Tuesday, July 23, 2013

Masters project proposal " Flat design - usability improvement or just stylistics hot air?"


I saw this debate and thought how wonderful it would be as a masters project
Working title: Flat design - usability improvement or just stylistics hot air?
General description 
Recently there has been a writing debate in the user interface community about a new style call flat design. Flat design has become a familiar on the web, under Windows 8 and the new IOS7. In many ways that design is a movement against previousskeumorphic designs, but more importantly claim that flat design produces more usable interfaces.
On the other hand leading user interface practitioners such as Jacob Nielsen have also criticized flat design. I have listed out some of the most useful websites and arguments below.
This masters project is to fairly evaluate both interfaces, and produce evidence to answer the question “does flat design improves the user experiance or not”?
 To do this your approach could be to produce two identical interfaces on the phone or web and then evaluate them with real users.
MSc and taught modules (Any relevant taught modules and specialist MSc title(s) if appropriate)
M864 Fundamentals of interaction design
Background reading 
Some open reading materials and the cases they introduce are.. 
http://techstream.org/Web-Development/Core-Principles-of-Flat-Design
  • suggest that flat design will improve functionality. 
  •  [The Flat design] "philosophy preaches that if something doesn't help usability, leave it out of the design. This suggests that minimal approach will improve usablity by removing unnecessary elements." 
  • The simpler time also Speeds up down load times - important for better 
  • Makes the content stand out which is an important factor on small screen phones/tablets.
  • Simpler design takes less time and lets you spend more time getting it right and usablity
  • Flat design includes more animation,sound,vibration. 
  • flat design looks friendly and approachable to users. But it does this by presenting a clear and engaging interface, rather than disguising and warping that interface to mimic something familiar
  • speedier pages, 
  • cleaner code, 
  • easy adaptability
  • Better experience by embracing the medium (2d) rather than trying to poorly simulate another one ( 3d). 
  • skeuomorphic introduces many unnecessary elements which consume valuable screen space. 
  • Flat design is easier to read ‘at a glance’ 
  • Improved conversion rates “What's more, flat design can help companies to achieve higher conversion rates on their website. Because of the uncomplicated design, the eye is drawn directly to calls to action buttons and the user is less likely to be distracted from their main reason for being at your website.”( http://www.wysi.co.uk/web-design-and-marketing-news-berkshire/the-benefits-of-flat-design)  
 Nielsen criticized flat style in Windows 8 as 
  • reducing discoverability.
  • low Information Density.
  • For new users hidden interface features (charms on W8) which need to be summoned with a swipe are difficult for new users. 
AA: Flat design is not memorable.
Research methods (An indication, but no need to be specific unless that is required by the topic)
To do this your approach could be to produce two identical interfaces on the phone or web and then evaluate them with real users.
Other specific information
  • Deliverables: Two programs or websites which embody flat vs skeumorphic or traditional design.
  • A series of tasks for the users to perform on both interfaces.
  • A series of questions asking each user their preference.
  • Timings and error rates for each user and each task

Friday, July 19, 2013

GPUImage : NSHipster

GPUImage : NSHipster:
While my central interest is human computer interaction I can't that help stop myself from getting highly excited about this kind of low level stuff. Basically in order to process sound and images in real time on an iPhone there is a separate group of tiny processes like a mini graphics card. Programming this GPU is pretty difficult compared to everyday programming. Indeed it'  so difficult we won't even mention it on the OU new algorithms course. 

However the Brad Larson has just created an IO library which hides all these GPU accelerated details behind a well-designed object. There is a mouth watering list of image processing operations which can be applied to video stream processing. Doing something like the tracking of all the fingers on a multi-touch table would be quite simple ( and fast) .  I could imagine hundreds of quite funky programs you could build using this. 

File under toolbox goodies

also here are some OPENCV info on ios programming 
http://docs.opencv.org/doc/tutorials/ios/table_of_content_ios/table_of_content_ios.html#table-of-content-ios

Thursday, July 18, 2013

Virtual architecture idea of the day.

Virtual architecture idea of the day.


  1. Get two cameras to view scene in a remote location - in real time.
  2. Using something like a Kinect find out where people are walking past a large display. 
  3. Warp the two images so that the moving person sees the virtual location change perspective as they might expect to do so when walking past a 'hole in the wall'. 
  4. does this give the impression of hole or is it seen as some kind of visual break down?
  5. Does this make you feel like the other place is more there than the flat case ? 

Tuesday, July 2, 2013

The Coq Proof Assistant

Welcome ! | The Coq Proof Assistant

I've always been a bit warey of formal mathematical proof. Most of the people who I have met who make big claims for the power of mathematical proof always seem to have very dominant/dogmatic  personalities. While it may be a small statistical sample I do keep wondering if mathematical proof just appeals to people who's personality type is they can't believe they can ever be wrong. 

Still I like the notion of constructing 'truth' like lego bricks and like someone who goes to a modern art gallery but doesn't get the art but does appreciate the elegance which this work involve, I like the notion of proof.  My biggest doubt about the power of proof was that no one ever managed to make an automated proof system ( which if it wasn't partly based on 'I can't see my error so I must be right' theory of mathematical proofs). Or rather the automated proof systems so far have always failed to do anything but the most trivial work ( as far as I know).

So you can imagine how excited I am that someone has managed to generate a formal proof assistant. Obviously as someone who is big into human computer interaction I would be strongly in favour of a machine which will augment the process of thinking about mathmatics. The big problem apparently with this cog system is it needs a different type of proof to be based on types not sets. Still very exciting work.