlines crashes. (original size 33x123cm //http://www.flickr.com/photos/25541021@N00/3682749730/)

Diagrammatic map created by Simon Rogers, visualising swine flu cases across the UK. (http://www.guardian.co.uk/world/datablog/2009/sep/18/uk-swine-flu-cases-data)

Task 1:

In systems analysis the use of the written word is not an effective way to communicate to end users or designers. This is because computer systems can be quite complex and trying to establish the structure of the system and its requirements needs to be done via a diagram that can clearly display the systems structure; its inputs and outputs etc… The created diagram of the system can then be produced to the end users, who can then communicate with the systems analyst as to what processes are missing from the system etc…

1) It is not uncommon for large projects to fail in the real world due to miscommunication of the written word. One example which I have researched is the FBI’s attempt at the creation of the Virtual Case file. The aim of this software was to “automate the FBI’s paper-based work environment, allow agents and intelligence analysts to share vital investigative information, and replace the obsolete Automated Case Support (ACS) system” (click link for details). However 700 000 lines of code was produced that was so bug-ridden and functionally off target that the US $170 million project had to be scraped.

It is obvious here that during the beginning of the SDLC, communication as to how the system should operate was not communicated effectively, hence the failure of the system.

2) From personal experience, there have been many times when the written word has caused miscommunication. For example, in my previous job I was in charge of maintaining and running the website for ‘Mahon Furnishings’. When I was starting university I was unable to maintain my job and so I had to train a new employee how to maintain the company’s website. In preparation for this I created a ‘how to use guide’ for the website. I had a short period of time to train the new employee and so after a few demonstrations I set the new employee a task to complete with the help of the guide I created and without my supervision.

 Obviously I found the guide I had created easy to follow; unfortunately the new employee did not. As a result of trying to follow my written instructions a module of the website was permanently deleted instead of a single image.

The above are just some examples of how the written word can lead to miscommunication and problems with the development of a system.  Next I am going to discuss some types of diagrams that can be used instead of the written word, to effectively communicate the systems structure to end users and designers.

Task 2:

Diagrams are an effective way of communicating to end users and designers. It is important that users understand how the systems analyst has interpreted the system. The use of a diagram to convey this is more effective than the written word as with the written word it is easier to misinterpret the meaning of something.  The systems analyst can then show this diagram to the end user; from this the user can decide if a process or a data flow is missing.

Some types of diagrams:

• Decisions trees

Use a simple tree format to represent the processes within a system.

• Use case diagram

Graphical diagram that summarises the relationship between end users and the system.

• Pert diagrams

Are a network model that shows the project as a series of activities that can be performed sequentially. From this the critical path can then be determined.

• Scope diagrams

Are used to help identify the scope of the project. In other words what problems can be solved using ICT and the development of a new system?

• E-R diagrams

Used to show the relationships between tables within a database.

• Flow chart diagrams

Show the order of processes within a system.

Use case diagram:

Claimed advantages of use case modelling:

One main advantage of the use case diagram is that it is used to show all of the processes that a system must support. In the diagram above the use cases are those within the rectangle or system as these are activities that the system must carry out. The use case diagram also clearly identifies those people that interact with the system. They are known as ‘actors’.  The use case diagram does not always have to show the ‘actor’ that is supplying the data, such as the customer in the example above. A use case diagram can just show those people that directly interact with the system. This gives the analyst the advantage of knowing exactly the precise interactions to which the computer system must respond. This then helps to define the system requirements.

From my own point of view I think the use case diagram is an effective tool for communicating to end user and designers. This is because the use case diagram shows the users needs and defines the requirements for the system. When an end user looks at a use case diagram they can clearly identify their own role and how they interact with the system. From this they can then report back to the analyst with any changes they feel should be made. Changes could include, identifying another process that is missing or removing certain processes. As A result the systems analyst gets a better understanding of how the system operates.

Decision Tree Diagram:

Claimed advantages of decision trees:

Decision trees are used to summarise how complex systems operate. A decision tree is more readable than the written word; this is because they keep “a string of decisions in a particular sequence”. A decision tree allows a systems analyst to identify and organise actions within a system in a structured way. Thus making it easier to identify how the system operates.

Sometimes within a system (like the one above) complex branching occurs. A decision tree allows the analyst to clearly communicate to designers the complexity of the system through the use of a tree diagram. If the analyst tried to describe the system to the designer using the written word; this would be extremely complicated and so the decision tree diagram allows the designer to be able to clearly see the structure of the system with out any misunderstandings or confusion.

In my opinion I think that yes, the tree diagram does effectively show the structure of a system and is easier to interoperate compared to the written word. However if the system is very complex then the tree diagram must become very large and complex also. Tree diagrams can take up considerable space and as a result the systems analyst may not have room to write descriptions fully on the tree diagram. This would make the diagram very hard for the designer to understand. I think Decision trees are most effective when used for simple system processes.

Overall, diagrams are a good way to communicate to end users and designers instead of the written word. The written word can be hard to understand (especially when trying to understand a complex computer system) and can easily be misunderstood as something that means one thing to somebody, can mean an entirely different thing to somebody else.

Resources:

- Design in a Changing World

(By John W. Satzinger, Robert B. Jackson, Stephen D)

- System analysis and design

“a string of decisions in a particular sequence”

(Kendall & Kendall)

- www.wikipedia.com

Problems caused by using the written word

Ensuring written communication is clearly worded is important especially when making a Will to ensure that it cannot be misconstrued. An example of a Will where its intent is unclear is my great great grandfather’s. In the Will property was left to his daughters, however a codicil added excluding one daughter from inheriting if she returned to her ex husband and also that her inheritance should pass to her sisters on her death. The Will failed to fully clarify the position should she not return to her husband prior to her death. This resulted in legal disagreement and to date the estate remains unsettled.

Another example of poor communication is on the following link. http://mcdowellab00518469.wordpress.com/2009/11/05/an-example-of-miscommunication/ This example shows the importance of not only stating clearly what you mean but also when unsure about the meaning it’s important to enquire to prevent mistakes.

Diagrams used by systems Analysts

There are a number of diagrams which have been used to communicate with end users and designers some examples include;

Flow Charts
A flow chart or flow diagram is a graphical representation of a process or system. It shows the necessary steps to generate the required output. Each flow diagram focuses on one particular process or system. Flowcharts use a set of symbols to represent various functions, arrows are used to connect functions. They can be used to show end users the flow of information through the system.

Sequence Diagrams
A sequence diagram in UML shows how processes interlink and in what order. This can be shown to the end user to see ensure it’s correct and when agreed shown to the project team to aid designing the system.

Decision Trees
Decision trees are a tree like graph used to represent simple processes. They are useful for programmers as they can represent what action a system should take in defined circumstances i.e. in relation to an insurance company it may show that if a client made a claim last year, £100 must be added to their premium.

Entity-Relationship Modelling
An entity relationship model is a graphical representation of the relationship between entities (a specific object of interest to the business) that need to be stored in a database. It can be used to communicate the degree of the relationship, the cardinality ratio and the participation constraints to the programmer and shown to the end user to verify the information is correct. An example diagram is shown below.

There are many claimed advantages of using ER Modelling firstly it’s a visual representation of the data rather than written therefore a smaller chance of confusion or misunderstanding. It’s easy to understand using businesses users’ language.

In a well designed model all database elements e.g. the entities, attributes, and relationships etc. are clearly laid out. Therefore prior to building the database end users can suggest amendments to better suit their business needs reducing modifications during the implementation and maintenance processes. It aids database design leading to faster development and increased productivity, overall reducing the chance of project failure.

There are limitations to using ER Modelling, as miscommunication can still occur. Human error can cause elements to be forgotten and poor communication between the end user and the systems analyst could lead to attributes which are required not being discovered until completion of the database. ER Modelling also doesn’t show all the information a programmer requires e.g. it doesn’t represent data manipulation e.g. queries or any validations which may be required.

To conclude ER Modelling is a useful tool and should be used when designing a database however it isn’t perfect or fool proof therefore other methods of data communication will also be required.

Use Case Modelling
A use case model gives a graphical overview of the requirements of the system by looking at the interaction (the lines) between end users (the stick men) and the system (everything inside the black box). The circles are the use cases.

There are several claimed advantages of using use case modelling. Firstly the focus of the diagram is on users rather than the system itself therefore the real system requirements are identified early on in the data gathering and design stages.

Another benefit is it focuses on one specific usage aspect at a time helping manage the complexity of the systems needs and requirements. It consists of mainly narrative text and therefore is easily understood by all stakeholders e.g. managers and end users rather than developers and system analysts. Including end users and managers during planning promotes a sense of buy – in from them helping the project team build a system suiting their needs and eliminating surprises.

The use case model describes the way the system is used which beneficial as it also helps identify problems which could occur early on in the project saving time. Once developed, it can be used to drive other aspects of the project, including project planning (cost, scope and timing) and user documentation.

However there are limitations to use case modelling. It is not suited to capturing non – interaction based requirements of the system (i.e. algorithm or mathematical requirements) or not functional requirements (such as performance). It doesn’t document interactions between requirements as requirements are studied separately.

There is a learning curve involved when interpreting use case models correctly for both end users and developers. As there is no standard definition of use cases, it’s more difficult for all stakeholders to understand possibly causing miscommunication. Larger systems will have more complex diagrams and be harder to understand.

The systems requirements may also be skewed by user motivations therefore systems requirements may be based on what they want rather than their system needs. The use case model may not capture all the mishaps and events that can occur. Developing a use case model can be very time consuming for both systems analysts and end users.

To conclude use case modelling is very useful however there are limitations e.g. it’s time consuming and may not identify the systems non-interaction based requirements. However it overall saves time by identifying requirements early in the project and it should be used alongside other tools to identify systems requirements.

Some doubts related to the meaning of diagrams reference elements and their geographical range variable seemed to emerge from the workshops (Barcelona-London). In order to clarify these concepts I will explain their differences and importance with an example of a recent personal experience:

I have just returned from a four-day break in south Spain. As it was my first time in Murcia, I found myself a bit lost and hard to find the correct route to get to the main square, Plaza Circular, from the bus station. So, I decided to ask a policeman from the station how to get there. He explained me the route and gave me some guidelines and references, however, after walked a couple of blocks I preferred to ensure whether I was in the correct direction or not. This time, I asked another random person from the street. As the previous one, she kindly explained me how to arrive to the main square using more or less the same particular words and expressions than the previous man, instead, for example, to common words like street names or a quantity of blocks. Both of them were referring to local directions and coordinates: a hairdresser’s name, a telephone store name, a square, another shop, traffic lights, sidewalks…
Fortunately, this woman noticed that, even though we were both speaking the same language, I wasn’t completely sure about her explanation. So, she took an A4 paper and a pen and drew a quick map of the zone, including all the local references that she considered key and useful to get my destination.

Explanation drew by a local person to go from the bus station to the main square, showing key referential elements: hairdresser, shops, traffic lights, gardens. (Murcia, Spain)

Although, I arrived to the main square, some days after that, I stopped in the tourist office to ask for a tourist map. When I looked for the bus-station-main-square route, I wasn’t surprise at all to find completely different references in that map: street names, blocks, parkings. That is to say, international references.

Map given by the Tourist office. (Murcia, Spain)

Tourists maps are created for tourists and they have to be understood for the more quantity of tourists as possible. They have to employ an international and conventional language, although, they would be harder to read for local people.

Local is defined as something that is understood by a restricted audience from a particular place, country, city… As an example, this means that foreign people would barely understand the references given by both the woman and policeman or would think that this information was meaningless.

Reference elements depict those key devices that enable readers/users to navigate into a diagram. References elements do not have to be always geographic elements. In this case, they are urban-rural references.
Geographical range variable refers to the audience’s degrees of comprehension of those reference elements.

Even two years ago many people wouldn’t have predicted that the primary selling point of a mobile handset in its own advertising would not be the hardware, the screen size, calltime or even brand message, but the fact that you can get Facebook (and other feeds) on it.

The current rapid growth of mobile social networks is a symptom of finally moving beyond the endless powerpoint based hype of “social media marketing” to a new realism. Mobile that takes social networking firmly away from sitting behind computers and means we can bring back the lost real social elements of face to face encounters and real world experiences.

Mobile coming together with social networks means that we can physically see these networks actual transformative value. It is one of the many things leading to what I hope is a sense of  ”won’t believe the hype” maturity in our industry (and wider society).

This “always on”, real-time world of Social periphery or ambient intimacy [explored in the diagram above] has been on the cards since the birth of Jyri Engeström’s Jaiku but it has taken easy and seamless mobile integration to really snowball into the mainstream.

Now the challenge is what do we do with it?

Recently I’ve been exploring the idea of brands needing to function as both enabler and filter for people in order to have a role beyond passive loyalty. They not only need a position within a market or category but must also have a clear and simple point of view on the world/culture in which they operate.

This evolving way we deliver and communicate is at the heart of the opportunity for mobile networks and technology providers in light of their move to become the enable and the filter for people’s social peripherial vision and networks.

But what will the work need to look like?

Technology lets the crowd raise-up the things it likes with links and tags and re-posts, and damn the things it doesn’t like with a pointed lack of attention. Old passive message, big idea, objective correlative creative with a big call to action, and series of key frame proof points doesn’t cut it anymore. There is too much noise: now things have to be good enough to share.

But just because something is good enough to share or inherently interesting doesn’t mean it will catch on and spread through networks. The work itself must be implicitly structured for the network (as illustrated by SharedEgg) and stimulate the growth or reshaping of the network, not just have an AddThis or send to a friend button.

• It must contain an idea that can be reprocessed and played with, passed on and owned.
• It needs to be an idea that people can use but not be completely dry and functional.
• It needs to provides more than the (important) phatic entertainment and brand story BUT ALSO vital real world reward and social currency.
• It must allow the crowd to create nodal points* within their part of the network.

Mobile is the tangible, first thing you see in the morning, last thing you check at night, always with you object, that will enable people to pull their networks and these smaller, higher frequency, locally relevant ideas together so that they can create their own digital nodal points or experience their friends’ ones while on the move.

This is why the work we do needs to be good enough to share with everyone, anywhere.

———————————————–

* If we can think of a Nodal Point as a (potentially distributed) collection of content, conversations and links that spread a meme/concept and cause the ideas, experiences and other journeys around it to be reshaped and “dragged” (just like a planet’s mass influences the passage of time around it), then it is a key point in an interactive experience or someone’s digitally enabled life.

An asterisk (*) indicates a correct answer.

1. Matrix Diagrams helps verify you have identified all possible relationships between your existing entities. True or False?
True (*)
False

2. Creating a Matrix Diagram is mandatory when doing Data Modeling. True or False?
True
False (*)

3. A Matrix Diagram will help you with all of the following except:
Defining Relationships Between Entities
Identifying Entities
Defining Instances of Entities (*)
Naming Relationships

I’m proceeding with the redesign of my diagrams library slowly but surely. I don’t have a lot of time to work on it, hence the “slowly”. But progress is being made. It’s still very much in the design phase, which makes it difficult for others to help, but Lennart has created a diagrams page on the Haskell wiki which I hope can be a good way to have an open design process and get input from the community. There’s a lot of stuff I have written down that I haven’t gotten around to putting on that page yet, but I hope to do that soon.

Occasionally I plan to write some blog posts about interesting issues that arise in the design; today is the first.

Here’s the background: a diagram is essentially a tree of sub-diagrams (although there will be a principled way to refer to subdiagrams in other parts of the tree; more on this in a future post), with leaves corresponding to atomic primitives. Every node in the tree can (optionally) have some associated style attributes, such as stroke width and color, fill color, font, and other such things. When we encounter a primitive at a leaf, how do we know what attributes to use when rendering it? It may have some associated attributes, but its parent node might also have some associated attributes, and other ancestor nodes higher up the tree might have attributes as well.

What we’d really like to do is have a Style data type which has a Monoid instance:

data Style = Style { strokeWidth :: Double
                   , strokeColour :: Colour
                   ... }

instance Monoid Style where
  ...

Then to determine the style to use for a leaf, we just combine the styles of all its ancestors using the monoid.

So, how to implement this Monoid instance? Well, first of all, what I wrote above is slightly bogus; at the very least each particular attribute ought to be optional, so it should look something more like this:

data Style = Style { strokeWidth :: Maybe Double
                   , strokeColour :: Maybe Colour
                   ... }

To combine two Style records, we match them up field-by-field, and Just trumps Nothing.

So far, so good. But how do we combine two fields containing Just? It seems we have two choices: we can be biased towards the top of the tree (i.e. parent attributes override child attributes) or towards the bottom (i.e. child attributes override parent attributes). Indeed, Data.Monoid contains two newtypes, First and Last, whose Monoid instances exhibit exactly this behavior.

But here’s the problem: in this application, one of these choices isn’t obviously better than the other. In fact, it’s easy to imagine situations where each would be the desired behavior. For example, imagine that we have created a subdiagram that we want to use many times throughout a larger diagram. Most of the time it will be blue, so it makes sense to specify that attribute as part of the subdiagram itself. However, in one place we want it to be red, so we’d like to be able to override its attributes with parent attributes. On the other hand, imagine a situation where a diagram is going to be composed of many different subdiagrams, which all share the property that they are blue. To avoid repeating ourselves, it makes sense to specify “blueness” as an attribute of the parent diagram and have all the subdiagrams inherit it. However, one subdiagram should be red, so we’d like to be able to override the parent attribute in this particular child.

What to do? A first cut might look something like this:

data Override a = Default | OverrideUp a | OverrideDown a

data Style = Style { strokeWidth :: Override Double
                   , strokeColour :: Override Colour
                   ... }

The intention is that OverrideUp a overrides any attributes above/before it, and OverrideDown a overrides any attributes below/after it. However, there’s a problem: what should

(OverrideDown a) `mappend` (OverrideUp b)

be? The OverrideDown a claims to override the OverrideUp b… and vice versa! So this doesn’t really work. We need a way to specify relative priorities. So, another solution would just be to assign each attribute with a priority:

data Prioritized a = Default | Priority Double a

data Style = Style { strokeWidth :: Prioritized Double
                   , strokeColour :: Prioritized Colour
                   ... }

For the Monoid instance, we just take the value with maximum priority. This allows us to do what we wanted—overriding parent or child attributes is done simply by assigning a higher priority. However, I really dislike this solution. Having to specify a priority is annoying—but not only that, figuring out what priority to use to achieve your desired effect requires global knowledge about the value of the priorities used elsewhere. One improvement we could make is to adopt the solution used by CSS: attributes are leaf-biased by default, but assigning a priority can override this. That is,

data Prioritized a = Default | Priority (Maybe Double) a

where the Monoid instance chooses the value with the highest priority, or the right/leaf-most value if no priorities are specified. This might be the best option—but it’s still somewhat unsatisfactory.

I wonder about a solution that allows you to say, “I want to override the attribute on THAT node”—where “THAT” represents some way to refer to a particular node by name (what these names look like will be the subject of another post). This might solve the problem of arbitrariness with the numerical priorities, but might also be veering into the realm of the overengineered…