Archive for October 22nd, 2013

Disorganisation, and a surprising amount of useful information   no comments

Posted at 6:08 pm in Uncategorized

Having had time to look in to the two subjects (Geography and Biology) which I’m applying to user interface (UI) design, I’ve found a surprisingly large amount of topics, methods and skill sets within them which could be used by geographers and biologists to help answer my question. My next task will be to sort through all of them and reject the weaker arguments and develop the stronger ones, this could take a while…

Below, for those interested, are my brief (and badly written) notes on each and how they could apply, along with the textbooks I used (which, conversely, are very well written):

Geography – most information from these notes has come from “Practising Human Geography” (Cloke, Cook, Crang, Goodwin, Painter and Philo) and “Earth’s Climate, Past and Future” (Ruddiman)

Sifting, sorting and simplifying data – geographers use large data sets which often contain lots of anomalous and irrelevant data, drawing out the most important elements of something is a key skill which could be transposed in to producing a simple and easy to use UI.

Studies and public surveys – part of human geography includes sampling public opinion, while other geographers often examine physical structures (manmade and natural). Both of these produce useful data and are similar to user studies which are a key part of user centred design (UCD).

Set theory and grouping items – grouping relevant items together and finding links between different things would help to make a well laid out UI.

Quantitative and qualitative methods – when analysing data collected from studies these are useful skills for extracting meaning.

Statistical approach with inference – more methods for discerning information from data which are transferrable.

Awareness of the limitations of measurements – e.g. Children are measured as belonging to a household, when in reality they may move between two – this is a useful part of UCD which is often overlooked (in my experience). Many user studies do not properly analyse the limitations of their measurement techniques.

Town planning – the layout of a town is not dissimilar to the layout of items on a screen, both must take in to account the different types of people using them, as well as the underlying restrictions of the system they are built upon.

Cartography – refining very complex data in to simple form which is usable by almost everyone is an important goal for UI design, and one which cartography has mostly solved for it’s own data sets. Applying similar refinement techniques to other data sets could yield similar results. Cartography also has to consider the audience and the abilities of the users, which is a core principle of UCD.

Transportation geography – underground railway maps are a simple example of where this has produced a very usable interface by creating a vastly simplified and more readable version of the actual data, without loss of any of the relevant data (intersections between lines, etc.). A map of the actual paths of the London Underground lines is barely visible due to the scale that has to be used, but the version shown to the public is accessible to everyone, and even takes in to account disabled users.

Geology – layers in rock which contain different information could be (very tenuously) linked to layers of a UI containing different data. Although the metaphor breaks down, the mental model could be useful.

Cultural and socio-economic differences/demographics – accessibility if computer interfaces is very important, and geographers are familiar with analysing and segmenting people in to different groups which could be used to provide different experiences and interface types when using a computer program.

Observation of the world and people – this is an important part of UCD and geography

Breaking down objects in to functions – e.g. a city as a transport hub – computer programs can be seen as a collection of functions working together, which is similar to a city. This type of understanding is the opposite side of UI design to the user side, but equally important to make it functional and efficient as well as easy to use.

Climate – patterns on a small scale can be applied to large scale problems. This doesn’t always hold true, but it is usually a good idea to attempt to try/extend existing solutions before reinventing everything from scratch.

Weather – inference of future trends based on current and past patterns/data. Looking at how people have behaved before with other programs, and how designers/developers have done things before is important to learn what works and what doesn’t.

Erosion and removal of soft rock, leaving structural/important rock in place – the metaphor here could be used to remove excess and wasteful parts of the UI, leaving just the parts which are important

Ethical considerations – part of any user study is deciding what is acceptable and reasonable to do.

Systems theory – the interaction between different systems and parts of a system – this is used heavily in geography and applies to lots of areas

Biology – based on notes from “Campbell Biology” (Reece, Urry, Cain, Wasserman, Minorsky and Jackson)

Cell theory – cells are the building blocks of all life – building a UI could be analogous to building an organism, the whole is made up of lots of parts, each part has a function and each part is made up of smaller parts, then all of it has to work together.

Evolution and natural selection – applying these to UI design could see several possible designs being tested then the best parts of each combined to a new generation which is then tested again, until the best survives. Natural selection also suggests that the surviver will be the one which best suits the environment it is tested in, so there would be a possibility to produce several final versions for different users and scenarios.

Mutation – taking a working UI and changing parts of it could be used to avoid the local maxima problem.

Genetics – a core set of instructions/features – a program has a set of requirements which it exists to complete, these could be seen as the core genes of it which determine to a certain extent what it will have to look like.

Link between an environment and life that exists there – different experiences may be required in different places, for example on a train people behave differently to when they are at home in the living room, even though they are probably sitting down looking at a screen in both.

Emergent properties – the arrangement of smaller parts affects the properties of the larger object – the functionality of the whole has the ability to be more than the sum of the functions of the parts

Component parts – the view that something is built up from smaller sections into a whole

Properties of life:

Order – organisation of items in the UI
Evolutionary adaptation – a UI must change as the users’ needs change
Regulation – the flow of data which is displayed has to be managed
Response to the environment – a UI must change if the user changes environment or it is used by a different user in a new environment
Reproduction – this probably doesn’t apply…
Growth and development – users may have new requirements over time, or it may be necessary to create a basic version to start with and build on it for new releases
Energy processing – efficiency is important

Written by Alex Owen on October 22nd, 2013